9.0 SOURCE TES rs - l)UPONT DeNEMOUl{S AND COMPANY ~-1 SITE 0VElĀ·!VIEW Summary feed Dupont, according to Antioch produces the reactions: Freon 11 and 12 from carbon tetrachloride CC1 HF..,...CCl/ + HCl (F-11) 4 + CCl + 2HF .- CCl / c + 2HC1 (F-12) 4 Carbon tetrach 1ori de (CT) is off-1 oaded pri mdrily from bottom emptying ra i 1road rail cars and stored in a tank vented to the atmosphere. Process feed pumps tranport CT into a reactor which is operated in a continuous manner. Reactor output is fed into a distillation column with recycle back to the reactor. Emission sources uf CT are expected to be the storage tank and the fugitive emissions from valves, fl,rnyes and pump seals. Facilities DescriĀµtions Tank cars containing 20U x 10 3 lbs of CT are off-loaded into a 570 x 3 10 lbs capacity storage tank on the order of 250 times per year~ CT is fed from the tank car by l)ottom un 1 oat.Ii ng and pumped into the storage tank. A feed pump delivers CT to the rectctor wi)ere it is reacted with hydrogen fluoride. Since the HF is nighly corrosive, considerable care is taken to contain all reactants. Material is output to the distillation column and chlorocarbons are recycled to the reactor from the column bottom. Beyond this point there are no enriched CT streams as hydrogen chloride absorbing, caustic scrubbing, d,hi scrubbing and distillation are accomplished. The single CT storage tank has a 3-inch U-leg vent to the atmosphere and thus has no vapor recovery system. Fittings associated with CT flow are inspected for leaks and maintained according to plant practices dnd Bay Area Air ijuality District rules on volatile organic emissions. The total number of fittings .1re li~ss than 100. 127
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
90
SOURCE TES rs - l)UPONT DeNEMOUlS AND COMPANY
~-1 SITE 0VElmiddotVIEW
Summary
feed
Dupont
according to
Antioch produces
the reactions
Freon 11 and 12 from carbon tetrachloride
CC1 HFCCl + HCl (F-11)4 +
CCl + 2HF - CCl c + 2HC1 (F-12)4
Carbon tetrach 1ori de (CT) is off-1 oaded pri mdrily from bottom emptying
ra i 1 road rail cars and stored in a tank vented to the atmosphere Process
feed pumps tranport CT into a reactor which is operated in a continuous
manner Reactor output is fed into a distillation column with recycle back to
the reactor
Emission sources uf CT are expected to be the storage tank and the
fugitive emissions from valves flrnyes and pump seals
Facilities Descrimicrotions
Tank cars containing 20U x 103 lbs of CT are off-loaded into a 570 x 310 lbs capacity storage tank on the order of 250 times per year~ CT is fed
from the tank car by l)ottom un 1oatIi ng and pumped into the storage tank A
feed pump delivers CT to the rectctor wi)ere it is reacted with hydrogen
fluoride Since the HF is nighly corrosive considerable care is taken to
contain all reactants Material is output to the distillation column and
chlorocarbons are recycled to the reactor from the column bottom Beyond this
point there are no enriched CT streams as hydrogen chloride absorbing caustic
scrubbing dhi scrubbing and distillation are accomplished
The single CT storage tank has a 3-inch U-leg vent to the atmosphere
and thus has no vapor recovery system Fittings associated with CT flow are
inspected for leaks and maintained according to plant practices dnd Bay Area Air ijuality District rules on volatile organic emissions The total
number of fittings 1re li~ss than 100
127
It is folt ttldt the sinSJlt~ 10st ir~portdnt source )f CT emissions is
the storage tdnk Tt1ere are two kinds of e1iission train storage tanks (a)
loss due to tcrnk bredti1in9 and (b) middot1orkin i luss due to tank cleaning and
f i 11 i ng Uupont tias made measurements of head spac~ concentrations of CT
during tank filling and also pertori1ed u1or2ticdl calculations oased on vapor
microressure These compliment one awu1er dd ctre 18800 and C(UUlJ lbsyr
respectively Specifically for me calcu dtions tne vamicroor microressure was
taken dt ~odeg C bulk liquid ternmicroerture ccording to f-P-42 the relationship
betv1een v1orling lJss and vapor p2ssure given Dy
wnere M - molecular weight p true vapor prtSSlffP n bu1k liquid conditions (psia) I( turnover fract1on(annual hrougnput)n
tank capacity I
crude oil fractmiddot1on (=l for CT)
working loss (lb)10 3 gal
Uupont did not microredict tne loss due to normal tank breattling during
the year Vapor is expelled from tJO pri11ary mechanisns (1) thermal
expansion of existing vdpors and (2) vctpomiddot expansion caused by barometric
pressun~ changes The AP-42 emission fori ula (EPA 1Y81) is
M( p )U 68 l) 1 7J HO 51 T )5 F C K
p C
147-fJ
where - Lt) = breathing loss lbday
li moleculltlr wt
p true vapor pressure dt bu Imiddot liyuiu condition (microSid)
D tank diameter (ft)
H = dVerafde vamicroor space tll i ~ llt (ft)
T dverage ambient t111p12gtratunmiddot change liurnal (OF I
F = pctint tc1ctur micro
C Sllld 11 tank ddjust111ent fact( r
l( c rude oi l f d c t o r C
128
foe rurrnulct 1s etiildtld to 11e within+ lU~ uf actudl measured values
Fuy1t1ve emissions from leak~ in the re1atiJely feJ (approximately 100)
Vdlves fldnges and microummicro fittings are likely to be of secondary importance to
the stora~e tank emissions rnese however vJill be directly111easured in the
field 1110n i tor nId
9 l 1111ASUKEI1ENT APPRUALH
Top priority is thlmiddot determin1tion of storage tank emissions H1is
Cdn De done for tank workinlJ loss L)_y otdir1ir1q drld drlcil_yLin~ ihnlti smicrodce
Sdrnmicroles duriny tdnk car unloctding lt was proposed to insert a sampling tube
into tanK t1ead space and co 1 Iect a ti me i ntey rated samp1 e over the off - Ioad i ng
period The samp1e would then be transferred to the La Jolla laboratory of
SAI and analyzed directly by gas chromatogrdmicrot1y Normal tank breatning can be
determined sufficiently precisely (_ lC~~) oy utilizing AP-42 with accurate
tank dimensions and meteoroloyy
Since relatively few fuyitive source components exist it unlikely
tndt sucl1 e111issions would be smiddotiynificant ~itn respect to the storage tanks
However it would be cost effectiVt to screen 100 of tne fittings with tt1e
Foxboro OVA since a team would be on site to perform the tank measure111ents
rt1e tugitive screening dmicromicroroacn vwtdltt tol Im- tn~ microrocedure described in
Section 7L
UETtlmiddotMINAT llJN Uf E111SSIUlS
931 Fuyitive Releases
Apmicroroximately llU (lrvice~ were screened and six leaks above the
establisned tt1rest1olltt of LO micropm were recorded Table 9J-1 summarizes tt1e
screeniny ddtd ano 111dss ~mission microrojections Since carbon tetrachloride is
microoorly detected Dy the UVA and TLV instrn1ients fdctors between approximately 5
and 14 ~ere ctpp Ii ed octsed on resiunse t unct i uns to deterrii ne the upper bound
of the leak rate mass emissions Total fuyitive emissions range between S8
and 61U lbyear
93L Storaye Tank tli1issio11s
Head smicroctce sa1i1ples were -caken uuriny tne nearly six t10ur off-loading
interval in order to directly deternine tne CT concentration in the displaced
129
) )
Table 93-1
DUPONT MASS EMISSION PREDICTION FROM OVA SCREENING VALVES
OVA SAi Actual Umicroper Bound Device Response Response Cone Source Leak Rtte Leak Rate Location (ppm) Factor (ppm) a b Se IRc Function lbyr 1byr
Valve-Reactor Heat Exchanger 200 009 2222 I I I 285 I 93 52
All measurement are for 100 CT streams
1--- w 0
air smicroace Saturation concentratiun at 20degc is amicroproxiriately 12 Measured
values were bY and 74 These values r~sulted from gas chromatographic
dn a I y s e s ot t rd n s t er re c1 sa111 p I e s fr() m t n e 1U U I i t e r Ted l a r b a y t i 11 e i nt e9r a t ed
sample ror tt1e tank car displdcement of 2008 ft 3 the total CT emitted based
on tile ctverctye is LUU8 x 007L = 144 ft 3bull
Tnen for a density of p =PMRT
P = (1)(1gt4) 64 gL = 04lbft 3
(082)(293)
Tl1erefore tne displacement weiyl1t of CT microer ()ff-load is equal to 144 ftJ x U4
l b ft J = ~ 7 bull 6 1 b bull F o r 2 5 U off - l o ad i n gs an n u c1 1I y we have 14 400 l b s bull Th i s
compares with UuPont I s rnectsure1nents of 9 3 vapor content arid 18800 lbyr
emissions Saturation vapor pressure concentration emission would yield
2400Ulo
l)reathing loss from the tank can be calculated from AP-42 as
Lt3(1bday) = 619 X 10-5 M( p ) 068 D 173 H 051 T 05F CKcmicro 147-P
5 0 68 173 = (619 X 10- )(154( 173 ) bull (155)
147-i73
(lj) 051 (26) 05 (1)(1)(1)
= S 63 lbday
3where a working range of zuu x 10 o 400 X 103 lbs were used to determine an
average l i quid nei grit Jn nu a 11 y e111 i s s ions would be 2057 lb Therefore
working losses dorrii nate tt1e totil elissions from tne microlant
Summary
The upper and lovJer bounos of caroori tetracriloride emission ~vere
determined as Ll467 lbyear (upper bound on fugitives DuPont measurement of
storage tank work i ny erni ss ions nur 1a I tanK oreattli ng) and lbgt 15 loyear
(lower bound on fuyitives s1 111eas--1re111ent of storage tank working emissions
no ri n a I tan k breatt1 i ny ) bull
13]
100
SOUfCE TESTS - STAUFFER CHEMICAL COMPANY
101 SITE OVtRVItw
Summary
Stauffer is located in the Carson area of Los Angeles County It is
a manufacturer of polyvinyl chloride from tl1e monomer (VCM) ll11hich is produced
on sit2 Stauffer is the only California producer of ethylene dichloride
(EUC) which is converted to VCM The VCM is regulated by several standards
including Cal OSHA EPA (emission standard) and CARB (ambient)
During Phase I the plant was inspected and possible emission
sources were identified Th~se cans~sted of EDC storage ta11ks fugitive
emissions from valves flanges and pumps process vat2r content gas
incinerator effluents and th~ loading of outbound tankers
Since the completion of Phase I several events have transpired which
a ffectebullj the test program to deterni ne foe i 1i ty emissions These wer2
~ the completion of a vent ~ldS incineration system tied to all significant EDC storage tdnk breathers
$ the completion of a comprehensive study by SAi staff to develop a nationwide material balance for EOC
bull inactivity in EOC importation for the plant and elimination of EDC exportation from the plant
Based 11pon this input and the plant inspection of February 3 1981
it was expected that a plant emission factor for EDC can be determined with
relatively little uncertainty It is further expected that si 1Jnificant
atmospheric release of EOC du~ to plant operation may not occur at the plant
itself but rather offsite These vmuld drise fro11 two sources - the process
water discharge from the plant and severcil off site EDC storaye tanks
Facilities Description
Ethylene dichloride is being produced at Stauffer by two processes
direct chlorination and oxychlorindtion In the former EDC is produced by
d i rec t ch I o r i nd t i on uf et 11y l t~ n e i n the pr 1~senc e of cHl Fe C 1 J crl t d lys t I n th e
latter process EUC is microroduced by the oxychl)rindtion of ethylene with
hydroyen chloride and oxyuen in tne prescrlCe of a cdtdlyst typically Cuc1 bull2
132
lhe microldrrt fluriti1~s tuC after tne microrimar y reactions by separation and digt ti l I dt i ur1 fgtr-ocJuLL 1 s ctured ctncl umiddot 2d ctS foed for VCM production All
sourclS agre~ tl1dt c1ission of EDC is (f concern in its productionstorage
process stages ctnJ O little importance in VCM production steps (J~B 1980)
At the timt of plant inspection some storage of EOC existed at three
ledsed storage tanks located at tne Port of Los lngeles 1-iaterial has not
been 1-Jithdravm from these tanks during the 1dst year and there had been
discussion to consolidate materictl into a single tank
All microrocess components handling EDC storage are now tied to a closed
ventilation incineration syste11 It is expected that virtually all
chlorinated hydrocarbons includiny EDC will be etfectively destroyed since the
system must demonstrdte VCM concentrations are reduced below 1 ppm Firebox
tempera tu re is aoodegF and residence ti me greater than 1 5 seconds under
heaviest flow conditions Note that by way of comparison combustion perforshy
mance datct for PC8s are 99995 destruction at 1832degF and 1 second residence
time and 9Y9Y9YY4 at 2 seconds dnd 2372degF (N Flynn SAi Personal
Communication)
Fugitive emissions from valves and flanges are monitored by the
plant according to the requirements of SCAQMD Rule 4661 Pumps and
compressor follow ~ule 466 Emission and control requirements for vinyl
chloride are specified by ~ules 1005 and 10051 EOC concentrations in
wastewater are monitored several times daily in tne primary EDC steam stripper
stream and once daily in the composite plant outflow stream Daily water
disct1arge limits are 2~ ppm wit11 typical montt1ly averages being in the 8 ppm
vicinity The discharge limit is ~mbodied in the discharge permit (number
5061) vJith tt1e Los Angeles County ~anitation District
Tnere are a number of points at which tt1e rnc can be released to the
dtmospnere and they are similar for both dir~ct chlorination and oxychlorinashy
tion These include the following along with their emissions factors for direct chlorination
A chlorinator vent 2xliJ- 5 IOdSS per unit mass EDC produced
I) I ight end column vent -52x lt)
C dist i I I at i un column vent lXlU-5
u storage tank breatning 7xlu- 5
133
E storaye tankworkiny loss 27x 0-u 5F fugitiv~ emission bxll -
-JLi ildSteJdter l gtX] U
Em~ssion factors are si~~lJ fer oxychlorination processes
Enlission factor vau-5 (~1rbulle Tdilt2~ frcn tne literature (JR13 198U) and
assume 9b iciency for incineration (A-E) It is also assumed above that
EIJC emissions to water are Lf~ uf ~rihsion to citir ard that in wastewater
treatment 100 of EOC discharged tc1 vater is rel easl~d to a i 1middot ~ These emission
factors 1bull1e1middote used to prioritize r-2ieaies iJut were cldrly uude approximations
to tre planL l)ecause of thlt ncine-xt~on system it was anticipated that
fdctors A-E would be ri2dtKcdft 1-middotactor - 1nii~t1t be reduced since a monitoring
program had been in force al20st o~e y2ar Factor G and the off site storage
tanks~ vni~cn are not tied into lt1n ircinerator system rniyht -iominate emissions
Emissmiddotion limits and concomitant regulations embodied in Rule l00S of
the SC-l)MU have necessitctted tlw incinera-cion system Ninety gas chronatograph
probes are located throughout t11e plant including the stack of the primary
incinerator Con cent rations of vc1v1 are reported es sent id I I y be I ow tt1e r~yu la -
tory limit of 10 ppm and in fact below the limit of detection somewhat less
tl1an 01 micropra
lUl MEASUK~MENT APPROACH
Tne objective of tt1e proyram is to determine a pl ant emissions for
EOC It is not accemicrotabl~ to dtmiddotvelop sucll informatmiddot1on basect upon published
indlllstry Jide esti11dtes of plant control efficiencivs Fortunately it was
possible to desiyn a monitoring and calculational progra1n to determine EUC
emissions for tne site and not ctbsorb a disproportionate share of program
resources
Hie re are tnre~ modes ot re I ease from tt12 micro Iant dnd d fourt11
offsite
bull Post-Process 1nci11er-dtion
Processes A-E of ecc ion lll 1 dre a 11 1ented into the pl ant
incinerdtio11 syste11 rt1e conceritrcttion of VCfmiddotl contir1uously iectsured in till=
stctck cts SJdS output remicroresents d rectsondble UjJper li11it to diJply tor EDC
con cent rat i ons s i nce i ts ef f i c i ency of i nc i nera t i on 1s d t 1 east c l1 u a I to tr1 at
134
ot VCibulll Tlerefore knowledge of the system dirflow rnd VCM concentrcttion are
tt1e necessary and sufficient conditions for determining the bounds of EDC
reledse fhe pldnt expn~sse I wi 11 inyne~)s to provide tnese data in order to
support the calculations
bull Fugitive Emisions - Valves Flanges Seals and Other Sources
Fugitive emission from the valves pumps and flanges can be
determined more precisely tnan was anticipated since Stauffer has completed a
comprehensive leak inventory of all such components in compliance with Rule
466 466 1 and 1005 The Inventory delineates all leaks uncovered by their
three man crew throughout tne year and indicates the screening level in ppmv
and component identity In consultation with Stauffer we will be able to
identify the total number of components associated with EDC handling systems
( 700) the distribution ot substance composition streams the distribution
and incidence of leaks by hardware component type and leak rate We will
utilize this information to provide historical data to compliment our Foxboro
OVA sampling at the site dased upon this monitoring we predict an EDC mass
emission rate for the fugitive releases The plant has agreed to provide the
necessary information We propose to meet witt plant personnel prior to the
start of monitoring and finlize the sdmpliny strategy to accomplish 100
coverage of the lines of hiuhest EDC composition The sampling approach and
calculation of mass emission are described in Section 72
bull ~~as tewater
From examination of the emission factors derived from published
literature (see Section 10~) it is clear that EDC release from wastewater to
air is potentially several 11rders of magnitude higher than any other plant
source l1ased upon microl ant ml~a sured concentrcJ ti ons of 8 ppm EDC in water a more
realistic emission factor wgtuld b 24xlo-4 1-iassunit mass EOC produced
Clearly this could still be the dominant source Furthermore the release
point would be expected to middotgte locctted between the plant and the sanitary
district treatment site whi1 his~ kmfrom the plant at 24501 S Figueroa We
believe it is necessary to ndemicroendently confirm the average 24 hour EDC
concentration in the dischar-qe valer by obt1ininy the refrigerated composite
sample We t1ave identified ttie I 1nes of int~rest and received agrt~ement to
sample and analyze for EDC in the stream ~e propose to draw a duplicate
sample from the compositor nd ani1lyze for its rnc content This will be
135
compared with Stauffers para11eI tnalysis It is not reliable to obtain
direct readings of EDC by survey instrument in the air above the water flow since concentrations at crny single point will be i 1 the near dmbient range
Emissions will be cc1lcJLtecl 1_sing the plants vulumetric daily flm
and ass~m~ng that complete degass40~ of the effluent will eventually occur
This assumption is based umicroon the relative volatility uf EDC and tne distances
involved However it should be noted that no direct experimental or
monitoring data is available with which to confirn this Th2 composition of
the discharge stream is unknomiddotwn since it merges into a larqe multisource flow
Conversation with the LA County Sanitation l)istrict (J f1ilne) reveals the
stream to be both exposed and covered Vents exist where me measurements
could De made to assess gross leaka~e at key points
We will obtain samp1es of several plant process discharge streams in
40 ml bottles witn no head space Transit time fr0m sample collection to
analysis will be minimized and wi l I not exceed 24 nours This time frame is
conservat i v e al though ED C i s a v o l ct ti l e mater i al ar1 ct vJi 11 undergo
concentration degradation and outgdssing lnalysi) will b(~ performed in tt1e
SAI Trace Environmental Chemistry Laborcttory utilizing pur 1Je and trap analysis
FID gas chromatography A trial a11alysmiddotis was conducted and the EOC
characteristic peak was distinctive down to the ppb level Therefore the
analysis should easily confirm concentrations in t1e 8 ppm range
Offsite Storage ranks
Three EOC leased storage tanks are located offsite at the Port
of Los Angeles and are owned by annther firm Annual emissions from the tanks
were very roughly estimated based un AP-42 JRB 1980) as 44 kkgyear
based on preliminary estimates of stored quantites
Considering tne magnitude of this source these storage tanks must
be investigated We will gather information about their configuration and
control i n order to ca I cu l ate their e111 i s s i on s bull
136
lU3 UEfERM I NAmiddot 10~~ OF MISS IONS
1031 Incinerator
Vinyl chl 11ride co11centration is monitored dt approximately 90
locations throughoul the pl trlt (Lanyner 19Bl) including the incinerator
output Conc~ntrations are reported by Stauffer dS less than 0l ppm at a
flow rate between 10000 ant 12000 SCFM Although no direct monitoring of
EDC is conducted it is posible to conservatively bound the concentration of
EDC at U 1 pp111 Tl1at conce 1ltrat ion is a suitable ci10i ce s i nee it represents a
conservative bound on tile VCM levels measured near the stack output
Furthermore the molar volume will be taken as 224 L rather than the higher
value it would nave because of the sl i~thtly elevated temperature at the
detector location
Using 12000 SCFM one has thE annual emission of EDC as 12 x 103
SCFM x 283 LSCF x 526x105 minyr x lmole x 97gmole x 10-7vv 224L
Therefore incinerator emissions of EOG are thought to be bounded by 773 kg or
170 1byear
10J2 Fugitive Emissions
Seven hundred (7UlJ) sources were surveyed comprising nearly 100 of
EDC service All accessibl~ plant areas with streams containing greater than
1 EUC were screened except for a small number of devices located in areas
11lere active maintenance was being condJCted It is believed unnecessary to
p2rf-gtrm any emission factor adjustment since it is estimated that gredter than
95 of the requisite components were screened
Three leaks above an arbitrary OVA reading threshold of 20 ppm were
parameters and mass emission numbers fhe upper bound leak rate was
determined to be close to twice the nominal leak rate which accounted for the
response fltlctor of amicroproximamiddotely 05 by Radirn for TLV detection of EDC
( 13 rown 1980) bull
Stauffer found and reported approximately 10 leaks in the EDC
service during 9 months previous to the pLrnt testing Assuming no111inal leak
137
) )
Table 103-1
STAUFFER CHEMICAL MASS EMISSIONS PREDICTION FROM OVA SCREENING VALUES
OVA SAI Actual Upper Device Response Response Cone Source Leak Rate Bound Location (ppm) Rae tor (ppm) a b Se I Re Function lbyr lbyr
Gate Valve Unit P1559A
Valve Heavy Ends Column Unit 14439
Compressor Seal EDC Storage Unit Tl062
I-- w co
300 11 273 -035 096 0 17 153 D 6
11 I500 114 439 II 241 C 1
r III II II o-is lUI bull2000 vl 1 1818 A 9~
All emissions 100 EDC (12-Dichloroethane)
11
values in the nei~hborhood of 2000 lbs~ total could be emitted annually
Therefore it appears ttiat major reductions in the mass emissions were achieved
by the company run inspection prugram and the fugitive emission source has now
become of seconqary importance as a fraction of total plant emissions
lUJJ Wastewater Discharge
Wastewater samples were collected in duplicate from four sites
within the plant for determination of ethylene dichloride (EDC) concentration
The samples were collected in EPP standard 40 ml VOA vials on July 1 1981
Analyses were performed using standard purge and trap techniques coupled with flame ionization detection gas chromatography The results are given in the
table below
Sample description EDC conc~ntration range (microJml)
EDC concentration average ( 1-19ml)
Approximate fl ow ( gpm)
PVC Interceptor Box 82 - 108 95 200
EDC Stripper number Chlorination area C1404
2
011 - 014 012 25
Final collection site for pH adjustmentP663 349 - 3B2 366 350
Final discharge site sanitary sewer 6 - 254 158 500
The water in the PVC lnterc1~ptor ~ox was warm and represents washings from the
PVC reactor which travels t 1J the Interceptor Box in a concrete drainage ditch
Water from the EDC ctripper was vPry t10t and because of its heat was difficult to collect The ~1eat ot trie wat11 r may in pdrt account for tne
relatively low concentratio11 of ElC here as the EDC would out-gas from the hot
~vater more readily The fir1ctl collection site tor pH adjustment is a large
concrete container middotvith mixers in it located just prior to the final discharge
site Water at the final dischdrg~ site is being constantly aerated due to
the speed that it flows thru~yn the concrete drainage trough This aeration
could account for tne relatively large range in the rnc conotent found here
The average EDC concentraticmiddot1 at tne final discharge site is middotbelow the daily
discharye requirement of 25 19ml EDC
139
Plant oersonnel indicdted monthly iverage readings are typically on
the order of 8 ppm but daily averages can re 1ch greater than 20 microgml In
addition LA County Sanitation Uistrict staff (J Milne 1981) indicated that
an on-site impoundment pond can become laden with EOC during certain abnormal
operating periods and discharge Vdriances arl requested by Stauffer This
mi 9t1t Jccur on the order of O1c0 aCii ecH c1 1d t1erefore is no-c expected to
signifcantly impact averag~ cnnJJI discharg ~ values lt s not expected at
this t1ime tEit evaporative eriss uns from this pond are significant except
infrequently during upset conditions and spi 11 control operations Program
staff ~~ 1er2 una111Jare of any periods ~h 1~n EDC cimtent in the ponds could be
appreciable and therefore no sampling was performed
Utilizing the average uf the two final discharge concentration
1readinJs (158 micrograrnsm~ 1 and 1G 1Ji gpm fl)W one has 34600 lbyear of EDC
released into the sanitary sEwer For the pJrposes of calculating population
exposures from plant releases it will be assJmed that the EDC is locally
emitted from the wastewater streams Tl1ere ire no monitoring data available
with which to develop a more accurate release profile
However emissions of me fron the plant wastewater discharge can be
calculated according to the method of Mcmiddotckay (197S) as modified by Dilling
(1977) It should be noted that this mLst be considered an estimate since
conditions of fl ow and the presence of other substances will influence
emission rates
Using Uilling (1977) for non~erated flow first recalculate
Henry 1 s law constant (dimensionless-my of chmiddot~mical per liter of air divided by
mg of chemical per liter of water) dS
1604 P M 1 Wl
TS l
)
wnere
H Henrys law constant dimensionlessl
Pi tile compounds pur1~ componen- vapor pressure in mm Hg at T
Mwi the 11olecular weiglt
T tile absol ~te temperature of he wastewater in K
Si tt1e compounds ~uluhility in myliter at T
Then for UJC
(l6u4)(7~)(9Y)H = = U0447 with 1
(2lJ4)(8690)
LIii bullul1Jl)i I 1 1y 1 middot UL i11 wt1Ler yiven -1L L0degC is 8u4JU UHI tile vapl)r pressure
1 4 psi ( 7 2mm) at 7 0 degF bull
Tl1e overall liquid mass-transfer coefficient Kil is given by Dilling (1975)
as
(2211)(06) in mhr
-f-0-4~ (M _) 12+ 100 Wl-H-
1
=U108111tH
Then from Mackay the percent desorption is given by
c 1 = exp (-Kil tL) where
c 0
c = the concentrd ti on at time t of EOC 1
co = tile initial concentration
L = the Ii quid depth (Ill)
t = tt1e retention time (in hr) of the liquid in the wastewater
system
Retention time is bc1s~d 11pon a flow velocity range of 3 ftsec as estimated by
the Los Angeles County Sdnitatii)middot1 Dibulltrict (1J Milne) over a middotdistance of
ctmicroprox i 1nate I y S km to the Joi n_t Wdter Po 11 ut ion Contra l Pl ant Sewer fl ow
depth throughout the entire route have been estimated by J Milne as typically
1 foot 030m) to the Davison Pump Plant and 3 feet (09m) to the Joint
Tredtment Plant distances assumed to bt~ 1 mile and 2 miles respectively
Trdnsit tiine l)eco111es between 05 tnd LU hours sequentially Then computing
the net emission reduction as the product of each leg
Therefore 26 of the EOC i erni t ted bdween the pl ant and the Joint Water
Pollution Control Plant Rtgtsfdence ti111e and conditions at the plant account
141
for additional releases and residudl content is forther emitted between the
plant cnd the ocean discharge poiiit ~-Herever the flow is r2tained aerated
or shallow emissions will be accentuated
Offsite Storage Tank Emissions
There are three storage tanks leased by Stauffer for EDC storage
1ocated at 22nd and Gaffey Sts ~ bull San Pedro These tan ks a ~e used for long
term storage rather than providing feed on a routine basis Therefore
breathing loss rather than v1orkir 1J lass is )f concern An tanks are white
two being 67 ft in diameter while t~2 third is 57 ft All are 40 ft 3 inches
high and have capacities of e~thi l0~10000 (2) or 840000 gallons The
tanks are cone roof type and are not tied into vapor recovery systems The
South Coast Air Quality r1dna~J~ment u~ str4 ct has based their estimate of
emissions on the specification of 12 foot vapor level Using the AP-42
formula for breathing loss with the vapor pressure of EDC at 20degc and an
average diurnal temperature variation of 26degF one hds for each of the two
larger tanks
l) 173 H bull51 T bull 5 LlI = (619 x 10-S) M( _P_ tH F CKcp 147-P
68 173 51 5 = (619 X 10-S) (9119~ 116 bull (67) ( I 5) (26) (1) (1)
14 7-1 16)
Thus for the two larger tanks LB= 472 lbday
For the third tank 0=57 and = 178 lbdayL8 The total emissions become 23724 lbyear ~ince he plant is in the process
of shutdown it is not clear what the liquid levels currently are Note that
if the material in the three tanks were combined into one totamiddot1 -~1nissions
would be reduced markedly Exposure to a population from this site was
determined separately since it is lbullgtcated over six miles fron tl1e plant
142
110
ASSESSMENT OF PUPULATlUN EXPUSUR~
111 OVEKVIEW
A major program goal wa~ to compare emissions from the various
sources and identify and rank any 11 tlot spots in California where the general
population was exposed to elevated concentrations of carcinogens A simple
Gaussian dispersion model was therefore used to obtain order-of-magnitude estimates of exposure of the genermiddotal population surrounding each source
Since this was essentially a screening study use of more sophisticated models
was not appropriate
It cannot be emphcsized too stronyly tnat most California residents
are exposed to emissions of hazardous substances from a variety of natural and
rnanmade sources Urban dwellers dre typically exposed to greater concentrashy
tions than rural residents however all are subjected to so called 11 background 11 levels from multiple sources In order to place stationary
source exposures in perspective the typical ambient levels of each substance
were identified from the literature and compdred with the concentrations due
to the emissions from each ~lant Exposures were thus expressed both as
absolute quantities and as increments above background
Comparison of plants presents a further difficulty in that various
substances are being consict~red No attempt was made to evaluate the relative
importance of exposure to two different substances such as chloroform versus carbon tetrachloride other than by ambient concentration
112 DATA SOURCES
1121 Meteorologicdl Llata
The dispersion model to be descritgted in Section 113 required input
of annual average wind speed and frequency of occurrence of wind from each
compass direction These data were obtained for most of the sites from the
South Codst Air wuality Management District (SCAQMD) In other cases (Kaiser
Jotms-Manvi l le l)ow and OuPlt)nt) only clai ly cJVerage wind speed and frequency
data were available In al I cases we used data from the meteorological
station nearest the modeled e11issrnn source Table 112-1 summarizes the
141
Table 112-1 METEOROLOGICAL DATA BASE
Observation Site Data Period of Pldnt (Distdnce Source Observation Remarks
from Plant)
A11 i ed Che11 i Ca 1 Lennox SCA~MD 1Y71-1974 Averaged hourly readings available ~ 1 Segundo (3 111iles)
S ff~r-- Uemical Long l3each SCA()MD 1962-1974 Averaged hourmiddoty leadir19s avai1able offsite Carson (3 miles) storage tanks appruxi1ntL~ly 5 miies from
Figure 114-3 Cumulative Population Exposure to Carcinogenic Polycyclic Aromatic Hydrocarbons From Coke Ovens at Kaiser Steel Corporation Steel Mill Fontana
CV)
0 -I X-V) V) a _J
s 0
C 0 micro co s micro C a u
l ~
14
1 --
J- t_bulli
_ C
uw m 0
1 -0 a bullbull micro
microdeg ()
0 micro tbullmiddot C QJ V)
0
X LLJ
C
middot4middot~middot C 0 micro rt1
-- - 0 ~ 0
0
n __ _______bull~ -______ --al_ - ~_
I bull
I I - bull
middot 1
-~1
l i l ~l
~ 0 I 1
Li bull I t_~I
i bull -1 _1-J
0fmiddotbullmiddot-middot+-middott-t-r+-t-+-+middotr+-t--+-1--t+ t f-t-J-t--r-+tmiddotmiddot~~bull+-t--1~~bull-+-1 Id middot t 4 E1 (
I ncrementa 1 Concentration (~gm3)
Figure 11 4-4 Cumulative Population Exposure to Asbestos From Johns-Manville Plant Stockton
I
)
--~ ~w t _ ---~- ----------~ --____~-=~ M
~
bull I bull I I I I ii i Q t I i E e- I Ii tl I I I 6 I a I I J I f t t bull I G fl I
a I e 1 1 a 1 1 1 1 iii bull bull 1 1 bull 1 1 bull 1 1 bull bull bull bull 1 1i bull bull 1 1 bull a bull 1 1 r ~ middot u ~ ~ I0
+H~ ~ ~~~~-~~~~-rmiddot-~~~~~~~~~-~--~~-~~-~t~~~~~-~J C
~1_ c bull 1 t~ bullbull bullabmiddot 11 l bullmiddot II bullbull C- I u~
0 rl
gtlt I I I- 20 () ()
CJ -1 ~_J
J Cbull s
I If I I I I0
C bull I Ii I I I I I I I I0 1E
bull-+l I t I I ro s Lower+l
I I I G I iii I bull
() 14 I I IC
u I0 I I II Iu
C
1-_~I fi
CJ ~ I-- (j) CJ I G I I p +l
ro +-l I I I1~3~V)
0 +) UDf)Pr u I I 11 I I
CJ 1iII
Vi I 0 I I It- I a X
0 ~
w I I I I bull
Lower C 0
middotr-
- 4+)
(1j
J Q_
0 bull1
ij a 3- ti I
D d I I I I
I I I bull
II B I t
i ] i amp I
1 ei ~ amp 3 ii
I I I I
e ft a a I bull G
I I I bull I
I I bull I I t If
UpPrf
I I
I I I-_L~ lI I I f
I I I I I
- I I
j I I II I -
lI ImiddotbullII
l I
l Li
Incremental Concentration gm3)
Figure 114-5 Cumulative Population Exposure to Perchloroethylene and Carbon Tetrachloride From Dow Pittsburg (Solid Lines) and to Carbon Tetrachloride From DuPont Antioch (Xs) Results For Upper Lower Bounds on Emission Estimates Are Shown
1i-1 - middotbullth~middot t =lttF--i-+-middotr+-r+middot~middoti-middot-~-4-4middoti t -~--is -~middot-rmiddotibull--+ u ~~ i--t4~ t-bull~~311 li-i1middoti- C - JJ _ CL
-- I 1 t middot1 c7 Ibull ) ~ I-middot bulli C bull 1-_ 11 11gt1l11 ~bull -a bull-bullbull t a- II bullbullbull bullI lt_I Iii _- I
Incremehtal Concentration (microgm 3)
Figure 114-7 Cumulative Population Exposure to Carbon Tetrachloride From Allied Chemical El Segundo (Curves Corresond to Upper and Lower Bounds on Emission Rate Estimate)
--- I
~__~-unt2wbull--o1WCUatJ41ia1
16 0amp (V)
0 ~
gtlt
14~11l 1l QJ _J
~
1 middot-0
c t1middotmiddotmiddot 0
micro ro
_ c micro ~ 1euro1pound1
0- QJ u-J c 0 u O =-middot ~-= QJ
c_1 micro ro +- V)
0 micro ~ t1 O QJ 1l 0 a 40gtlt w c 0
-
ro --
micro
20middot l ---a I ~ I a 0
0
pound1r-t k
Figure 114-8
8tWI-S~1111111N~-
a I __ I wi~ i awLi--~~ 2 I I
1--+ I ~--if bull =~~--bullJ-+--f-1-----middot+--+middot-middot--r9c- 1 c- J
bull _ bull -~
Incremental Concentration (microgm 3)
Cumulative Population Exposure to Carbon Tetrachloride and Chlorofonn From Allied Chemical El Segundo Assumingsix-months Operation With Each Feedstock (Curves Correspond to Upper and Lower Bounds on Emission Rate Estimate)
i u---J--bull-+ i i--+ --4--+-imf-bullbullQ~-+-+-+~bullbull-4 171 i R = Fi middot1 1bull1 r1__ bullltaa ~a
Incre0ental Concentration (microgm 3)
Figure 114-9 Cumualtive Population Exposure to Carbon Tetrachloride From Allied Chemical El Segundo During Offloading From Midnight to 6 am middot-middotmiddot------middot--middot
--bull ft ampLA 1111-~lliaIN_WPl1A 1JIU_Mt91 MJ1WWlaquoPI-W1111~ --------__----Et1
M gt~middotI
0 -I
-X
~ t~ bullM
V) middot-middot ~ V)
QJ _J
f 0
0 -
C -4 0middot~middotmicro re f micro C a u
10 Cdeg u 0 3 0middotmiddotmiddotbull~ -0 a
+-gt ro
+J V)
+-gt 0
2E1-0 QJ V)
0 a X
LLJ
C 0 1pound1--middot+-gt ro
-- Q_ 0 0
I I
~ -~middot bull I -bull-_ bull I
middot lbullgtbull___I I
frac14 llk I
I bull
-~
I
-1_ I
Ibull II__ JI
I~
(1 f---1-+ bull-r_--f bull middot~-bull +-- bull i -bull 1 -1-+--+bull-+-t-+bull--t1 ~ _ 4middot 6 E 1t1 1middot2
1 3 5 I middot~ _ 11
Incremental Concentration (ngm3)
Figure 11 4-10 Cumu1ative Population Exposure to Ethylene Dichloride From Stauffer Plant Carson
_~ ~ t M
C X E k1~
-
- c- 1-1 ~ C middot-middot -_J
I--- -J 4t1middotbullmiddotmiddota
D
-+-shy --middot liiit1J
r ~
ltL
t-1 _ V
0 -+-1
middotr -bull ~1
X
C
+- bull1 ~1 atr l -=
~~lll(k~IHmiddot~~-Wa~middot~nL1middot~~~ta KF u n P middot~dE-tveRS ~-tl~~~-u-~~~tmiddot_-~~~~liil-~-~~~ tc r
I bull
bull -r~l-yen--~1l~a l fl __ A4o bullbull bullbullbull ~ 111bull--middotbull
8 NJ industrial sites 28 samples in industrial areas with benzene consumption facilities Torrance CA Tuscaloosa AL National Forest 1000 sarip Ies-Torontt) Azusa CA El Monte CA Long Beach CA Upland CA Los Angeles CA Riverside CA
bull Assumes all year operation on either feedstock If plant operates 50 on each feed the population exposed to greater than 50 increment over background goes to 16025 - 19377 and 4044 - 16025 for 100
Ambient concentrations of benzene vdry over one order of magnitude in the literature and therefore make tnis calculation questionable For Kaiser therefore the carcinogenic PAH assessment was used to evaluate incremental population exposure above background
+ The partition of emissions from Oow are 78 carbon tetrachloride and 22 perchloroethylene by weight
173
1142 Comparison of Incremental and l3ackground Concentratiors
Those sites whicn elevate background concentrations greater than SO~~
City Kentucky 6600 The latter two are associated wi t11 EOC pl ants Data
taken in senFice s~ations ano lrctffic areas ir various cities range from -~
300-3640 ngm~ As previous~J mentioned the Stauffer plant is discontinuing
operations These two sources should be examined as part of any possible
start--up permitting activity
11422 Kaiser Steel Corporation
The Kaiser steel plant polycyclic aromatic hydrocarbon (PAH)
emissions cffmiddotise from the coke oven omicroerations Comfdrison of the cumulative
population exposure Figure 114-3 and Table 114-4 with the specified
background levels for urban areas illustrates the breadth of the exposure
distribution The five known PAH carcinogens that were isolated in the coke
oven emissions were quantified in tile ambient air of Los Angeles by Gordon
1Y76
Benzo[a~pyrene 046 ngm J
t3enzo~e]pyrene osio Benzaanthracene U lo
Chrysene U60
lndeno Cl23-cd~pyrene 134
Althouyn these concentrations ctr~ low co1npared with 1 number of other cities
cited in the literature and repres2nt a very I imited data base tt1e predicted
concentrations from the Kaiser plant ~enerally exceed these levels by d
174
significant margin ie greater t11an 30000 are predict2d to be exposed to
4redter than l()-lt t 1tbull 1111tlit~nt background of 3j nqm 3 It ic 1 iklly tlat
lHrllerte exmicroo~urt 11 Ilie drld dre also 11evatPd ovr r11nbient I1owever since
backyround concer1tr1tions of t)enzlmiddotne show large variation no population
calculation was specified
11 4 bull 2 bull 3 Dmv
Carl1on tetrdcnloride releases from Dow constitute approximately 78
of tl1e total CT plus perc emissions in Table 114-6 and were found to elevate
urban background concentrations greater than 50 in five census tracts Emissions are predominately from storage and check tank working and brething remiddot1 eases
1142~4 Allied Chemical
The Al lied plant was modeled several ways since the plant can
operrttt~ wiUl chlorotor111 dr cc1rbon tetrctlt hloridt fo(~d The c1bull-i pr~middot11ntlid in
ldlgtlt~o ll4- drld llil-B represent dflrlUil 01Hrdtion with eitt1er teed Partial
yedr operation with each feed can be scaled from the individual annual modes
and is presented for equal hltllf year operation in Table 114-9 As with the
Du Pont plant carbon tetrachloride emissions arise from feed tank working
loss fable 114-10 illustrates peak exposures predicted to arise during an
off loading cycle As expected concentrations in that six hour period far
~xceed annual average values Insufficient data were available to contrast
levels with backyround transient concentrations
175
l~O
AUHLAlHJ~ CONTROL n CHNil)UES
Alternative control approacnes for the most significant emission
sources dinmig ~he various pl1ants n ees fitie(J n the follo11 ric suJsections
EmpiiasVi nas Deen placed in dedlinq ~riU1 Uiose sources of 91reatest absolute
inaynHud2 (lbyr) and those const1 ut 0ing trie largest increm121middot1t middot] bulli~he
bdc kgrnund con cent ration of Hie em~ tted subs tcince i~o ctt tention Wi 11 be ~i ven
to the s~iondc1tY sources witt1middotn cK1 microant or to trlL case of J011ns-Manvi I le
since -UH imiddotajor source is less tran LOO 1byr and is not microrejicted to raise
bctckgro 1wd exposure levels to the ~~ntr21 population Furt12rmore a11
emission sources dt Kaiser Ste 1 1iC(1 v1 1~n directly dedlt -1ith in this
program r1re elltlteci ro t11e cc1lt1bull ovcmiddotr nperatmiddotions These faci 1ities are to be
closed du1rm and all primary reel 11i~ oH~rdtions discontinmiddot 2d Kaiser
forecasters have predicted further deterioration of the plant economics and
the phased closure has been accelerdted for primary steel making operations
Note that this closure is essentially irreversible cince differential
exmicroansion and contraction of the coke oven structur1 occurs in the cooliny
process and it would be improbable that ovens could be reheated without
extensive reuuildiny at major expense
12l STORAGE TANK EMISSIONS - fAUFFrn IHJW OUPUNf ANU ALLIED
At c1ll four sites emissions from storage tanks constitute either the
primary or near dominant (Stauffer) source of carcinogen release andor
genero1i iJOpulation exposure Curreritly the tanks of interest at each site are
permitted by the local Air Quality IJistricts hJwever they do not require
emission ontrol systems for vdrious reasons Tl1e estimated releases grounds
for exemption and other pertinent information are given in Table 121-1
In order to amicromicroreciate tth~ practicdl dlternatives for emission
controls the provisions of SCAWMU Rile 463 ar~ listed below which specify the
acceptable alternatives for tdnks r4uirin9 controls ie tanks 11aving
capacities greater than 39630 yal with suostances of true vapor pressure
176
Table 121-1
Site
Stduffer Uff-Site Tanks (J) (Sdn Pedro)
Allied (El Seyunrlo) - KdJ
1bulllctLer1 d I reld iturctjc
-J -J
Dow (Pittsbury)-Product Check Tanks (4)
Uow - microroduct sturag~
DuPont (Antioch) - Raw Materidl Feed Stordge
Approx Capacity~
6lUS x 1g (c) 84 X 10 (1)
lt3Y630
18000 each
70000L) 3UUOUO
30000 working 42636 liquid full
STORAGE TANK
Substance
Ethylene Uichloride
Carbon Tetrctci1 I Jr i J1
Perchloreshythylene and Carbon Tetrashychloride
CT perc
Carbon Tetracr1 l ori de
SUMMARIES
Emission Mode
Tank breattli ng
Tank v10rk i ng Tank breatttiny
PERC-~JOrk i ng PERC-b r2a t ~ i ng CT-working CT-breattling
Tank breathing Tank breatt1i ng
Tank workin9 Tank breathing
Vamicroor Pressure
014 at 70 F psi
1 63 psi at 68 F
16d psi at 68 F(CT) UJ9 psi at 68 F(perc)
168 i)Si dt 68 F
Grounds for Exemption
Exempt from control re4uirements to SCAQNJ Rule 463 since vapor microressur-2 dues nut iXC~ec
l~ io at sturdJ~ conditions
txe11pt TlJil fule -+b3 since tank capaL ~J ~ s u~der 39630 gal (lSO 11 )
Exempt from Regulation Rule v~~vu u~~~~~shytank capacity is under 39630 ga 1 andor substances are not classified as organic 1i4uicts
Exempt f ro1n Ru l ~ 85300 because subsLctnce is not classified as dn organic li4uid accordin to Regulation-Rule 8120
exceeding l~ microsi at storage conditions
~ floating roof tanks
~ fixed roof tanks with an internal floating type cover
bull a vapor recovery system with vapor collection and retllrn (or disposal) proce5sinJ lXctbullerlin~ 95 efficiency
ln a cllemicai plant a typical vapor recovery systein (KR Evans
SCAQMU) migm consist of a collection inanifold to d recovery system (suctl as a
vapor sphere) to a compression syst2m dnd subsequently to absorption and
recovery systems The absorpt~on st~n c~nsisting possibly of (rubber
stripper or activated carbon
In dealing with speciti( carcinogenic substances such as vinyl
chloride hiyhly efficient vapor control system perf0rmance has sometimes been
stipulated necessitating incineration systems
Fer th~ cases of corccrn realistic alternltltives are as follows
Stauffer Off-Site TanKs - Tnere are a number of options which can 1~arkedly 3
reduce ei1issicns from these tanks from th~ calculaL~d value of over 20 x 10
lbyr One alternative is to consolidat~ the material in tne three tanks
One of the large tdnks can contain the currently stured material and reduce 3emissions to apmicroroximately 13 0 x 10 lbyr Anotl1er alternative is to
transfer dll E0C to a single floating root tank Various types of such tanks
exist dlld are reviewed in the EPA report Lrganic Cht~mical Manufacturing Volume
J Storage Fugitive dnd Secondary Sources EPA-450J-80-025 eg internal and
external floating roofs and a variety of seal desiyn configurcttions Generalshy
ly such designs would be expected to reduce emissions to the order of
one-fourth to one-fifth the current level Cost of 1nstalliny ct contact
single seal internal floating roof wds estimated by BB Lumquist Pentrex
for EPA as $27770 in 1Y8U dollars for 7U ft diameter tdnk Cost to build an
external floating roof tank was estimated by G Stilt of Pittsburg Oes Moines
for EPA as ~14UUUO for D=67 H=4Uft Anottler com111on approach is to utilize
cdroon ctdsorption This works we11 witn nonpolar hydrocarbons dS VvC is
removed from the vapor phase A basic syte111 consists ot tvo carbon beds and
a regeneration system Regeneration is typically performed with steam or
vacuum Figure 121-1 illustrattS rhe systems (Basd2kis 1~110) Steam raises
tne VOC vctpor microressure The resulting sti~am-VOC mixture is condensed and
173
AC-TJA~D CA-e0-l
-----~---
LOW- ffiESiJ~C ~T~H----
PiltOCESS COOL-J~ At-JO oLOWER DRY t-J~ 2gtLOWE~
COgtJDENSER
J
COOLIN(a - --------
WATi=-A
RECOVeRE SOLVENT
WASTE WATER
Fi l 1 1 middot gure middot - Activated-c arbon Adsorption System
179
rout2d to c ~epordtor decanted and returned to storage In vacuum regenerashy
tion VOC vapor is desorbed by pu1 ling a vact1um on the carbon bed then condens-
ed a1d returned System efficiency ~ ~ est ii1a ted at 96 f(duc middot i m from fixed
roof levels (EPA 4503-80-025)
H~fri 9ctated vent concicr1sers dre une of the most cu1mon emission
reduction prJcesses -for conttol ~ ng hx2-d-- storage tai vUC Figure
12 1-- 2 i l 1ust rates a u n i t ( Er i scmiddot n 19HU ) eff i c i en cy of middot middot~ ~ c middotlt middotmiddot _y i s rate ct
bet1Jee~middoti 60-90 for the vapor pressure range of concern For such a large tank
the capital costs would be high ~igure 121-3 illustrat~s EPA estimates
(EPA l ~~SOb) for the condenser section Condenser sys tern a ea mu l d be in
excess of 1000 ftL
It should be noted t -at Jtc1 uf fer Uiemi cc1 l has arH1ounced the c 1 osu re
of its VCPVC plart PrevfoJsly tt12) had planned to purge tlle off-site tanks
of EUC Since any future po~sible micro1ant stdrt-umicro will necessitate a compreshy
hensive SCAQMD review this ltoument can assist in evaluating proposed control
measures
Allied and DuPont Feed Tanks - In bott1 of these cases the nore significant
quantity of emissions arise from tank workinltJ ie during the off-loading
activity rather than tank breathing Contnl ~easures taken for working
emissions are thus of primary concern Ther~fore no detailed discussion will
be provided on the alternatives for control gtf bredthiny emissions Additionshy
ally both tanks are in the range of 20U00 gallons which is a capacity where
floating roof tanks are almost nonexistent Out of 670 floating roof tanks
surveyd by EPA less then 15 were smaller than 30000 gallons in capacity
Therefore the utilization of any floating roof concept and seal combination
will not be considered
Commmerc i a11 y it is estimated by Ou Pont that carbon tetrachloride
feed costs approximately $017lb (E Taylor personal communication)
Ttierefore less than i2500 is lost to l)uPont annud l ly as a result of working
l eve l em i s s fo ns and 1es s fr om Al l i ed bull Thus i t is u n l i kel y that any
appreciable economic incentive exists to dev1~lop ci vapor recovery system
However a Cdndidate system could be pattern~d after the chloroform
feed-storage unit currently at Allied This is a dedicated vapor balance
system Chloroform is off-lndded from tank cars and stored in a closed
unvented tdnk As the 5tordye tank is filled the c1ir spdce d1spldced is fed
Figure 121-3 Installed Capital Cost vs Condenser Area for Various Materials of Cunstruction for a Complete Condenser Section
182
back to the tank car Also in this case the storage tank is not vented in its
normal breathing mode and is part of a closed feed system to the reactor
Vapor recovery system alternatives which are particularly effective
in loading and handling include r~frigeration adsorption andor absorption
Control efficienci~- 1re est1mateJ in the rlt1nge of 90-95 (EPA 1980b) and of
course depend on the speci tic su1)Stance and equipment used Carbon adsorpshy
tion systems were discussed above The smallest capacity carbon adsorptionmiddot
system priced by EPA (EPA 1980b) has 2 vertical beds of carbon (900 lbs - 4ft
diameter by 3 ft depth) witl1 an ilstalled cctpital cost of $135000 based on
December 1979 dollars
Dow Tanks - Dow has two pair of p~oduct check tanks which alternately are
filled dnd off-loaded Emhsions from th~sl tanks were calculated based upon
operating cycle (3 day fi 11) satubull-dtion vapor pressure and physicdl
chardcteristics Direct hec1d-spa1e testing was planned however it could not
be accomodated because of r 1stri c ed access due to unscheduled mai ntendnce on
the field test day
Dow has indicated that chey are studying the option of installing a
vapor controlrecovery systt~m in 1hese and their largerproduct storage tanks
The extent of their engi neri ng ind assessment work is unknown as are their
current plans It may be possibk to incorporate a vapor balance design into
the system whereby the disp1aced apor is trdnsferred to another process point
within tl1e system Alterna1ively a vapor r1middotcovery system such as carbon
adsmiddotorption is fedsible Hot-1ever since emissions for t11e check tanks are
primarily due to working loss tht~ use of a conservation vent or an adjustment
of its operational differential p~essure would be ineffective for tr1e
reduction of the bulk of such emissions Furthermore since check tank size
is re1at i ve 1y sma 11 no cons i derctt I on was given to conversion to a fl oat i ng
roof configuration for those tanks Conversion would be possible for
the large storage tanks
12~ WASTEWATE~ EMISSIONS - STAUFFER
Emissions of me throug11 ~-a~)te-1dter discharge are the largest single
source identified at the plctnt sites The discharge limit was set at 25 ppm
by the Los Angeles County Sc1nitation iistrict and was estdblished with the
occupational limit in mind of 50 ppm JVer an 8 hour work shift dt the District
18~
tredtrient center~ T11ere hctd been suine di~cuss1on between pdrti~s of possibly
raisiny tne discharge limit since it is felt by Stauffer that dilution of the
stream is sufficient to d 11 ow it Itmiddot shou Id he noted t10-1ever tna NIUSH has
recommended the permissible exposure limit be reduced to 5 ppm averaged over a
work period of 10 nours microer ddy 40 ours per Ieek witt1 a cei l iny level of 15
pmicroin dVeraged over a 15-mi nute microet~ v(J
It is not certain at 1~hrt rdte me 1rill be reledscd from the
wastewater stream At the microlant discharge points wastewater is both hot and
aerated thus favoring release~ No measurements have been ta~en downstream of
the plant after considerable dilut1on has occured The distance to the water
treatment pa11t is approximately 5 kin at 1i1ic~1 point anerobic diyestion is
conducted lt is presumed that a11 ELlC will be released before final ocean
di scnar~Ji
A possible emi ss i D~ cont ro process for nmiddotduct ion of the EDC in the
effluent is by ~recess adjust~Pnts or additio~s For example process
modifications to the EUC stripping stage coula dramatically reduce discharge
lev~ls A control alternative is the use of octivdted carnon or XA0-2 resin
to recover EOC in tne discharge stream Tests of Gulf South Researd1
Institute on XA0-2 and activated carbon (Coco 1980) show high recovery yields
for nonpolar organic carcinogens unaer a range of concentrations Viable
suggested alternatives oy Srnitn included regeneration of Ue trapping
materials dnu even consideration of burning tt1e concentrate carbon media (at
greater than 1000 pmicrom)
Control approaches to reduce emissions from so called secondary
sources in general and waste water emissions in particular fa 11 into four
categories - waste source control resour(e recovery alternative disposal and
add-on controls Alternative control pro(~esses which were considered but
appear to be inappropriate to this case i11clude chtmiddotmica means eg
neutralization precipitation coagulatior1 and chemical oxidation thermal
destruction of tne unconcentrated ~dste s1ream is ii1practical biological
treatment eg aerdtion and biomass-wasteater contcct ~ienerally relates to
the treatment of soluole degrddable organics in the concentration range
bet~~een U01 and l~~ terminal storagt eg lanctfilliny urface impoundment
and deep well injection dre either indppl 1cable or 1mmicroractical Therefore in
summary the poss i b I e contra l approac ties fur this ca e inc 1ude the improvement
r
184
of semicrodration efficiencies in steam stripmicroing tl1e internal recycle of ~aste
stredms ctnd the ddsorption b activdted carbon The design configuration
~fficiency and cost )bviously depend upon numerous microlarit specific factors and
their determination vJOuld rec-Jire detailed analyses
n1e ~~dsrewctter systbull-m of a model cl1emical production plant based
upon the average properties uf a composite of 30 chemicals was evaluated for
EPA by IT Enviroscience (EPA 1980b) Included prominantly among the 30 was
tUC with the hignest uncontrolled secondary emission wastewater release rate
ie ~ percent of tt1e production and 34 perc~nt of the emission Cost and
impact analyses were evaluated for alternative control systems to reduce
secondary voe emissions from wastewater Four systems were considered a
carbon adsorption system (eAS) for recovery of the voe from the wastewater a
cover to reduce secondary VOC emissions from the wastewater clarifier a cover
for the clarifier plus a carbon desorption system and a cover for the
clarifier plus ct eAS system usinu a fume incinerator The scale of the model
system was greatly in excess of the Stauffer plant thus further making
detailed comparison i mpract i cal bull However emission reduction factors ~-1ere 6diven as ~9 and cost effectiveness per 10 g reduction genera1middot1y ranged
between $450 to 1733 These factors would likely grossly underestimate the
system cost if scaled down to the range of the Stauffer plant ie the order
of 34600 lb annual discharge
The alternative approaches of improved steam stripping efficiency
and internal recycling of the stripper iischarge stream with a reduction in
makeup re4uirements could decreas~ net ~DC wastewater content release
123 CONTROLS FUR SECONDARY LEAD S~1tLTER STACK EMISSIONS
Given our findiny of low (16 kgyr) ~missions of arsenic from the
reverberatory furndce at RSR it middotJOuld ippear that the arsenic content of themiddot
lead feedstock is low andor that the microants system for reducing lead
emissions is dlso quite effective for arsenic ~SR it will be recalled from
Section 32 uses a quenchinu cnariber ctnd Dayiwuse filters to remove
pctrticuldte matter and a carbonate scruooer to remove sulfur dioxide
There are no federal new source microertormance standards for lead
eini ssions per se t10wever the Standdrds of Pbullrformance for Secondary Lead
SmeHers (40 CFR 6Ull2) limit total particuLte emission from a blast furnace
185
or reverberatory furnace to SU mgm 3 According to Augenstein et al (1978)
who reviewed the techno1ogy for control ing lead ~missions from ttwse sources
baghouse filters or wet scrubbers are yenerally used to contro1 particulate
emissions When fabric filters are used to control blast furnace emissions
they are nurma11y preceded by an afterburner which inciner~tes hydrocarbons
that bull~GU Id otnerwi se b 1 ind t ne f o1 ur-i c After-burners a re nor necessary for
reverb2middotmiddotatcry furnace emiss-ion smiddotnce the excess air and temperature are
usually sufficient to oxidize tn2 hydrocarbons
According to Augenstein et ali 11 sr1aker--type baghouse filters are
the 1nost effective means of controlling ead fume emissions from secondary
furnace opErations 11 Collection etficiencies can exceed 99 oercent One
advantage to this control approact is that l~ad oxide dust can be recovered
easily cmd recycled in the smeller Flue gases must be coated to below 300 degF
for dacro~ bags and to below ~00 degF for fiberglass bags (Hi~h et al 1977)
Although wet scrubbers are effecti1e under some circumstances in
controlling lead emissions it is more difficult to recover the lead oxide for
recycling In addition sulfur dioxide presi~nt in the flue gases becomes
oxidized to sulfuric acid and can cause corr1sion problems For this reason
sodium carbonate or other basic reagents ar~ added to the scrubber solution
Although it concerns a yold smelter an approach described by
Marchant dnd Meek (1980) provides an exampli~ of an arsenic control
alternative which might be apmicrolicatgtle to secmdary lead smelters At the
Campbe1 1 Red lake Gold Smelter in Balmerton Ontarion Canada Smelter gases
are first passed throuyh a hot electrostatic precipitator (ESP) The ESP is
heated so that the arsenic (which is principally in the form of As ) remains2o3 gaseous and is not yet collected This exclusion of arsenic allo~~s the ESP to
recover particulate gold rnore easily The ESP exhaust is then quenched with
ambient air to condense the arsenic trioxide Baghouse filters then remove
the arsenic along with other particulate matter
124 CUNT~OLS FUR STEEL MILL EMISSIONS
Given the imminent and irreversibli~ cessdtion of coking activities
at Ka 15 er Stee 1 Corpora t i on a rev I e~ o r t e c Imo 1 o gi es for ( on t r o 1 l i ng
~missions from the coke ovens rnd he Clt1ke byproduct recovimiddotry pldnt WdS not
deemed to be necessary Tnis is t1e only primary steel mill in California
1Pi6
KEFUENCE)
Al lf~n Jr Ll~ llJBU1 J 111odel to 1middotstimite hi1drdous (bullmissions from coke oven Juurs pr~pdrelt1 DY 1-kSecircn Tr1dnye7nrffut fur U~ Environmental Protection Agency Uffi ce of Air ~ua 1i ty and Standards fltesearch Tri ang I e Park Nortn Cdrolina KTl17362-UL
Jl len Jr CC 1~8Ub tstination of chdr in(i emissions for by-product coke oven microrepared by fltesedrch Tr1anglt Insc1tute for US nvironmental Protection 1-yency Uf f ice of Air l)ua l i ty and Standards Ke search Triangle Park North Cdrolina RTI17362-0
Allen Jr CC 198Uc 11 Environmeritdl c1ssess1ent of coke by-product recovery plants Proceedings First Symposium on Iron c1nd Steel Pollution Abatement Technology Ctncayo Illrno1s--Tin1ctober - 1 November 1979 EPA-600-9-80-012 pp 7~-dd
Augenstein ur1 T Cornin et al 1978 Con_trol techniques for lead air emissions prepared by PEDCU Envircnmental Inc for US Environmental Protection Agency Research Triangle Park North Carolina EPA-4502-77-012-B ( NT IS P88U-197551)
jarber WC 1977 Interim air pcllution a~sessment for eight t1igh-volume industrial organic chemicals 11 US Environin~ntal Protection Agency memorandum to Air and Hazardous Materials Uivision Kegions I III-X and to Uirector Environmental middotPrograms Uivision Region II
t3drrett KE WL Margard et al 1977 Sampling and analysis of coke-oven door emissions Prepared by ~dttelle-Columbus Laboratories for US Env 1 ronmenta I Protection Agen y Industri a I Environmental Research Laboratory Research Triangle Park North Carolina EPA-b002-77-213
t3 a r ret t flt bull r bull and P bull K bull ~~ebb 1Y78 bull 11 E f f e ct i venes s of a wet el e ct r o stat i c precipitator for controlling PUM emissions from coke oven door leakage Presented at 71st Meetiny of the Air Pollution Control Association Houston Texas 25-29 June 1978
l3asdekis HS 1980 Carbon Adsorption Control Uevice Evaluation IT Enviroscience Inc report in prepdration for US Environmental Protection Agency ESEU
8ee fltW et al 1974 Coke oven charging emission control test program Vol I c1nd II US Environmental Protection Jgency EPA-6502-74-062
~eimer RG 1978 Air pollution emission test Analysis of polynuclear aromatic hydrocarbons from coke oven effluents Prepared by TRW Uefense and Smicrodce Systems Groumicro for US tnvirorunental Protection Agency Office of Air wuality Planning and Standards Kes~arch Triangle Park North Carolina EMB Kemicroort 8-CKU-12
187
Braman RS 1976 Application of the arsine evolution methods to environmental analyses presented at the International Conference on Environmental Arsenic Ft Lauderdale Florida 5-8 October
Bratina J E 1979 11 Fabric filter applications on coke oven pushing operations J Air Poll Cont Assoc 29(9) lll6-920
tiuonicore AJ 1980 Environmental assessment of coke quench towers in proceedings First Symposium on Iron and Steel Pollution Abltlternent Technology Chicago~ Uirnois JO October - f N~vernber 1979 EPA-6009-~1b-62~ pp 112-142
t3usse A D and dR Zimmerrao1 1973 User 1 s guide for tdegh~ Climatological Uisper~ion Model US Environmental Protection Agency Resea~ch Triangle Park Nortl Carolina) EPA-R4--73~0024
California Air Resources 8oard 1976 Coke oven emissions miscellaneous emissions and their control at Kaiser Steel Corporations Fontana Steel making facihty Enforcement l3ranch) S2crcmento California L amp E-76-11
Calvert~ CA 1976 Envir __ Sci_ i( __ Technol 10256
Coco L~i 2t aL~ 1979~ 11 02c10pmert of trec1tment and control technology for 11refractory pet rochemi co was tt=1s r middotl 1d l Report Gu 1f South Research In st it ute
to U S tnv i ronmenta l Protection Agency EPA-ti002- 79-080
Colucci~ LM and CR Begemi 1971 Palynuclear aromatic hydrocarbons and other pollutants in Los Angeles air In Proceedings of the International Clean Jir Congress Vol 2 Academic Press pp 28-35
Cooper JA and JG Watson Jr 1980 Receptor oriented methods of air partkulate source apportionment 11 J Air Pollution Control Assoc 30(10) 1116-1125
Coutant R W JS McNulty and RD Grammar 1975 Final report on determi 11at ion of trace elements in a combustion sys tern Prepared by Batte11 e Columbus Laboratories for the Electric Power Research Institute EPRI 122-1
11 11Dilling W 1977 lnterphase transfer processes Envir Sci TechnoL 11 4 pp 405-409
Oow1irgl) MP JO Jeffry and AH Laube 1978 11 Reduction of quench tower emissions Presented at the 71st Air Pollution Conotrol Association Conference Houston Texas 25-30 June APCA No 78-92
EPA 1980a Review of Criteria for Vapor Phase Hydrocarbons 11 US Environmental Protection Agency ~eport 6008-80-045
EPA 1980b Organic Chemical Manufacturing Volume J Storage Fugitive and Secondary Source EPA report 450J-80-025
EPA 1981 Compilation of Air Pol lutdnt Emission Fdctors (Including Suppleshyments 1-7) Third Edition Suppl1111ent 12 Utfice of ir l)uality Planning dnd Stdndlt1rds AP-42-ED-3-Supp I - lt
188
Erikson Uli 1~80 11 Control Uevice Evaluation Condensdtion 11 IT Envioscience lnc report in prepdration for US Environmental Protection Agency ESEU
lrtel liL 197Y Quench tower particuldte emissions J Air Poll Cont Assoc 2Y(9) 913-916
Ev a n s K bull P bull 1981 Pe r son a l Commun i cat i on v i th R bull Zi s k i n d bull
Goodwin DR 1980 11 Air pollution emissions standards in Proceedings First Symposium on Iron and Steel Pollution Abatement Technology Chicago 11 lrno1s 30 October - 1 November 1979 EPA-6009-80-012 pp 37-45
Gordon GE 10 Receptor Models Environ Sci Tech 14(7)792-800
Gordon KJ 1976 11 lJistribution of airborne polycyclic aromatic hydrocarbons throuyhout Los Angeles Envir Sci Technol 370-373
Great Lakes Cdrbon Corporation 1977 Study of coke side coke-oven emissions~ Vol 1 Source testing of a stationary coke side enclosure EPA-340l-77-014A
lirimsrud EP and HA Rasmussen 1975 Atmos Envir Y 1010
Hartnan MW 1~80 Source test c1t US Steel Clairton Coke Ovens Clairton P~nnsylvania Prepared by rRW Environmental Engrneerrny D1v1s1on for US lnv1ronrnental Protection Agency Emissions Standards and Engineering l)ivision Research Trianyle Pdrk North Carolina EM~ Report 78-CKU-13
Hendriks Ilt V et a I 1Y7Y Urgani c air emissions from coke quench towers 11
Presented at 2nd Arrnual M(etiny ot the ir IJol lution Control Associdtiont Cir1cinndti Utlio June 1~7lJ l-dmicroer NU 9-391
Higt1 MU ME Lukey dnd TA Li Puma llJ77 Inspection manual for secondary lead smelters premicrolt1red b Enyineeriny Science lnc for US lnvironmentdl Protection Ayency Office 0f ~nforc~nent EPA-3401-77-001
Holzer G et al 197 11 Collection amp analyses of trace organic emissions from natura I sources 11 Journal of Cnromatogra flhY 142 pp 755-764
Science Applications lnc Final lemicroort to US Environmental Protection Agency tPA-560lJ-80-UU2
J bull Mi l n e l lJ 81 Person a I Cornm un i cat i on wi t t1 bull Zi ski nd bull
Kemner WF 1979 Cost tffectiven2ss model for pollution control at coking facilities Premicroarect oy PEDCo Environmental lnc for US Environmental Protection Agency Industrial Environmental Research Laboratory Research Tridnyle Park Nortll Carolind EPA-GU02-7l-1U5 (NTS PB 8U-118706)
Kessler T AG Sharkey Jr and kA Friedel 1973 Analysis of trace ele111ents in coal by smicroark-source mass spectrometry US 8ureau of Mines Report ot I n v es t 1 gd t 1on s No bull 77 14 bull
189
Kolak NP J Hyde and R Forrest~r 19 1 9 Particulate source contributions in tt1e Niltlgara Frontierffi Prepared t)_y Nr2bulll York State IJepartrnent of Env i ronm~ntc i Con serv d ti on for US En vi r inment ct l Protect ion Agency Region 11 EPA-9024-79-006
Mackay U and PJ Leinonen llt175 Hate of Evamicroordcion of low - solubility contaminants from vJater bodies to c1tmospmre 11 Envir Sci Tecnnol 9 13 pp 1178-libU
Magee tiv HJ Hctll and GM Vc~rlt_a Jr 1973 Potential Doimiddotiutants in fossi1 f1Jel Prepared by ESSO fbullt~scarcr middotnd Engineeriny Co tor US Enviromihm~ct Protection Agency Ufl-R2-7 --2l1~L (NTIS Pt3 as u~9)
Marcriant middotG0H ard RL Meek 1~gu Eva uation of technology for control of arsenic tmissfons at the CdlmpJel Hd Lakmiddot G~ld Smelter prepared for the US
1Env i ronmentd l Protection Agenc)-l-l(lU-sfrT l lnv i ronmenta l kesedrCiI Laboratory Cincirrn2ti Uh70 lYA-60U2-8U--141 (NTIS 8 oU-21Y36J)
Margler L$ HA Ziskind M8 lcgozen (Ind M Axelrod 1979 11 An inventory of care i nog)n i c substances re i ea sect into ne ambient air of California Vol une I - Scn~enin~ and id2nt~fication o-f r~1rci1ogens of greatest concern Science J- pp 1i ca middott~ L J s In c bull Fi nd l kcmicro~ rt ~ i I - U 6 C) - U- 5 0 4 to U1 e Ca I i ~- o r n i a Ai r kesources 130ard
Milne J 198L Persond1 C~mmunicdtmiddotion middot1itt1 R Ziskind
llurchio JC WrC Cooper and A ue Leo11 1970 Asbestos fibers in ambient air of California 11 Cal Horn i ct Air lesourc2s l3oa rd ARl3 4-0S4- l EHS Report 73-2 (March)
National Academy of Sciences 1Y72 Part1culate polycyclic organic matter Washington~ DC pmicro 4-ll
Nati ona I Kesedrcn Counc i I 197b Arsenic prepi1 red by Subcommittee on Arsenic Co111mittee on 1bull1edical ctnd t)ioloyicai Effects of Environmental Pollutants washin~ton uc
Oliver JF and JT Lane 1979 11 Control of visible emissions at CFampl 1 s coke plant-Pueblo Colorado J Air PolL Cont Assoc 29(9) 920-925
Pe I I i z z a r i E bull 0 bull anct J bull E bull l$ u n c h 1s 19 1 nbi en t ai r ca r c i n o gen i c v a po r s Improved samplin~ and analysis tectrnology EPA Contract 68-02-2764
Pellizzari E U 1977 11 Tle 11easure11ent of carcinogenic vamicroors in ambient middotatmospheres EPA-b007-77-J55
Phelps RG 1Y79 lSI-CPA-oatLlle c-ilte oven door sealing proyram J Air Pol 1 Cont Assoc 29(J) JlJ8--JlL
Porter RA 1976 Uismicroersion ~yuation s lucions Dy calculator I guide for air pollution engineers ctnd sci12nthts S cond edition rexas Air Control tioard ~MT1S P~ 262 lJ~)
Redburn T 1980 Kaiser battles aginst l~gacy of woes 11 Los Angeles Timesmiddot (7 Semicroternber irno) Part VI pp 111
Research Triangle lnstittJte 1977 11 Qudntification of benzene in 15D -1bullnoient air samples 11 for US Environmental 11rotection Agency Office of Air Quality Planning a11d Stitndards
Rinkus SJ and MS Legator 1979 Chemical characterization of 465 known or suspected carcinogens and their correlation with mutagnic activity in the Salmonella typhimurium system Cancer Research 393289-3318
Koberts R M 1980 11 An inventory of arc i nogen i c substances r~ leased nto the ambient air of California Tasks 11 and IV KVB Final Report KVB 26900-836 to Science Applications Inc
Rogozen MB LW Margler P Mankiemiddott1icz cind M Axelrod 1978 Water pollution control for coal slurry pipeiines Prepared by Science Appl1cat10ns Inc for US Department of Energy (NTIS SAI-068-79-516
Rogozen MB DF Hausknecht and RA Ziskind 1976 Methodology for ranking elements in fossil fuels according to their potential health impact Science Applications Inc Final Report 260-77-539 to the Electric Power ~esearch Institute
Russel 1 JW and LA Shadoff The sdmpl ing and determination of halocarbons in ambient air usin~ ~rncentration on porous polymers
Siebert PC CA Craig and Eb Coffey 1978 Prelimary assessment of the sources control and population to airoorne polycyclic organic matter as indicated by benzo(d)pyrene Prepare( by lnergy and Environmental Analyses Inc for US Environmental Protection Agency Office of Air Quality Planning and Standards Resedrch Triangle Park North Carolina
Simmonds PG et amiddot1 1974 Distribution of atmospheric halocarbons in the air over Los Angeles basin Atmos Envir Vol 8 pp 209-216
Singh HB LJ Salas and RE Stiles Distribution of Selected Gaseous Oryanic Mutagens and Suspected Carcinoyens in Ambient Air in Proceedings Annual Meeting Air Pollution Controrl Association 82-651 June 20-25 1982 New Url eans
Smith WM 1979 Evaluation of cok oven emissions in Yearbook of the American Iron and Steel Institute pp 163-179
SRI International 1980 Asessment oi human exposures to atmospheric benzene SJ Mara and SS Lee finai report to US Environmental Protection Agency EPA-4503-78-031
Taback HJ 1978 Control uf Hydrocarbon emissions from stationary sources in the California South Codst Air Basir Prepared by KVB Inc for the Cal1fornid Air Resources Bod-rdmiddotKVB No 5804-714
191
Trenholm J-f dnd LL )eek 1Y78 11 Assessment of t1azardous organic emissions fror1 slot type coke oven Datteries unpublished paper US Environmental l)rotection Ayency Ernssion Standdrds and Engineeriny Jivision
Turner U~ J970 Workbook of atmusphe~ic dispersion estiimiddot1Jt2s IJS Environmental Protection Agency k~sedrch Tri1ngle Park North Carolina
US Bureau of Mines lYl f-inercls yearbook 197U Vol 11 p 4~3
Van Os dE I l U bull iJ bull lL Ma rs l and et ct 1Y7~1 o En i r onmt~ nta I a s s es smen t of co k e Dy-product recovery plants prepareu by ReseMcn Trictngle lnstitute for US Environmental Protection AgerKy 11cJustrial Environ11ental R~search Laboratory Kesearcll Tri2n~e Idrk NorU1 Caroline EPA-6U02-7J-Ul6 (NTfS PB 293-278)
Westbrool( C W 1979a Level l ass~ssment of uncontrolled sinter plant emissions Prepared by fesearch Triangle Institute for US Environmenta1 Protection Agcncy Industria1 tnvirnnmental fltesearc-1 Laboratory kesearch Trianyle Park North Carolind EPA-bOU2-7Y-112
Westbroollt C~~ 197ltJb Level l css~ssment of uncontro1led l)-BOP emissions Prepared by Researcn Triangle institute for US Environmental Protection Agency Industrial Environmental Hesearch Laboratory Research Triangle Park ~Orth Carolina EPA-6002-79-190
Wetherold rLb LP Provost and CD S111ith 198U Assessment of atmospneric emissions from microetroleu111 refinin~ Volume 3 Appendix l3 Radian Corp Fina Report to US tnv i ron111enta I Protection Agency EPA-600 2-80-07 5c
Ziskind RA OF S1ilith JL Hdhn etnd G Smicroivey (1980) Determinants of Cancer and Carctiovasculdr Uisease Mortality in Asbestos Mining counties of California SAi Report 068-dl-514 l May
192
APPENDIX A
Rapid Screening and Identification of Airborne Carcinogens of Greatest Concern in California
Lawrence W Margler Michael B Rogozen Richard A Ziskind and Robert Reynolds Science Applications Inc Los Angeles California
This paper describes a method for establishing a priority list of airborne carcinogens within a
state jurisdiction In this case it was necessary to identify from among hundreds of potential
candidates the live to ten materials of greatest potential concern In California as airborne
carcinogens
Because no previous Inventory of carcinogens In California existed published lists rankings
and assessments of national scope were used to Identify candidates By systematic manipulation
and comparison of these data sources 47 materials of some notoriety were chosen for closer
scrutiny This selection was pared to 22 candidates largely by eliminating those which had
very little production and use In California (Substances primarily used as pesticides were
excluded from the scope of this study) The remc1ining candidates were then ranked by additive
and multiplicative algorithms and by a panel of experts The results of these rankings were
combined to produce a single selection of 11 priority candidates In alphabetical order they
ethylene dichloride N-nllrosoamlncs perchloroethylene and polycyclk aromatic hydroiarbons
In continuing studies a baseline emissions inventory Is being prepared and a source testing
program is being designed
In recent years concern has grown over su~pected arcinug-ens New Jersey seshythe possibility that certain materials lected ten volatile organic compounds released to the atmosphere through inshy and five htbullavy metals to be examined dustrial and commercial activity may be further and is currently measuring responsible for a significant portion of ambient atmospheric concentrations of the incidence of cancer in the general these substances in a variety of areas In population This concern is manifested the second year of the study the state at the federal level in the National h1s incre1sed the volatile organics Emissiltm Standards for Hazardous Air stt1died to 20 and begun measuring Pollutants (NESHAP) which limit heavier orianics associated with parshyemissions of the known carcinogens asshy ticulatcs2 In California a very different bestos beryllium and vinyl chloride 1 approach-was taken
Only a few states including New Jersey and California have begun efshy Overview of Californias Approach forts to identify airborne carcinogens of concern to the general public for the The California Air Resources Board purpose of setting state emission regushy (( ARB) is sponsoring a three-stage lations for these substances After reshy st 11d~bull of airbornC rmissions of carcinoshyviewing national use data for known and g1middot11s from anthropogenic atlivities The
November 1979 Volume 29 No 11
first stage which is the subject of this paper was to identify roughly five to ten materials which of the hundreds of known or suspected airborne carcinoshygens are most likely to be of greatest concern to Californias general populashytion Also of interest were those which in order to satisfy occupational health and safety regulations might be transshyferred from the workplace air to the outside atmosphere The second stage which is now underway includes pinshypointing of emission sources for each of the carcinogens of importance quantishyfiltation of emissions nnd design of source-testing methods A subsequent stage will consist of source testing and measuring public exposures to those substances for which data are unavailshyable The basis for regulatory action if appropriate will include the results of this program and other related reshysearch
Screening of Candidate Carcinogens
The National Institute for Occupashytional Safety and Health (NIOSH) lists 1905 chemicals which have reported neoplastigenic or carcinogenic effects and 510 which have otherwise received attention for their neoplastigenic poshytential1 The need to select five to ten materials from such a large number of potential candidates dictated that we devise a way to rapidly eliminate from further consideration the vast majority of the substances Given the paucity of published data on most of the candidate substances the screening method was designed to make best use of readilymiddot
C1111yrich1 19i9-Air Pullulion Control A~~ociatlon
1153
A- I
----- ---- ----- ---------------------available information The screening process was as follows (1) eight comshyplications of known and suspected carshycinogens were reviewed and those subshystances which were not used in Califorshynia were highly unstable in air or were very doubtfully carcinogenic were eliminated (2) after more detailed inshyformation was obtained for the reshymaining 25 substances candidates were rated by two different analytical methshyods (3) an expert panel was convened to review dossiers on the candidates and to independeitly nmk them (-i) from the eight to deven substances ranked highest by all three approaches eleven were selected for the emission identifishycation and source-testing design stages of the CAHB effort
lnitiI $cre~ng ot Potential Candidate Carcingtgens
65 compnunds middotwe ire selected from 642 industrial ur~nnic air pollutants comshypiled by MrTRE Corporation for the US Envircmnental Protection Agency (USEPA) in U1at study pollutants were scored by multiplying four explicshyitly defined rating factors annual US production fraction of production lost to the environment volaflity and toxshyicity To adapt this york to our pmpose we first selected the 114 substances listed as being carcinogeni 1~ or neoplasshytigenic Then the scores of each of these compounds under the criteria annual US production fraction of producshytion lost volatility and carcinogeshynicity were multiplied together Seshylected for further consideration were those substances which had a product score above 50 Another 15 substances listed as being carcinogenic but lacking information for one of the other rating factors were also selected This list of 65 was then compar~d with seven other lists of corcinogeris1
-middot 11 Materials comshy
mon to the reduced MITRE list and at least one of the other lists were chosen for further consideration Added as candidates were those suhstances which are regulated as occupational carcino-
gens by the Occupational Safety and Health Administration (OSHA) and certain inorganic carcinogens Finally substances were added which in our judgment should be investigated but had been eliminated at this point Exshyamples of these are bis(chloromethyl)shyether epichlorohydrin and hydrazine
Next the refined list which now contained 47 substances or chemical groups was pared further hy another rapid screening process Eliminated
were all candidates (1) whose production andor use in California was very low (under 10~ lbyr) and wns not thought likely to pose a risk to a localized popushylation (2) which are very unstable in 1ir or CH whilh should not on t lw hnsis of CUtrltbullnL cidlnre liebull considtbullnmiddotd r11rci-
1154
Tahllt I Suh tances n-iewed in letail
Candidate Substances
Arsenic Inorganic lead Ashestos Alkyl lead Benzene Maieic anhydride
nogenic The result of the initial the rating under that criterion and the screening was a list of 22 candidate mashy corresponding criterion weight The terials which is presented in Table I overall rating for pollutant i is then the
sum of the scores under all the crishyteriaRanking Candidates by AddiUve ancll
Mulllpllcatlve AigorUhms The additive approach has several virtues the main one being that it forces
Many screening or ranking systems the user to make all assumptions exshyfaH into one oft wo cat~middotgories additive plicit In the process of setting up such and multiplicative Some systems are a a ranking system new insights into the combination of the two while others problem under consideration may be combine nn ol1jective appronc 11 with g1ined Once the system is set up it is suhjlctive cvnluition of the remiddot ults 1l rdatively easy to use Where data for The L2 substances surviving th1middot initial scoring pollutants are unavailable arshyscreening were ranked independtmiddotntly by tificial scales can be constructed to the tvo approaches If the same subshy quantify subjective factors Finally the stance rated high~middot under both systems sensitivity of the results to the systems its importance to California was judged subjective aspects may be measured For to be more likely than if it had scored example one can determine the effect of highly in only one method changing criteria weights upon the final
Additive Approach In the additive pollutant ranking Similarly an appreshyapproach the user identifies one or more ciation may be gained of the significance criteria and rates each alternatiw subshy of the range of uncertainty for a particshystance against each criterion while sishy ular required data element by varying _ multaneously deciding the relative imshy rating factor values portance of the criteria Eq (1) shows its A fundamental problem with the apshymathematical formulation proach is that there is no logical basis for
adding the individual scores assignedRating for pollutant i = f W1 R11 under the criteria other than the asshy
rbullI sumption that this simulates or even
(1) improves upon the users thought proshyEach criterion or rating factor (1 ) is crss A major operational problem is assined a value for each pollutant and that of weighting the criteria A common each rating factor is weighted (b W1 ) practice is to give all critNia equal -icnrding to its importance reht veto wltgtight hut that is in itstbulllf a st1Ubullm1middotnl lhe otlwr rritnia llw sc11rtbull for 1111 aLnut the 1rcl11tive importance of Ow 1n1 1 undmiddotr nitn1 n j is the prod11middott of n1ttria
Journal of the Air Pollution Control Association
A-2
---------
Multiplicative Approach In the multiplicative approach the rating for each alternative is the product of the rat ihgs under each criterion
Hating for pollutant i --- 11Ill
UJ (~) Jbull I
A multiplicative approach can have ~omc altlvrntnges over additive onrs First in some cases the ratings can be physical parameters such as concenshytrations or volatilities there is then no need to weight the criteria and hence less controversy over subjective judgshyments Second multiplication generally provides a wider range of scores than does addition allowing clearer disshycrimination among alternatives Finally this approach provides results which are more intuitively acceptable As an exshyample of this last point suppose that exposure and harmfulness levels for candidate substance are each converted to values on aO to IO scale and that a certain substance is both extremely toxic and extremely rare An additive approach would give the compound a rating ofO + 10 = 10 which is equivalent to that of a moderately prevalent subshystance (rated say at 5) which is modshyerately harmful (rated also at =-) A multiplicative approach on the other hand would rate the first substance at 0 and the second at 25
Criteria The six criteria used in the additive and multiplicative ranking procedures are defined in Table 11 Asshysignment of values to the Rij was based upon data gathered from published litshyerature personal communications and panel discussion and has been fully documented 13
Because the purpose of this exercise was to determine the relative imporshytance of the suspected candidate carshycinogens R 1 was scaled to the most heavily used candidate substance benshyzene whose annual production and use in California is nearly 109 lb Materials with a use under 105 lbyr would be rated zero for R 1 and rejected Howevshyerbefore rejecting a substance by this criterion we considered whether its emissions could in particular circumshystances result in high exposures to a Joshycalized population
R2 takes into account the fact that the chemical industry is in continual change Substances of concern today may be phased out while the use of others may rise dramatically increasing their imshyporta nee asmiddot poll utan ts Information on developments which could likely result in a change in the growth rate was facshytored into the choice of a value for R2- As an example asbestos consumption in California has been stable in recent years However the pending phaseout of asbestos in motor vehicle friction materials will hasten the decline in asshybestos consumption hence we assigned a value of 1 for R2bull
November 1979 Volume 29 No 11
Ideally one cou1 1 use pouu1c11u
sion as a criterion However in this case emissions were to l e estimated in detail only for the five t1 ten carcinogens fishynally ~elected Thmiddotrtfor(bull a 11w1st1n of tt II iim pol cint ial vus 1~1middotd 1atbulld llpon
knowledge of the s11bstances manufacshyture and USP for R The higlwst rating went to substanc middots which are v idely used especially in consumer products A slightly lower rating went to subshystances which an~ routinely emitted from industrial processes during proshyduction and use Some materials are employed in such a way that emissions are quite low even though tight emission control may not be required by law Materials in this cntegory were assigned a value of two for H3bull Substances which under federal or st ate regulations may not be discharged to the exterior envishyronment but which could be dischJrged by accident received the lowest rating
Each candidate was evaluated (n the basis of its 1ropen-ity to decompbullgtSe in ambient air Materials with half-lives greater than eight hours were considered moderately to highly stable and rated five for R4 bull Low to moderate stability was assigned to substances with halfshylives between zero and eight hours Compounds known to exist in air for only a few minutes would be rated zero
Ta hie II Dcfinitinns of the criteria used
R1 Prbullbullsent use in California
100 of max (109 lbyr) 5 10 1)f max (108 lbyr) 4
1 of max (107 lbvr) 3 01 of max (106 lbyr) 2
001 of max (105 lbyr) l lt001 of max (lt105 lbyr) O
R2 Growth in California use
+ 20 5 +Oolti to +20 4 Positive growth to 0 3 Stahle or unknown 2 DPcline 1 B1middotin~ phased out 0
R Emision potential
Widespread use in ltonsumer products 5 R1middotlatively p1or con rot over emissions 4 Rdatively good control over emissions 2 Tihtly c1int rolled 1
R4 8tabilit~- in ambient Air4 tvlnderale t1 high st 1bility (l 1 middot2 gt 8 hr) 5 Lw to modmiddotrate s1 bility (t1 l -0-8 hr) 3 Unstahle(t2~febull minutes) 0
R Dispcision Potential
Emitted lar~ely as apor or iine 5 particulnte
Emit led largely as rnarse particulate
Rlt Evidence f Carcinu~enici ty
Known or suspected human carcinogen 5 Known mammalian carcinogtmiddotn 4 ~usptctcd 111ammalan carcino~en or 3
known m mmalin muta~tmiddotn Ames ltmiddotst p sitive 2 Prpoundgtcursor 01 co-car inogen
a t11 is the half-liftmiddot
A-3
tion state or anion associations may change in the atmosphere metals do not degrade and were considered stable Asbestos is likewise itahlc Many of the clcc-ompo~ilion rtbullnctions oi orinnic molecules are mediated by light Such substances if released at night would have several hours to disperse in the surrounding area
A rapid way of assessing the relative potential of different substances to spread from a release point is to note their physical state under normal amshybient conditions Accordingly we scored materials emitted as vapors or fine particulates the highest for R5 and coarse particulates the lowest Intershymediate values are possible for varying amounts of fine and coarse particulate emissions from the same source or from different sources
There is as yet no widely agreed upon measure of the relative potenciemiddots of carcinogens although some ranking systems have been proposed14 Extrapshyolating data from in vitro techniques such as the Ames bacterial mutagenicity test and from laboratory animal studies to humans is problematic Therefore a less quantitative measure of the carcishynogenic potential of each candidate substance was used The candidates receiving the highest scores for Rs would be those for which there is strong evishydence of carcinogenesis in humans Exshyamples are asbestos which is implicated in mesothelioma vinyl chloride which has been identified as the agent of liver cancer in exposed workers and bisshy(chloroinethyl)ether shown by epideshymiological studies to cause lung cancer in resin workers The next highest rated substances are those for which human carcinogenicity is unknown but which have produced cancer in one or more mammalian species in laboratory tests Next are those which have not been shown to be carcinogens but which have proven to be mutagenic in test animals Substances for which the only knowlshyedge of carcinogenic potential is a posishytive Ames test (producing mutations in histidine-requiring strains of Salmoshynella) are rated 2 Finally substances which are implicated only as precursors or co-carcinogens would be rated lowest
Substances unequivocally associated with carcinogenesis were considered as carcinogens in this study Conditions of emission and exposure including the presence of co-carcinogens were facshytored into the evaluation of each candishydate where possible Carcinogenic subshystances derived from the metabolism of a precursor were considered as carcinoshygens However ubiquitous substances which have been hypothesized to be precursors (eg secondary amines nishytrous acid and nitric oxide which combine under certain circumstances to
1155
form N-nitrosoamines) were not conshysidered because of uncertainties in the importance of their link to the carcinoshygenic compound and the practical conshysiderations demanded by the scope of the study
It was beyond the scope of this study to jud~e the validity and interpietation of the experimental ad epidemiological evidence upon which the carcinogenicity of candidate substances has been esshytablished We accepted the conclusions about carcinogenicity drawn by the Inshyternational Agency for Research on Cancer or the National Cancer Institute and did not consider dosage or route of administratioll of tested substances However considerations of test validity did enter into the subjective evaluation by the panel of experts
Weight lr1 the sdditive ranking scheme each rating criterion is weighted according to its limportance relative to the other criteria Little precedent exists for assigning these weights In our judgment and as generally agreed by the panel of experts (see below) R 1 RJ and R6 are more important than the other criteria Evidence of carcinogeshynicity was considered to be the most important criterion of aB so W 15 was assigned a valtze of 3 W 1 aid W 3 were set at 2 and W W-i and W5 were set at 1 In order to discern the potential senshysitivity of the rankings to the weight assignments the candidates were also ranked using equaily weighted crishyteria
Ranking Candidates by Panel o~ Experts
Anine-member panel of experienced governmental industrial and academic scientists whose disciplines include~ -organic and physical chemistry indusshytrial hygiene toxicology epidemiology and iegulatory control of toxic sub~ stances was convened to provide addishytional data for our ranking algorithms
to discuss our candidate --ubstanc -sand rejections to suggest pos-ibie ne suushystances for consiceration and Lt rank the candidates indepen11mtly ( f our own ranking Tvo v-ee ilts he fore the meeting pa-d uember-s were jven one-to three-page dossiers on eac I canshydidaie substance
At th~ start of the meeting- before any group discussion the pand was asked to rate each candidate suhtance with a score from Oto 5 Next cmiddotach candidate was discussed at length e provided ~n ovrview and summarized critical issues identifi2d up to that Pbullgtnnt Through materials brought to thtmiddot meetin and their persoril experience panel memshyber Veimiddote able to provide much useful information on the candidates and adshyditional insight in to our ~a ting criteria
At the encl of the two-d1y sc~sion the pa-iel ugain -ated the cai 1didates
Flnal SelediDn
Tabe III show- the highest-scoring substance a~ det(rraine1 by the 1ddishytive and mlltiplictive approaches and by th~ panel in ~e additive approach 21 single r1nking as obuined by avershyaging the two ran kings resulting from using equal and uneguai weights The ran kings ltif most candidi tes were unafshyfected hut carbo11 tetralt hloride chloshyroform chromium and inorganic lead changed more tban thmiddotee positions_ Because uncertainties in the data base predude imputing signifiance to small differences in the Lnal ordering the lists in Table III are prtmiddotsented in alphabetishycal order However it is of interest to point out that benzene consistently ranked hihest
Becausimiddot some cindidates had equal rating scores we ltould nut choose exshyactly ten candidat ~s from the additive and multiplicative rankings Instead the
top nine and eleven were selected from the two exercises respectively We also considered the ten substances scored highest hy the panel at the end of the session The final consensus selection cornsistt-d of the 11 candidate substances appearing on at least two of the three lisL- For the substances included in the cun~ensus ranking a baseline emissions inventory is being conducted and a source testing program is being deshysigned
1ejected Substances
Some comments about certain subshySiF1ces not appearing on the final list are in order inasmuch as they include known carcinogens and compounds which hae received considerable atshyllttion in recent years as occupational c11c111ogens
rouisionally Rejected Substances Appended to the consensus ranking (T1ble III) were vinyl chloride gasoline and engine exhausts tobacco smoke and pesticides No further action by the CRB is recommended at this time for vinyl chloride because it is already subject tu the USEPA emissions stanshydard a CARB ambient air quality strndard and an OSHA standard
Gasoline and tobacco smoke were appcndlmiddotd to this list because each ocshycurs very widely and contains several of the candidate substances reviewed in this study some of which are in the final listing For example gasoline contains benzene ethylene dibromide ethylene dilmiddothloride and alkyl lead compounds the last three being in leaded grades only Both gasoline and diesel combusshytion products include PAHs Tobacco smoke contains among other neoplasshytigenic substances nitrosamines PAHs nickel arsenic cadmium and other heavv metals Manv individuals are inshyvolu~tarily and i~ many situations
Table HI Hi~hest ranked candidates from each rnnkin~ methodbull
Highly ranked but no inventory
l I ighesl consensus recommended Additive Multiplicativ(~ Pan I of eqwrts rnnkmg at this time
1156 Journal of the Air Pollution Control Association
A-4
lirtualy unavoidably exposed to t11-k1ltco smoke Because the sources of gasoline its combustion products and tobacco smoke emissions are well known no specific action was recomshymended for these materials during the tbullmissions innnlory and -ource testing design stages of the present study It was considered importint ho-vemiddoter to draw attrntion to the general publics exposhysure tot hrse substances PrsticidltS are listed for the same reason thou~h ad(bullmiddot tailed examination of pesticides w1s beyond the scope of this study Many pesticides are widely used and some of them are known to be carcinogenic
Other Rejected Substances Acryshylonitrile and vinylidene chloride were placed in a provisionally rejected group because of the panels suspicion that imports of these compounds to Cnlifornia from Japan may be appreshyciable yet are hard to substantiate Should such imports be verified in the future these two compounds would take on greater importance Formaldehyde was also provisionally rejected because the preponderance of evidence indicates that it is not carcinogenic and that bisshy(chloromethyl)ether is not formed from formaldehyde in appreciable quintities in industrial environments 15 The ocshycupationally controlled carcinogens ethyleneimine and beta-propiolactone were rejected in part because of their reactivity in air At the time of this study DBCP a pesticide was no longer heing middotproduced in California and was therefore rejected from further considshyeration
Occupational Regulations and Community Exposure
A question of interest to the CARB was whether the regulation of acknowlshyedged occupational carcinogens adshyverselv affects the ambient air outside the v~rkplace Our general findings can be illustrated by the example of asshybestos
Asbestos is a very widely used mateshyrial for which no ambient air standard exists Concentrations in the workplace are limited to an eight-hour timtmiddotshyweighted average concentration of two fiberscm=1 of air and a ceiling concenshytration of ten fihcrscm 1bull In meeting this standard exhausting air containin~ asshybestos to the ambient air is not reshystricted except by the USEPAs reshyquirement that there shall he no visible emissions containing asbestos piut ides from s11eh foci lit i1bulls (bullxd11din~ hrak(bull shops 1 Considning that undc-r certain conditions 101 asbestos fiherscm1 could ht Pmitted without being visihl( 11 this standard mamiddot allow considerable asshybestos emissicns It is unlikely however that emissions would actually approach these levels First since the OSHA standard cannot generally be achieved
November 1979 Volume 29 No 11
by ventilation n ine the generation or asbestos partid in the workplace must be greatly reltlu ed Second the air is usually filtered 11gt prevent recirculation of asbestos to 1 workplace Asbestos waste must bed 1sposed of in sealed imshypermeable bag~ or containers Thus under current onupational regulations the amhient air 1enerally appCars to be afford(bulld greatn protection than it would without s11ch regulations 1
Conclusion
The screenin and ranilting methodshyology presented in this paper proved to be a feasible approach to establishing a priority list of Jirborne carcinc)gens in California We ftel that it is an efficient means of focu~ing further efforts on emissi()ns inventories source testing and ambient measurement for it not only identifie all the carcinogens of potential concern but it also permits the state regulatory agency to direct its reshysearch resourCtmiddots toward those subshystances of partilmiddotular interest within its jurisdiction
References l National Elllission Standards for Hazshy
ardous Air P llutants A Compilation as of April 1 19 3 PEDCO Environmenshytal Inc for US Environmental Proshytection Ag1middotncy EPA3401-78008 (NTIS No I B 288 205) 19i6
S Gray Ne Jersey Department of Enshyvironmental l 1rotection Trenton NJ private com111unication 1979
1 H E Chrislllsen E J Fairchild and R J Lewis Sr middotmiddotSuspected Carcinogens A Suhfile of th1middot NIOSH Regii-try of Toxic Effects of Chemical Substances 2nd Ed National Institute for Occupational Safety and HPalth NIOSH i7-149
4 B 8 Fuller J Hushon 1 Kornreich R Ouellette L Thomas and P Valker Scoring of rganic air pollut1nts for US Environrnental Protection Agency Mitre Corpt ration T1middotchnical Report MTR-6248 I 9i6
S F R Lozan11 Toxicants Selected for Priority En- 1ronmental Assessment corresponde11ce with EPA Region VI Administrat r 19i7
1gt Survey of inyl Chloride Levels in the Vicinity of Keysor-Century Saugus California US Environmental Proshytection AJet middoty National Enforcement lnvestiiatim Center Denn-rand San Francisco E A-n02-77-017a 19i8
0 Bardin l middotstimon~middot in Hearings on Helationshi1 Bet wfen Cancer and the Environmei 1 US llltuse of Hepreshysentatives bull th Congress 2nd Session Committee (In Interstate and Foreicn Commerce ~crial No 94-141 1976 pp 7~i0
middot T H Ma11middoth Carcinogens in the workplace lwrE to start d1middotaning up Sc-imiddotrmmiddot l17 1-1110) lfi8 (197)
I CH ar1111 Classif1r11t1on of the Exshyposure Haza d Prtbullstgtnled Ii~middot Thirt~middot-Six Clwmirals 011 I llixllmmiddots lternorandum from Clllm1middot I Offi(middote of thtgt Assmmiddotiate lgt1rector for 1 middotarcinoltnesis Division of Cancer Cau~ l and Prevention lo SushyJwrmiddoti-or~middot ( 0 Auditor Gtmiddotneral Acshycounting Oft lmiddote19ii1
I0 Chemicals ith Sufficient Evidence of ( arcinogenit ty in Experimental Ani-
A-5
ma1s Jnlernauona1 Agency 1or 1c-curcn on Cancer IAHC Working Group Reshyport Lyon France 1978
11 Threshold Limit Values for Chemical Substances and Physical Agents in the Workroom Environment with Intended Changes for 1977 American Conference of Governmental Industrial Hygienists 1977
12 M H Rogozen D 1 Hausknecht and R A Zisk ind 1976 A Methodology for Ranking Trace Elements in Fossil Fuels According to Their Potential Health Impact by Science Applications Inc for Electric Power Research Institute 1976
1t L Margler R Ziskind M Rogozen and M Axelrod An Inventory of Carcinoshygenic Substances Released into the Amshybient Air of California Vol I Final Reshyport Screening and Identification of Carcinogens of Greatest Concern by Science Applications Inc for California Air Resources Board 1979
14 T H Maugh Estimating potency of carcinogens is an inexact science Science 202 (4363) 38(1978)
15 C C Yao and G C Miller Research Study on ais(Chloromethyl)Ether Forshymation and Detection in Selected Work Environments by the Bendix Corposhyration for National Institute for Occushypational Safety and Health Cincinnati Ohio Contract No 210-75-0056 1979
16 Occupational Cancer Control Unit Calshyifornia Department of Industrial Relashytions Los Angeles California private communications with staff members 19i9
This paper was developed while Dr Margler was a Staff SriPntist in the Energy-Environment Systems Division of Science Applications Inc (SAIi 1801 Avenue of the Stars Suite 1205 Los Angeles CA 90067 and Mr Reynolds was a Project Manager with the California Air Reshysources Board Sacramento CA 95812 Dr ltar~ler is currently emshyployed as DirEgtrtnr of Bnvironmental Scitbulllltcs for WinzlN nnd Kelly Conshysultini EnintNi t1 Third Stretbullt EurekH CA 9gtgt01 and Mr Heynolds is Air Pollution Control Diretbulltor for the Lake County Air Pollution Conshytrol District 55 N Forhes Street Lakeport CA 95-13 Dr Rogozpn is a Staff Scientist and Dr Ziskind is Deputy Manager of the Energy-Enshyvironment Systems Division of SAi
1157
Appendix f~
_l)ispobullition of Miscellaneous Sites
Gould Inc Ver_n)_~ St~condary Lead S~_n_elter
The emissions of arsenic from the four large secondary lead smelt~rs
in Cali forni) J1~1 ~ bull~tiilnted in the program This estimate was based upon a
uniform fugitive emission f1cJr nt WPll supported by measurement inforriation
Therefore source monitoring was r2cq11)nded 1nd Gould Inc Vernon being
~he largest was singled out It was however proposed to monitor both Gouli
and RSR Corp (Quemetco) in the City of Industry since analysis of the plants
revealed significant differences in plant equipment and engineering The
latter being more typical of 1 nodern facility
As a r2sult Jf pr-~-middott -iiscussions and plant inspections we became
aware that Gould was actively in the process 1)f constructing a new facility
which would completely replace the existing one We have monitored progress
on the new site and concluded it would he inappropriate to utilize program
resources to conduct field tests at Gould ~rr1issi 1rns from the new Gould
faci 1ity wi 11 be bas~rl 11p11) test results from RSR
PG ampE - Pittsburg PG ampE - Salinas a~d So Cal Edison - Long Beach Oil Fuel Power Plants
It was app ro p ri ate tJ cons i dP r the (~mission of arsenic from power
plants during the initial study stage inc~ imiddotrace quantities in the fuel oil
are known to be ernitted Because of the population distribution in tile
vicinity of three plants and some 11realistically conservative estimat2s of
erlissiJn c-1ctors tht~ facilities appearc~d among the top seventeen stationary
sources of potential c1Jn-r-n HovJ~ver as a result of a reexamination of
emission data it was concluded that an error ~dj ~een made in the material
balance of arsenic - resulting in tn emission factor equivalent of a 300
release ~io lit~rature was found to justify 1reater than a 30 transfer
function Thus we estimated at the outside th~ emission factor should be
reduced from 013 lb per 1000 lb to approximately 0013 lb and resulting
ars9nic 2missions for the entire state w~r~ cJ11servatively estimated to be
1760 lbs (from 17600) divided amongst 311 the states power plants Clearly
then the four secondary lPad smelters estimat~ of S9410 lbs of arsenic(
B-1
emissions per year makes consideration of poer pl int ernissio1s if ffsenic a
very low priority
In a literdture review by SAI - Methodokgy __f_or Ranking Trace
Elements in Fossil Fuels Accordi~9__t_~_their Potential Health Iripact (Roqozen
1976) - emissions estimates f0r 15 trac~ substances were researched for c11l
and fuel oil pomiddot1er plant c 1J1v---r-sion Source content combustion pr-)~~
transfer functions and contrcl methodology were co1~idered to develop e~ission
factors The output to input ~atio computed and utilized n -~i-i~ study for
arsenic was 002 to 03 (ie tra1sfer function betw~en 2 11d JO~) reflecti~q
the wide variety of fuels processses and controls nationwiJe The upper end
reflecti1g both high arsenic coals and poor e11iss10n contro 1 devices Clearly
neither of those conditions accurtely apply to th2 thr~~ si~~~ r1nrl thus
substantiate the rlecision not to perform emissions neasur-i-~n~s
Calaveras Asbestos Company - Copperopo l middotis Asbestos Mini n(i and Mi 11 i nq
In the past (late sixties and early seventies) ~h2 ~3laveras
Asbestos Company came under heavy criti-is1 1fter imiddot1sp~ction rieasurements
revealed serious problems However significant reduction of e1nissi0ns have
occured prompted by NESHAP regulation and occupcttional standards SAI
inspected the site in f)~c~1b-2r 1979 under an EPA contract An missions
inventory was published under that work (Ztskind 1980) and the bottom line
conclusion is that currently no significant ernissioi1 are gteing released as a
r~sult of hl-isting which would reach the publ le i 1le 0pen pit is some 900
feet deep with blasting at the bottom 0ver 80 percent of emissions in the
CARB Emission Inventory System were attributej to pit blasting In fact at
its current depth blasted 1aterial does not reach the mine surface with tile
explosive implacement used Additionally the site is more remote than might
be realized The ~ituation is vastly different today than in the past and
currently attention should be paid to the issues of occupational exposure and
breakdown of ventilation syste111 controls in t1e inilling operationsmiddot He
recorrrnended that no furtw~ corsideration be Jiven to this site for the
purposes of tl1is w)JJil ie identification )f silt~middot1ificant releases vhich
might he responsible for causi1J middot1 fmiddot middotgtbullgt middot 1= 1(11 1bull11 tinn Pxposure
8-2
Vctri0J5 Refineries
It was est=tblished early in the anrilysis prmiddotJr111 V1at gtenzene
emissions from refinery operations coulrl in th~ 1guregate constitute a
significrnt source Since there ilre a large rrnrnber 0f refineri~s in
California (46 in Los Angeles County itself -it t(~ i 1~ middot)f r-1ltariination) with a
wide variety of types sizes ages etc their eva l uat 11) 1 i =r~sent -3~ i)7
monumental task It was noted immediately t 13t ~~it hi n the total scope of the
study the design and conduct )f c r~f i r1~ y testing program which would develop
a complete benzene emissions in1~ntory for 01e site was impractical let alone
to d1aracteri ze emissions fro111 three ie those listed among the 17 rnost
s i g n if i ca n t potent i a l st a t i on a r y sou r c e em it 1-_ l rs bull
The three refineries were singled out for special attention in the
previol1s phase ~ecause they uriiquely had components which process materials
containing 10 or more per~nt hy middot1-bulli1 11t )-~112~ne (46 FR 1165 Jan 5 1981 pg
1491 ) Est i mates by E P A ( F e d bull r l Rr-~ g i s t ~ r 1q81 ) i n d i cate that 90 p e r cent o r
more of the total benzene fu 1Jitive emissions arise from such components
Therefore attention is appropriately focused on the three henzene production
consumption refineries Chevron (Richmond) Arco (Crson) and Chevron (El
Segundo) SAI staff considered the possibliJ )f -1 testing program at one of
these refineries and cone l url 1middotd it would not 1)e cost-effect i v2 for o number of
reasons
California is a minor producer and consumer of benzene with
approximately 15 of the 114 hill ion pounds produced and
consumed nation1lly in 1977 This can 1)e riY1pared with the fact
that California has approximately 10 of th~ ~ations population
Benzene exposur~ to the generil nJmiddot1lation has been partitioned
among the various sources Aproximately 90 of exposure is
estimated (SRI 1978) to be caused hy gasoline distribution
activities and 1ehiclEmiddot emissions in 1 ir~an areas with nearly all
of the balance )Y benene hrndl ing operations (refineries and
chemical plants) In the case middotlf California this percent1g~ Jill
be even JT1ore di middotpruJor-tionate b~cause of its greater s1are of the
population and ~esser share of the henzene ha~dling Furthermore
since the three refin 1bullry cites ire heavily urbanized no new rural
population cegmmiddotnt is beinq exposed
B-3
Californias most heavily urbanized Air Quality Management
Di st r i cts have a l ready adopted t en zen e f ugit i v e em i s s i on
staridards comparable and vith S( 1ine features VJr stringent than
the proposed national (bullmission lttandarrl (4f-i FR 1165 Jan 5
1981) Thus the concmiddotlusion rleri1ed above (Le refinery emissions
are secondary cause of exposure to the urb~n microopu~ation) is
further reinforced since H is irojected that re1eases from
components in benzene service (gt10 benzene) will be reduced by
73 by the proposed F~deral Standards T~~ ~istrict rules (eg
South Coast Air Quality Management District 456 and 4fi6l) are
not restricted to berz~ne per senor to only components in
benzene service The rjLs 1lso i-lclude flanges in addition to
the components ca i led uut in the propose nat i 01a l standard
The EPA estimated that if the proposed emission standard were adopted the
1~aximum annual benzene concertr0Vioi1 -~01middot 1 plant would be 36 ppb at a
distance of 0 1 kilomeL~~ avay Cor1paring this ground level concentration
with the general urban background i1 California of 19 ppb shows the latt2r t1
dominate In recognition of the secondary im0ortance of these thr~~ i~~s i~
was decided to utilize program resources to d~velop field data at other sites
1111here little emissions informati1)i1 1as cll-1aila1)le
8-4
APPENlJIX C
US tn v 1 ronrnen ta I Protection Agency
middotJc k
METHOD 108 - UETERMINATION OF PA~TICULATE ANlJ GASEOUS ARSENIC EMISSION
FRUM NONFE~RUUS SMELTERS
1 Applicaoil ity and Principle
11 Apmicroli1ability This rnetnod amicropliltS to tt1e determination of
i nor~an i c d rsen i c ( s) e1n is sions trom non f errou Sm(~ I ters dnd other sources as
smicroecified in the reuulations
12 Principle Particulate and gaseous As emissions are withdrawn
isokinetically from the source and collected on a glass mat filter and in
water The collected As is ther1 analyzed by means of atomic aosorption
spectrophotometry
2 Apparatus
21 Sa111pliny Train A schematic of tt1e sampling train is shown in
fiyure 421 it is similar to tt1e Method 5 train of 40 CRF 60 Appendix A
Tne sampliny trdin consists of tt1e followiny components
Note H1i s and a11 subse~uent nferences to ottler methods refer to the Methods in 40 CFflt 60 Appendix I
3 Annual Book of AST11 Stdndarctbull Part 31 Water Atmospheric
Analysis American Society for T~sting and 1-1aterials Philadelphia PA
1974 micromicro 40-4~
C-15
APPENDIX D
~f-~__d_y_ e s f o r Pr o c e s s i ng a n d An a l y s i s of PA H i n
Co~e Oven Emissions from Kaiser Steel
Global Geochemistry Corporation
10 Samples expected
l 1 Connective tubes between section welded to oven port cover and remainder of sampling system - These are to be stainless steel tubing disconnected and capped off for each sample They will be about 10-15 long
1 2 Impingers in series first two water-filled third dry - The center tubes will be removed and wrapped in clean foil and the impingers stoppered as is for shipping One set for each sample Protect from light during shipping
20 Extraction -procedure
Sample fractions are to be covered with foil or in opaque containers with inert gas flush during storage in freezer Solvents used in extraction will be BampJ glass distilled pesticide grade low residue) Working bench area will be screened with amber plastic and light bulbs
2 l Connective tubes will be rinsed with dichloromethane by partial filling capping tilting several times and draining into a beaker This will be repeated until the washings are colorless but at least three times
22 Impinger contents will be transferred to a separatory funnel The impingers and center tubes will be rinsed inlo the funnel using a measured amount of dichloroshymethane in portions The funnel is shaken well and the extract drawn The rinsing and shaking is repeated with two more portions of solvent The extracts and washshyings are combined over Na 2 S0 4 bull
A separate portion of the water used for impingers is extracted in the same way
For each liter of water 100 ml dichloromethane will be used per extraction ie 300 ml total
Internal standard is added to the combined exshytracts at this p0int
D-1
_rmiddot
30
40
Concentration of extracts
Bulk solvent is removed on a rotary evaporator in a water bath at not over 40deg stopp~ng short of dryness The res~dual volume should be just less than the f i n a l v o l ti 11 e o wl1 i ch the s o l uti on i s to be d i l u t e d (accurately) The object is a final solution with roughly five mgml solids concentration in dichloromethane After the vol w1 e ~ s made up to the mar k i n a vol umet r i c flask an aliquot is removed and evaporated on a tared disk in a nitrogen stream and weighed to determine yields The remaining solution is stored in a ial or ampoule sealed with a Teflon-lined septum A split for SAI QA may be made at this point (see Section 43 also)
Gravity column c1eanup chromatogram
4 1 Into a 10 min 1 d column with co2rse sinter and Teflon stopcoc~ 6 g 120 mesh silica gel (activated at 120deg) 1s sluiry packed in pentane cooling the column by evaporation from a tissue wrapped around it and wet with acetone) This minimizes vapor gaps in the column A cap of 3g sodium sulshyfate is added to the top The column must not dry out before the chromatogram is complete
42 A sample of the extract concentrate is taken This is to be based on experience of expected PAH conshytents but initially would be 5-10 mg This is exchanged with pentane by adding 10 ml portions of pentane and evaporating over about 100 mg silica to small volume Three to five portions of pentane are used The final residue is rinsed onto the column with pentane
4 3 The chromatogram is developed with four solvent steps
l 25 ml pentane 2 l 0 ml 2 0 ~~ dichloromethane in pentane
II II ii II3 l 0 ml 50 u II II4 l 0 ml 5 0 ~~ methanol
The column volume is approximately 8 cc This is allowed for while collecting fractions During the first chromatogram the graduated cylinder collector is observed for visible solvent changes and a more precise estimate of the column volume
T h e f o u r f r a c t i o n ~ a r e e v a p o r a t e d i n a n 1 t r o g e n stream at not oveimiddot 40 to l 0 ml then stored under nitrogen in the freezer The principal constituents
0-2
of the fractions will be
l Aliphatics 2 Lower molecular weight PAH
11 113 Higher 11
4 Polar materials
Fractions 2 and 3 are to be analyzed by HPLC the others retained for future interest or in case of an incomplete or defective separation Splits for SAI QA may also be made from these fractions
50 HPLC Analysis
5 l The column is a reverse phase partition column C18 bonded to microparticle silica 4 x 250 mm The solvent system is a gradient from 6040 acetonitrile water to 100 acetonitrile initial slope setting l 5min The flow rate is one mlmin the injecshytion sample volume one to ten microl depending on conshycentration Detectors are UV absorbance and fluorescence in series The absorbance is set routinely at 296 nm but may be varied in certain instances to aid in identification of peaks The fluorescence filtars are narrow pass excitation at 360 nm and cut-off emission above ~10 nm Both d e t e c to rs a re rec o rd e d s i 1i ul ta n e o u s l y
52 Reference calibration is based initially on indishyvidual chromatograms of single compounds and roushytinely interspersed chromatograms of multicomponent reference mixtures The proposed calibration stanshydards are the following at minimum
Phenathrene Pyrene Chrysene Perylene Benzo(a)pyrene Benzo (ghi)pcrylene I n d e n o ( l 2 ~ c d ) p y r e n e Coronene
The internal standards (added in Section 22) are 910-Dimethylanthracene
9-Phenylanthracene 9 10-Diphenylanthracene
60 Peak Identification
Significant peaks not recognizable as present in the reference standard will be investigated by alternate methods for identification
D-3
6 bull I Stopped-flow examination of UV absorbance for maxima
by manual wavelength varmiddotiation allows checking against known reference spectra (published or measured) This is usually sufficient confirm a component already suspected to be present
62 Collection of the fraction contairino the unknown peak allows later determination o~ t~e full absorbance andor the fluorescence spectrum for comparison with reference compounds
63 If not already run on GCMS the sample can be sent to SAI for that purpose possibly allowing the comshyponent to be identified Since the HPLC elution order is not the same as the GC retention sequence it may be necessary in affibiguous cases to collect the HPLC fraction and run it separately by GCMS
The~e may be numerous minor constituents not identified The decision on how many of these to track down and identify will be made in consultation with CARB adv~sors a~d with SAI since at some point the returns ute not worth further expense
_64 IdeAtified peaks are quantified by comparing their areas to the areas of the known concentrations of reference compounds A recovery factor based on the added internal standard nearest in elution order to the sample component is then applied the amount of internal standard originally added to the sample is divided by the amount found in the HPLC analysis This ratio is the multiplier used to correct for losses in concentration and chromatography steps
70 Quality Assurance
71 Transmittal of sample ~plits to SAI provides for external QA This should be done with 15 (or one in seven) of all samples
72 Blank water extracts are carried through the same analytical procedure (10 of total)
73 The cleanup and HPLC analysis are duplicated on 15 (or one in seven) of all extracts
74 The multicomponent reference is run first every day then elution volumes and peak area for each comshyponent are plotted on a control chdrt to pinpoint developing problems At least evetmiddoty tenth chromato-gram will be this standard It wi 11 also be rerun after any service on the instrument~ such as a column change
D-4
APPENDIX E
Gas chromatograph 1mass spcmiddotctromlt~try protocols used by SAI Trace
Organic Chemistry Laborator1
Aliquots of sampl=s received from Global Geochemistry Corporation
were analyzed by combined c11s chromatography mass spectrometry (GCMS)
Liquid extracts were spiked with an internal standard compound (deuteriumshy
labeled phenanthrene) priot to analysis by gas chromatography mass spectroshy
metrymiddot A Finnigan model 4( 21 Quadrupole GCMS system including an Incas data
system was used for all analysis Liquid samples (1 microliter) were injected
into a 30-miter x 025-mm ID SE-54 fused silica open tubular 9as chromatoshy
graphy column (JampW Scientific) A programmed GC oven rate of a 4 minute hold
at 30degc followed by a 40degCmin temerature increase to 160degc and a s0 cmin
temperature increase to 270degc was used for sample analysis The detector
system was a quadrupole mass spectrometer operated in the electron ionization
mode at 70eV The quadrupole mass filter was scanned from 35 to 475 atomic
mass units in 095 seconds A hold time of 005 seconds was used to stabilize
the electronics prior tote next scan Data were continuously acquired using
a Finnigan Incas Data system which writes time intensity mass spectral data
to a magnetic disc The mass spectrometer was operated under computer control
during acquisition Figures 33-7 through 33-9 show the total ionization
chromatograms for each of the three samples that were analyzed
Following analyses the data files were searched for priority
pollutant polynuclear aromatic hydrocarbons (PNAs) In addition to the
quantitative data reduction used for priority pollutant PNAs a survey analysis
was performed on all significant GC peaks (ie signal to noise of greater
than 101) The survey analysis was based upon a computerized library search of
individual background subtracted peak maxima scans for qualifying GC peaks
The library used for qualitative identification AJas a combined NIHEPA
National Bureau of Standards library 1hich contains approximately 26000
entries
Tables 33-11 through 3-13 present the results of this survey analysis As expected a series of al~yl Slbstituted PNAs were observed in addition to heterocyc l i c compounds con ta i ni n~ nitrogen iind sulfur moieties Further
investigation of the mass spectre of the nitrogen substituted identifications
confirmed the presence of benzonitrile isoquinoline andor quinoline acridine
carbazole and alkyl substibted ~yridines Ion specific searches for N-nitroso
substituted amines were neg1tive
E-1
APPENDIX F
Detailed Fiber and Mass Concentration Data - Johns Manville
=-- -i l L - [- - =-- -= f- - =-- -l J middot - -=-
i-- L =-- - ~ ~ L ~ ~ =--shy r---- --
- 7 l -=----7- -- - - - -- =-- - - ~ =-- - 1- -~- l
g~~~~ r_-middot~ ~-i [-- L ~ L Ldeg
middot--~ l - --------=-- r- L r- r-
~~~ =-- -- r- = - -r---
a- - - - - _ - -I~=-- L -- l 7 -~--
gtZlt _
- _(_ -lt lt =- = ZJZ~
-
__ _
~ -- __ middot- -~ ~~~ = L ~f~ Z ~ C
middot -
__ ~ lt lt =- _~~ijE
gt = z lt gt
rbullshy__
F-2
SI I HMllLfc J H 1 I OBr I lll-H ANI) NS~ coHcrrrllliTJ ON CUfllllTI VImiddot SIJMMIIY
TIITIL JnI~ SfiNtlEO TOTtl Ill- or FI ITETI TOTI Illlt ~111-11 11 ltllOT FIIACT I Ori 1111 OF ~middotWI MUllllNJbull lOTI 01IIMImiddot 01- JII TOTI FI lilmiddott~ UgtIJH nD
= t nwo 00 so MI CllONH = t I 1JOO SO (M = II 900 SO CM t bull (0000 l1T~
I I JO SO CM 70000 CUil JC Ml-TEHS
= ~lt O V l lWfl~
Manville D-2 Baghouse
723 347 -633
CIIHYSOTI LE JfllII I nou J11BICUOUS NO PTTFllN NON-SBI-STOS tLL FIIIEIIS
FI 111-11 COllflT ltFIIWll-
ITllClmiddotrir r lllrf
2gt()
) ( I f S bull1 bullbull
I 0
l ~-Irbullmiddot
00
fl JI)(~
00
0 OUU~~
00
0 OflO
~60
1110 000~~
FI IWII ( ()[IClmiddotNTHT I ON ltI- I mn ll11 ~41 GI OF FI LTEH) (FIBlmiddotn~ ll-11 CIJB f-11bullTl-H OF IH)
I (2fi71bull+0G (aG(l-+05
( gt027 F+04 OfmiddotHH
OOOOOE+OO OOOOOEtOO
OOOOOF+00 o uuoul~middotH)O
0 O(OOE+OO o 0000I+oo
J bull f 1)07 E+O( fi r71HE+OG
I
w
TcfLI ~~lllllmiddotCE 1lllmiddot lt~O Ctl lFII ~~O UI OF FIiTEi) t q UI 11-11 Cll II fllmiddotTlbullll OF J I H)
I 590)1+07 6 t 90llbull-HH
( HMOE+O 2 40411bull-1 o
0 OIHHlF+OO o 00001+oo
0 OUO(I 1 00 OOOOOJi+00
() 10110~ Ji)
OOOOOE+OO 1 lHHllJ i 00
0 00001middotgt1 00
M ~~~ middotor-ic Irrrn TI or
( PGRAMS 1bull111 ~c UI OF F J LTEH) (PGRAMS 1111 UJI illI Lil ()j Ill)
IUitTII Ml-Ml ltMI tJIOfl~~) ~Tlgt lgtEV
Ml- H Lot c1middoton nr COlmiddotF I II
I) I MWTEll Ml-N ltMI c11or1~~) STI) tHV
MEtl IOC cImiddotor1 r1N CCWF V11
AIt lbullI rt 1bulli I~ (Sil MIC) STIgt 1gt1-V
Ml- N 101 c1-ori r1u COlbulltmiddot V1ll
I 0 I ~if1+07 bull tJf I (11bull-HH
r 1gta 1 1) bull 2 (10()
0 lt700 l 1Hi4~ 2G44
() ~( 1 027B
- l bull 6 I 1bull1middot 0 J l)J 09J~fgt
1a 1071 1~ 09rn
0 2(jlt)fi l ~HVil 4040fi
~ HltF+O~ U (H~61bull+o I
o rooo 0()()00
-0GJOB 0 (000 0 (000
OOfiOO 00000
- )()57
OOGOO 00000
0 0 1)02 00000
- I J 0 0 0 1W 00000
00000 0 ()000 00000 00000 00000
00000 00000 00000 0 (WHO 00000
0()000 0()()()()
0 ()()()) 00000 00000
00000 00000 0000() 00000 00000
00000 00000 00000 00000 00000
00000 00000 00000 00000 00000
00000 00000 00000 00000 00000
00000 00000 00000 00000 00000
00000 00000 0 (WOO 00000 ()0000
( ~1701 l B bullJl~H 0 lt~( I BIJI ~ ~i Ill
o~-~ 0 ~j
- I ult-fc 0 I Hbull)I o bull)abullgt~
12 (1 41 17bullgt7
0 1 l I H) 4 ltmiddotI 1
VOUJME (ClJB MIC)
MFiN ~Tl) 111-V ru-1N 10C Ct-OM MN COE- Vtll
2lHH7 gt HUB
-2 7m 0 060 1)
( IJl)()C)
0 001 00000
-ltj 74a 1gt 0001~ 00000
00000 0()000 00000 00000 00000
00000 00000 00000 00000 00000
ooono 0 1)000 00000 00000 0 ltWOO
2 7 0 1J lJ I J J
-2J+H () ()~(
B 77gtB
lt ~
lt
C
J c
~ lf
_
ii
(]j (fJ
J I 0
a1 c r--- l)-- Clltr (Y) r- ro C)(V) gt deg CN C1 (0 I ~ C M
N --- r--
w J
s rJ gt-
c
~
LJ
t-
~ - -lt-[-bull
0- ~ l L L l~ -18~~
L a- - ~ I-
-I - ---
-
-- -- -- - - ~l
-1_ - - --
--~--- [-middot-~--- t-- -- [- -_ 3J ~I r-
z g ---
l
t--[-
l l c +-shyc
~ Ci - - I
-1--r-middot~- ~
L l- L ~
L
II V
C
c
t--
-= -- - - ---- - _
middot= - middot-= l J l~ - ~ ~ L - ~ L ~ [--
II II II II II II II
_
---
--
z
- I
==== lt _ --middot --~ 7 _
- ~~- _
z ~ ~ =lt ~~~ti3
f
gt =2 lt_
2 z _ lt=lt i~iS
-- z ~ lt
~~~t3
~lt~ =--~~~sect~ _ -- -middotImiddot_
L -_ -
F-4
(lJ II) I
~ Or--M
ltlJCsr-M ~ O) bullbullbullbull ~ lCMIO bull- ci gt (= N lC I M ~ 0 N
_
J J
---r--co - - ~l r-j = ~i
~~ 2-
88 rJ--_-ltI -
lt ~lt
z
Cl J ~ _ e-l ~~
- ~ middot~ e-- l ~ l~ [- - ~ -~=----~
_ L
L
--o---
-o-~shyc L~ ~ - 0 - Ll-C d o~-
11middot II 11 11 II II II
-_ 0- _
--
-- lt== ~~ ltZ z-== -~~=E~8C~l~-=rI - lt ~ fJ
ltZ lt---=lt===~~~- _ ltltlt=gt ~~=~~~~~ -
~ ~ lt -=z~ z _ -shy middot--z =-
-middotmiddotmiddotmiddot middot-middot -----
r bull bull _
- _ - ~-middot __ _
~ _
middot- - --- middotmiddot
_
- l
~=~r -~= ==
lt =~~~ti
F-5
FIHJ-ll Nil MAS~ CONCFNTllJTlOi SJ I SMllLE I l) Al OJ CUMJIATI JIbull t-llHMHY
TOTL iHf- SCNNFD lOTiL Illmiddot OF FILTFTI TOJI Ill- JSlllmiddot~ll I I OIIOT FlltCT [ ON 1111middot OImiddot ~1-M iIHlllllJNE TO 1 1 I ()lIJMImiddot ()I t I H TOTI V 1111-11~ COUNTED
= 1080000 SQ MI~RlozNS = 11 900 SQ CM = 1 l bull )00 SO CM = 1 00000 liHlS
11 IH) so CM 2G70000 CUBIC METfiHS
1 ) fi FI JllIIS
Man vi l1 e D-2 Baghouse
723 0 347 -633
Cl Ill Y-OJ l LE MIPII I BOLF Jf1BI(UOUS NO PATTEnN NON-ARHFSTOS ALL FI BFllf~
F 11l-ll f()llrlr ( I 1IFH-- l 17 G o 00 00 00 19fi
Tl fL P1 Sl~JNNFD Tl lI ltl- OF 1bullrn~n H 11I HF ~IJlbullI) lll 1 ll(IT 1-ILCTON IL CW ~U-1 HErlllllNE
THl1 (11llrJI~ OF 1IH Tt1T-d 11111middotH COIJrfflI)
= 697~oo so MICRONS l l 1)00 so CM t I JOO t-o CM
= 1 00(00 lHT~ = I l 1)0 ~o GI
ant()Oo cun1r NETEns = 7 0 I l lWllS
Manville (upwind)
CIIIIYsOllY AMPII I nou JMfl[CUOUS NO PA1vrrHN NON-A~~HFSTOS ALL FIBFHS
v1mI1 Cttllffl l 1-- I II I fl~ 1 () () oo l 0 00 40 70
1middot1111unmiddot ToT1L f J BFllS o ooo~ ooow- t OTT~ ( 01rn 57 I 1~m too ooo~
1middot 1111--11 1or111rnnT I ON I Imiddot I 111 II~~ 1111 ~o CM OF f I LTFTl) 1 11111 1middot111 CUil middot11bullnn Illmiddot J l ll)
O OOOlJEmiddotH)O UUUUOL+OO
0 00001-1-00 0 00011 lbull-Imiddot00
G 1 I nn+OG 1 HJ llmiddot~+OG
00001+oo 0001101-1-00
r n~+H+uG l 7llHmiddotlE+Ofl
l bull l 1)1-n+O(i I J ~ 1)(11-middotl()i
T11T I ~Ull FiCImiddot 1 ll f ( ~ l lrl ll11 tn Ii OF F 1ITFll)
f ill I I I i 11 I U l ( i I Ill J
0 OOl)Ol ll() () U()0 I Imiddot()()
0 UOOOlmiddotgt1-00 O OOlgtUlbullgt100
l 0 1) ) I 1 lJ J ~ tl( 1H-HH
0 UUUOlmiddotH)I) o 00001-+oo
~ lt1Hlt1(E+OG 7(H(Ml-middotHM
0 OOOOJbullH)O 0 0000lbullgtHlO
77 I
N ~
rl gt- 10NCJfffll1T I ON
r orR AMS ITll ~q lfI (II- FIT n~ll) - LIi LIIll 111middotTU OI 1lll)(PGRAMS II i1 Ill MlmiddotN
bull I t bull I 1 f I ~ middotr1 Ill lmiddotl 1- [~ IOC ( Hl1I f r~ Ull-F ll
0 0000 Imiddot bull-110 0 0000 Ibull HO
0 ltHgtOO 000110 t) (HJ()()
0()()()()
0()000
llOOOOF+Oo 0 00001bull-100
0 ()000 OOtlOO 00000 0 (WOO 00000
0 (((7
0 ~0-i -0 middotlbullribull)bulll-
0 ltl l lt 0 oHi
00000 00000 00000 0(0HO 0000()
077[0 0 -1117
-0 ~bull)I)()
0 7middot1middot1 (i () L~bull)7
0 7~B( 0 I I
-o alt17n 0 (1)~1 0 17lt1 I
11 I 11ITlt I j i 1l 1 111bull I
f 11middotgt N gt I ii ii I V 111 ii IOC tHlfl flf1 Clll-tmiddotmiddot VAIi
00000 0()000 0000() 00000 O 0000
041000 ()0000 () 0000 o 0(1()0
0000()
0 llfl(i7 OltHJ
-~ lt-11 I 0 07 I 1 () I)lt)
0 (H)ll()
00000 00000 0 ()()()() 0 0(100
0 0 Omiddotlmiddot l
-c07 1)
0 I 1 I( 0 1101
0 l l j--
OOGli -~bullI) ( (J
0 1lt11 I 077G7
I 1 (71 rJI C)
tll N -r11 111-V rmiddotllfl trn CI llf I 111bull1 COIT rll
() ())(l(J
00000 0 ()(1()()
0000() () (1()(10
00000 00000 00000 0 (I()(()
00000
0 ~ I Ifdeggt 0 I Imiddot~ I
- I H 1)7
O I GOO I I i ) l)
0 0001) 00000 00000 00000 00000
0 1 1)17 o~ViH
-1 0 1)gtG 01lbullH 0 7 I till
0 l l lt G 0 ~r1
- J bull J1 1) I 0 17 I bull at-lt
lllllrmiddot lt Clll ii IC)
~I Imiddotmiddot N ~middot1 D 1)1bullI f-1 Imiddotgt f IOC ClltHI rm uw-- i ll
O(lOOO 00000 00000 () ()(J()(J
(l(t(()()
0()()00 00000 00000 (1(1(1(()
00000
000fi7 OOOGO
-G Jll7l 0011G 1 H20
00000 00000 00000 000()() 00000
0 0 I bulllCi 001w
-4 iHd 00101 I ~700
OOIOH 0 0 I Ilt
-(1 j l)~G
000(i 1tHH
--
--
L c~
U c t c
(
J
J -
C
C
c g 0 0
+ + e--c~ =~ - r~ C- l~
--
+ +---
- -C 0 o- -c--1
I
lt t w
lt
C
ta J ~ ltCi
lt ~
a---~ -- c gt C 0 rtl 5 -
VJ C f-rJJ c
[J_
ltI ~ C z
z _ ~
~ lte 0 ~
C
tJ
e
C
~ c C
L N
~~ C 0
C
l~ 1 0
t CI ~ ----~ -
i-
cc cc + + t=~= g~ --
-~ C + + t~ ---[w (~ lC
-
ccc ~t c5 ~ ee 00000
co--oc I
- - - - -_____
------ _ - _ _--__ ---__ ---- _ - _
ooocc
rr ~ c-~ E lt
C
-
~ C g L-
gt o + + ~~ ( L C- lN o -l
t l - -I
~ - L~ [ti e- [
l _ 1
- - ~_ C
gg -- -- ~ -0000-tr--- ____
----C __
- __ - _ --ggggg
- -
C
+ + -- --- -~_shy-----_ _ -----___ ooo~
-----__ ___ _ - - - _ - _ _- - - - ___ 0~000
--__ ---_ __ -- --c5 ooco
C cc
1 II II 11 11 II II
z
-
r ~ c
z
~ t lt i C z __ -- -~tt ~ middot~ ~
-middot middotmiddotmiddot _ - -----=___
_ =~ - _ -=shy ~
-- -~== 2~= ~ () V)
_middot ~ c( c(
0 er ~- ~ ~ lt 0 0___
-
- amp lt~ ~~~t8
F-21
) ) ) )
TOT1L AllE SCiNrmn TIIT1 L ll lbull1 tW FI 1lFll TIIL1 11bulli 1~lllbullll lllIIOT lIIJClION -11- OV TUI Mlbullrllll1Nft TOT1 I VOIIHIIbull 01- I Ill TOTmiddot I Imiddot I I 11-11~~ COi l NTlbulln
_LI__l_ ___l 4 bull P 7 bull P A R l n A 4 R 1 n 4 4 ___ 7 A ___ _RA____ 1 A 1A 8 1 SL__ 1 R 5 1 ~ 3 _1] ________ 111~____JJ__) 1 ~---~-~ 4 9 1 ----1___11
A0 1A4 JR 17 lAR A 41 4C 174 A ll~ JnR ll7 9 5 R12
----- --------- - ---- ---- -- middot-middot --- ---- -- ----middot-- ----middot ----~ 7 c ~ bull n 1 1 2 l 9 -s 1 -i s s ~ c c__ bull A____02 1 1 2 1 r n 1 c 0 r-- l ~ 1 o 1 0 1 t 1 o c- ----~--- -Pl Ol OP lA2 11A 61 c 0 ~ cq03 l 5 4 l 4 4 1 i r 7 -i r) c q 1 deg 1n
qA RO 107 ~_12_7 SA c 7 A04 _l _ s_ _________l 7 l RA____ 7 ______l _o 4 __ l nn_________ 0 _7________ R_l __ o ~ _____ J(__7 11__r ___ _t-__ o ____ ln ___ h_ ______ An____ c-_n __05 ]lA 140 170 7cA __2n1 qc- 7q 1n1
c 2c 7 lR 1 O 406 11 c A 0 4 n 3 4 7 l 4 R ~ 7 7 5
___ ___~ c Z bull A ~R_ ---2_h 2 bull bull 4 e R bull C07 l_ H l_ ri ------- R1 --- c ----------- AQ _middotbull 4_p____ r 7 ___1[_2_ _r--_______ _ A____________ )_3-____ _____ __t 9 ______ ~_____ _3 __ --~___()_______ __I__08 C7 lR9 9A 1n~ ll7 S AS 84 ~Q 33 11 11 1n R 3A 3309 7RJ 1R lA 14-1 17fI 77 4R 7(_
---~- L~ 1 7 -~ 1 7 4 1 l 1 9 S 3 9 C) 4 9 A ~ 9 q10 __ 7 c _ ___ _ 2 RA l A ____J__6_R 1 C 4 7 P C 1 rq I
-----~~a-R___-=4bull2_1 e __ middot-middot middot bulls __ - __ IL 9 1 ~_1_ ___3 7 40__19 1 ~~- gt q6_ (- l Q 144 43 _37 _ A f ---middot ~ bull r bull q ~ 9 IL ~ 1 _0 -~ bull 1 A 3420 l r -- I- I q4 l qo l c c 0
Valve-Reactor Heat Exchanger 200 009 2222 I I I 285 I 93 52
All measurement are for 100 CT streams
1--- w 0
air smicroace Saturation concentratiun at 20degc is amicroproxiriately 12 Measured
values were bY and 74 These values r~sulted from gas chromatographic
dn a I y s e s ot t rd n s t er re c1 sa111 p I e s fr() m t n e 1U U I i t e r Ted l a r b a y t i 11 e i nt e9r a t ed
sample ror tt1e tank car displdcement of 2008 ft 3 the total CT emitted based
on tile ctverctye is LUU8 x 007L = 144 ft 3bull
Tnen for a density of p =PMRT
P = (1)(1gt4) 64 gL = 04lbft 3
(082)(293)
Tl1erefore tne displacement weiyl1t of CT microer ()ff-load is equal to 144 ftJ x U4
l b ft J = ~ 7 bull 6 1 b bull F o r 2 5 U off - l o ad i n gs an n u c1 1I y we have 14 400 l b s bull Th i s
compares with UuPont I s rnectsure1nents of 9 3 vapor content arid 18800 lbyr
emissions Saturation vapor pressure concentration emission would yield
2400Ulo
l)reathing loss from the tank can be calculated from AP-42 as
Lt3(1bday) = 619 X 10-5 M( p ) 068 D 173 H 051 T 05F CKcmicro 147-P
5 0 68 173 = (619 X 10- )(154( 173 ) bull (155)
147-i73
(lj) 051 (26) 05 (1)(1)(1)
= S 63 lbday
3where a working range of zuu x 10 o 400 X 103 lbs were used to determine an
average l i quid nei grit Jn nu a 11 y e111 i s s ions would be 2057 lb Therefore
working losses dorrii nate tt1e totil elissions from tne microlant
Summary
The upper and lovJer bounos of caroori tetracriloride emission ~vere
determined as Ll467 lbyear (upper bound on fugitives DuPont measurement of
storage tank work i ny erni ss ions nur 1a I tanK oreattli ng) and lbgt 15 loyear
(lower bound on fuyitives s1 111eas--1re111ent of storage tank working emissions
no ri n a I tan k breatt1 i ny ) bull
13]
100
SOUfCE TESTS - STAUFFER CHEMICAL COMPANY
101 SITE OVtRVItw
Summary
Stauffer is located in the Carson area of Los Angeles County It is
a manufacturer of polyvinyl chloride from tl1e monomer (VCM) ll11hich is produced
on sit2 Stauffer is the only California producer of ethylene dichloride
(EUC) which is converted to VCM The VCM is regulated by several standards
including Cal OSHA EPA (emission standard) and CARB (ambient)
During Phase I the plant was inspected and possible emission
sources were identified Th~se cans~sted of EDC storage ta11ks fugitive
emissions from valves flanges and pumps process vat2r content gas
incinerator effluents and th~ loading of outbound tankers
Since the completion of Phase I several events have transpired which
a ffectebullj the test program to deterni ne foe i 1i ty emissions These wer2
~ the completion of a vent ~ldS incineration system tied to all significant EDC storage tdnk breathers
$ the completion of a comprehensive study by SAi staff to develop a nationwide material balance for EOC
bull inactivity in EOC importation for the plant and elimination of EDC exportation from the plant
Based 11pon this input and the plant inspection of February 3 1981
it was expected that a plant emission factor for EDC can be determined with
relatively little uncertainty It is further expected that si 1Jnificant
atmospheric release of EOC du~ to plant operation may not occur at the plant
itself but rather offsite These vmuld drise fro11 two sources - the process
water discharge from the plant and severcil off site EDC storaye tanks
Facilities Description
Ethylene dichloride is being produced at Stauffer by two processes
direct chlorination and oxychlorindtion In the former EDC is produced by
d i rec t ch I o r i nd t i on uf et 11y l t~ n e i n the pr 1~senc e of cHl Fe C 1 J crl t d lys t I n th e
latter process EUC is microroduced by the oxychl)rindtion of ethylene with
hydroyen chloride and oxyuen in tne prescrlCe of a cdtdlyst typically Cuc1 bull2
132
lhe microldrrt fluriti1~s tuC after tne microrimar y reactions by separation and digt ti l I dt i ur1 fgtr-ocJuLL 1 s ctured ctncl umiddot 2d ctS foed for VCM production All
sourclS agre~ tl1dt c1ission of EDC is (f concern in its productionstorage
process stages ctnJ O little importance in VCM production steps (J~B 1980)
At the timt of plant inspection some storage of EOC existed at three
ledsed storage tanks located at tne Port of Los lngeles 1-iaterial has not
been 1-Jithdravm from these tanks during the 1dst year and there had been
discussion to consolidate materictl into a single tank
All microrocess components handling EDC storage are now tied to a closed
ventilation incineration syste11 It is expected that virtually all
chlorinated hydrocarbons includiny EDC will be etfectively destroyed since the
system must demonstrdte VCM concentrations are reduced below 1 ppm Firebox
tempera tu re is aoodegF and residence ti me greater than 1 5 seconds under
heaviest flow conditions Note that by way of comparison combustion perforshy
mance datct for PC8s are 99995 destruction at 1832degF and 1 second residence
time and 9Y9Y9YY4 at 2 seconds dnd 2372degF (N Flynn SAi Personal
Communication)
Fugitive emissions from valves and flanges are monitored by the
plant according to the requirements of SCAQMD Rule 4661 Pumps and
compressor follow ~ule 466 Emission and control requirements for vinyl
chloride are specified by ~ules 1005 and 10051 EOC concentrations in
wastewater are monitored several times daily in tne primary EDC steam stripper
stream and once daily in the composite plant outflow stream Daily water
disct1arge limits are 2~ ppm wit11 typical montt1ly averages being in the 8 ppm
vicinity The discharge limit is ~mbodied in the discharge permit (number
5061) vJith tt1e Los Angeles County ~anitation District
Tnere are a number of points at which tt1e rnc can be released to the
dtmospnere and they are similar for both dir~ct chlorination and oxychlorinashy
tion These include the following along with their emissions factors for direct chlorination
A chlorinator vent 2xliJ- 5 IOdSS per unit mass EDC produced
I) I ight end column vent -52x lt)
C dist i I I at i un column vent lXlU-5
u storage tank breatning 7xlu- 5
133
E storaye tankworkiny loss 27x 0-u 5F fugitiv~ emission bxll -
-JLi ildSteJdter l gtX] U
Em~ssion factors are si~~lJ fer oxychlorination processes
Enlission factor vau-5 (~1rbulle Tdilt2~ frcn tne literature (JR13 198U) and
assume 9b iciency for incineration (A-E) It is also assumed above that
EIJC emissions to water are Lf~ uf ~rihsion to citir ard that in wastewater
treatment 100 of EOC discharged tc1 vater is rel easl~d to a i 1middot ~ These emission
factors 1bull1e1middote used to prioritize r-2ieaies iJut were cldrly uude approximations
to tre planL l)ecause of thlt ncine-xt~on system it was anticipated that
fdctors A-E would be ri2dtKcdft 1-middotactor - 1nii~t1t be reduced since a monitoring
program had been in force al20st o~e y2ar Factor G and the off site storage
tanks~ vni~cn are not tied into lt1n ircinerator system rniyht -iominate emissions
Emissmiddotion limits and concomitant regulations embodied in Rule l00S of
the SC-l)MU have necessitctted tlw incinera-cion system Ninety gas chronatograph
probes are located throughout t11e plant including the stack of the primary
incinerator Con cent rations of vc1v1 are reported es sent id I I y be I ow tt1e r~yu la -
tory limit of 10 ppm and in fact below the limit of detection somewhat less
tl1an 01 micropra
lUl MEASUK~MENT APPROACH
Tne objective of tt1e proyram is to determine a pl ant emissions for
EOC It is not accemicrotabl~ to dtmiddotvelop sucll informatmiddot1on basect upon published
indlllstry Jide esti11dtes of plant control efficiencivs Fortunately it was
possible to desiyn a monitoring and calculational progra1n to determine EUC
emissions for tne site and not ctbsorb a disproportionate share of program
resources
Hie re are tnre~ modes ot re I ease from tt12 micro Iant dnd d fourt11
offsite
bull Post-Process 1nci11er-dtion
Processes A-E of ecc ion lll 1 dre a 11 1ented into the pl ant
incinerdtio11 syste11 rt1e conceritrcttion of VCfmiddotl contir1uously iectsured in till=
stctck cts SJdS output remicroresents d rectsondble UjJper li11it to diJply tor EDC
con cent rat i ons s i nce i ts ef f i c i ency of i nc i nera t i on 1s d t 1 east c l1 u a I to tr1 at
134
ot VCibulll Tlerefore knowledge of the system dirflow rnd VCM concentrcttion are
tt1e necessary and sufficient conditions for determining the bounds of EDC
reledse fhe pldnt expn~sse I wi 11 inyne~)s to provide tnese data in order to
support the calculations
bull Fugitive Emisions - Valves Flanges Seals and Other Sources
Fugitive emission from the valves pumps and flanges can be
determined more precisely tnan was anticipated since Stauffer has completed a
comprehensive leak inventory of all such components in compliance with Rule
466 466 1 and 1005 The Inventory delineates all leaks uncovered by their
three man crew throughout tne year and indicates the screening level in ppmv
and component identity In consultation with Stauffer we will be able to
identify the total number of components associated with EDC handling systems
( 700) the distribution ot substance composition streams the distribution
and incidence of leaks by hardware component type and leak rate We will
utilize this information to provide historical data to compliment our Foxboro
OVA sampling at the site dased upon this monitoring we predict an EDC mass
emission rate for the fugitive releases The plant has agreed to provide the
necessary information We propose to meet witt plant personnel prior to the
start of monitoring and finlize the sdmpliny strategy to accomplish 100
coverage of the lines of hiuhest EDC composition The sampling approach and
calculation of mass emission are described in Section 72
bull ~~as tewater
From examination of the emission factors derived from published
literature (see Section 10~) it is clear that EDC release from wastewater to
air is potentially several 11rders of magnitude higher than any other plant
source l1ased upon microl ant ml~a sured concentrcJ ti ons of 8 ppm EDC in water a more
realistic emission factor wgtuld b 24xlo-4 1-iassunit mass EOC produced
Clearly this could still be the dominant source Furthermore the release
point would be expected to middotgte locctted between the plant and the sanitary
district treatment site whi1 his~ kmfrom the plant at 24501 S Figueroa We
believe it is necessary to ndemicroendently confirm the average 24 hour EDC
concentration in the dischar-qe valer by obt1ininy the refrigerated composite
sample We t1ave identified ttie I 1nes of int~rest and received agrt~ement to
sample and analyze for EDC in the stream ~e propose to draw a duplicate
sample from the compositor nd ani1lyze for its rnc content This will be
135
compared with Stauffers para11eI tnalysis It is not reliable to obtain
direct readings of EDC by survey instrument in the air above the water flow since concentrations at crny single point will be i 1 the near dmbient range
Emissions will be cc1lcJLtecl 1_sing the plants vulumetric daily flm
and ass~m~ng that complete degass40~ of the effluent will eventually occur
This assumption is based umicroon the relative volatility uf EDC and tne distances
involved However it should be noted that no direct experimental or
monitoring data is available with which to confirn this Th2 composition of
the discharge stream is unknomiddotwn since it merges into a larqe multisource flow
Conversation with the LA County Sanitation l)istrict (J f1ilne) reveals the
stream to be both exposed and covered Vents exist where me measurements
could De made to assess gross leaka~e at key points
We will obtain samp1es of several plant process discharge streams in
40 ml bottles witn no head space Transit time fr0m sample collection to
analysis will be minimized and wi l I not exceed 24 nours This time frame is
conservat i v e al though ED C i s a v o l ct ti l e mater i al ar1 ct vJi 11 undergo
concentration degradation and outgdssing lnalysi) will b(~ performed in tt1e
SAI Trace Environmental Chemistry Laborcttory utilizing pur 1Je and trap analysis
FID gas chromatography A trial a11alysmiddotis was conducted and the EOC
characteristic peak was distinctive down to the ppb level Therefore the
analysis should easily confirm concentrations in t1e 8 ppm range
Offsite Storage ranks
Three EOC leased storage tanks are located offsite at the Port
of Los Angeles and are owned by annther firm Annual emissions from the tanks
were very roughly estimated based un AP-42 JRB 1980) as 44 kkgyear
based on preliminary estimates of stored quantites
Considering tne magnitude of this source these storage tanks must
be investigated We will gather information about their configuration and
control i n order to ca I cu l ate their e111 i s s i on s bull
136
lU3 UEfERM I NAmiddot 10~~ OF MISS IONS
1031 Incinerator
Vinyl chl 11ride co11centration is monitored dt approximately 90
locations throughoul the pl trlt (Lanyner 19Bl) including the incinerator
output Conc~ntrations are reported by Stauffer dS less than 0l ppm at a
flow rate between 10000 ant 12000 SCFM Although no direct monitoring of
EDC is conducted it is posible to conservatively bound the concentration of
EDC at U 1 pp111 Tl1at conce 1ltrat ion is a suitable ci10i ce s i nee it represents a
conservative bound on tile VCM levels measured near the stack output
Furthermore the molar volume will be taken as 224 L rather than the higher
value it would nave because of the sl i~thtly elevated temperature at the
detector location
Using 12000 SCFM one has thE annual emission of EDC as 12 x 103
SCFM x 283 LSCF x 526x105 minyr x lmole x 97gmole x 10-7vv 224L
Therefore incinerator emissions of EOG are thought to be bounded by 773 kg or
170 1byear
10J2 Fugitive Emissions
Seven hundred (7UlJ) sources were surveyed comprising nearly 100 of
EDC service All accessibl~ plant areas with streams containing greater than
1 EUC were screened except for a small number of devices located in areas
11lere active maintenance was being condJCted It is believed unnecessary to
p2rf-gtrm any emission factor adjustment since it is estimated that gredter than
95 of the requisite components were screened
Three leaks above an arbitrary OVA reading threshold of 20 ppm were
parameters and mass emission numbers fhe upper bound leak rate was
determined to be close to twice the nominal leak rate which accounted for the
response fltlctor of amicroproximamiddotely 05 by Radirn for TLV detection of EDC
( 13 rown 1980) bull
Stauffer found and reported approximately 10 leaks in the EDC
service during 9 months previous to the pLrnt testing Assuming no111inal leak
137
) )
Table 103-1
STAUFFER CHEMICAL MASS EMISSIONS PREDICTION FROM OVA SCREENING VALUES
OVA SAI Actual Upper Device Response Response Cone Source Leak Rate Bound Location (ppm) Rae tor (ppm) a b Se I Re Function lbyr lbyr
Gate Valve Unit P1559A
Valve Heavy Ends Column Unit 14439
Compressor Seal EDC Storage Unit Tl062
I-- w co
300 11 273 -035 096 0 17 153 D 6
11 I500 114 439 II 241 C 1
r III II II o-is lUI bull2000 vl 1 1818 A 9~
All emissions 100 EDC (12-Dichloroethane)
11
values in the nei~hborhood of 2000 lbs~ total could be emitted annually
Therefore it appears ttiat major reductions in the mass emissions were achieved
by the company run inspection prugram and the fugitive emission source has now
become of seconqary importance as a fraction of total plant emissions
lUJJ Wastewater Discharge
Wastewater samples were collected in duplicate from four sites
within the plant for determination of ethylene dichloride (EDC) concentration
The samples were collected in EPP standard 40 ml VOA vials on July 1 1981
Analyses were performed using standard purge and trap techniques coupled with flame ionization detection gas chromatography The results are given in the
table below
Sample description EDC conc~ntration range (microJml)
EDC concentration average ( 1-19ml)
Approximate fl ow ( gpm)
PVC Interceptor Box 82 - 108 95 200
EDC Stripper number Chlorination area C1404
2
011 - 014 012 25
Final collection site for pH adjustmentP663 349 - 3B2 366 350
Final discharge site sanitary sewer 6 - 254 158 500
The water in the PVC lnterc1~ptor ~ox was warm and represents washings from the
PVC reactor which travels t 1J the Interceptor Box in a concrete drainage ditch
Water from the EDC ctripper was vPry t10t and because of its heat was difficult to collect The ~1eat ot trie wat11 r may in pdrt account for tne
relatively low concentratio11 of ElC here as the EDC would out-gas from the hot
~vater more readily The fir1ctl collection site tor pH adjustment is a large
concrete container middotvith mixers in it located just prior to the final discharge
site Water at the final dischdrg~ site is being constantly aerated due to
the speed that it flows thru~yn the concrete drainage trough This aeration
could account for tne relatively large range in the rnc conotent found here
The average EDC concentraticmiddot1 at tne final discharge site is middotbelow the daily
discharye requirement of 25 19ml EDC
139
Plant oersonnel indicdted monthly iverage readings are typically on
the order of 8 ppm but daily averages can re 1ch greater than 20 microgml In
addition LA County Sanitation Uistrict staff (J Milne 1981) indicated that
an on-site impoundment pond can become laden with EOC during certain abnormal
operating periods and discharge Vdriances arl requested by Stauffer This
mi 9t1t Jccur on the order of O1c0 aCii ecH c1 1d t1erefore is no-c expected to
signifcantly impact averag~ cnnJJI discharg ~ values lt s not expected at
this t1ime tEit evaporative eriss uns from this pond are significant except
infrequently during upset conditions and spi 11 control operations Program
staff ~~ 1er2 una111Jare of any periods ~h 1~n EDC cimtent in the ponds could be
appreciable and therefore no sampling was performed
Utilizing the average uf the two final discharge concentration
1readinJs (158 micrograrnsm~ 1 and 1G 1Ji gpm fl)W one has 34600 lbyear of EDC
released into the sanitary sEwer For the pJrposes of calculating population
exposures from plant releases it will be assJmed that the EDC is locally
emitted from the wastewater streams Tl1ere ire no monitoring data available
with which to develop a more accurate release profile
However emissions of me fron the plant wastewater discharge can be
calculated according to the method of Mcmiddotckay (197S) as modified by Dilling
(1977) It should be noted that this mLst be considered an estimate since
conditions of fl ow and the presence of other substances will influence
emission rates
Using Uilling (1977) for non~erated flow first recalculate
Henry 1 s law constant (dimensionless-my of chmiddot~mical per liter of air divided by
mg of chemical per liter of water) dS
1604 P M 1 Wl
TS l
)
wnere
H Henrys law constant dimensionlessl
Pi tile compounds pur1~ componen- vapor pressure in mm Hg at T
Mwi the 11olecular weiglt
T tile absol ~te temperature of he wastewater in K
Si tt1e compounds ~uluhility in myliter at T
Then for UJC
(l6u4)(7~)(9Y)H = = U0447 with 1
(2lJ4)(8690)
LIii bullul1Jl)i I 1 1y 1 middot UL i11 wt1Ler yiven -1L L0degC is 8u4JU UHI tile vapl)r pressure
1 4 psi ( 7 2mm) at 7 0 degF bull
Tl1e overall liquid mass-transfer coefficient Kil is given by Dilling (1975)
as
(2211)(06) in mhr
-f-0-4~ (M _) 12+ 100 Wl-H-
1
=U108111tH
Then from Mackay the percent desorption is given by
c 1 = exp (-Kil tL) where
c 0
c = the concentrd ti on at time t of EOC 1
co = tile initial concentration
L = the Ii quid depth (Ill)
t = tt1e retention time (in hr) of the liquid in the wastewater
system
Retention time is bc1s~d 11pon a flow velocity range of 3 ftsec as estimated by
the Los Angeles County Sdnitatii)middot1 Dibulltrict (1J Milne) over a middotdistance of
ctmicroprox i 1nate I y S km to the Joi n_t Wdter Po 11 ut ion Contra l Pl ant Sewer fl ow
depth throughout the entire route have been estimated by J Milne as typically
1 foot 030m) to the Davison Pump Plant and 3 feet (09m) to the Joint
Tredtment Plant distances assumed to bt~ 1 mile and 2 miles respectively
Trdnsit tiine l)eco111es between 05 tnd LU hours sequentially Then computing
the net emission reduction as the product of each leg
Therefore 26 of the EOC i erni t ted bdween the pl ant and the Joint Water
Pollution Control Plant Rtgtsfdence ti111e and conditions at the plant account
141
for additional releases and residudl content is forther emitted between the
plant cnd the ocean discharge poiiit ~-Herever the flow is r2tained aerated
or shallow emissions will be accentuated
Offsite Storage Tank Emissions
There are three storage tanks leased by Stauffer for EDC storage
1ocated at 22nd and Gaffey Sts ~ bull San Pedro These tan ks a ~e used for long
term storage rather than providing feed on a routine basis Therefore
breathing loss rather than v1orkir 1J lass is )f concern An tanks are white
two being 67 ft in diameter while t~2 third is 57 ft All are 40 ft 3 inches
high and have capacities of e~thi l0~10000 (2) or 840000 gallons The
tanks are cone roof type and are not tied into vapor recovery systems The
South Coast Air Quality r1dna~J~ment u~ str4 ct has based their estimate of
emissions on the specification of 12 foot vapor level Using the AP-42
formula for breathing loss with the vapor pressure of EDC at 20degc and an
average diurnal temperature variation of 26degF one hds for each of the two
larger tanks
l) 173 H bull51 T bull 5 LlI = (619 x 10-S) M( _P_ tH F CKcp 147-P
68 173 51 5 = (619 X 10-S) (9119~ 116 bull (67) ( I 5) (26) (1) (1)
14 7-1 16)
Thus for the two larger tanks LB= 472 lbday
For the third tank 0=57 and = 178 lbdayL8 The total emissions become 23724 lbyear ~ince he plant is in the process
of shutdown it is not clear what the liquid levels currently are Note that
if the material in the three tanks were combined into one totamiddot1 -~1nissions
would be reduced markedly Exposure to a population from this site was
determined separately since it is lbullgtcated over six miles fron tl1e plant
142
110
ASSESSMENT OF PUPULATlUN EXPUSUR~
111 OVEKVIEW
A major program goal wa~ to compare emissions from the various
sources and identify and rank any 11 tlot spots in California where the general
population was exposed to elevated concentrations of carcinogens A simple
Gaussian dispersion model was therefore used to obtain order-of-magnitude estimates of exposure of the genermiddotal population surrounding each source
Since this was essentially a screening study use of more sophisticated models
was not appropriate
It cannot be emphcsized too stronyly tnat most California residents
are exposed to emissions of hazardous substances from a variety of natural and
rnanmade sources Urban dwellers dre typically exposed to greater concentrashy
tions than rural residents however all are subjected to so called 11 background 11 levels from multiple sources In order to place stationary
source exposures in perspective the typical ambient levels of each substance
were identified from the literature and compdred with the concentrations due
to the emissions from each ~lant Exposures were thus expressed both as
absolute quantities and as increments above background
Comparison of plants presents a further difficulty in that various
substances are being consict~red No attempt was made to evaluate the relative
importance of exposure to two different substances such as chloroform versus carbon tetrachloride other than by ambient concentration
112 DATA SOURCES
1121 Meteorologicdl Llata
The dispersion model to be descritgted in Section 113 required input
of annual average wind speed and frequency of occurrence of wind from each
compass direction These data were obtained for most of the sites from the
South Codst Air wuality Management District (SCAQMD) In other cases (Kaiser
Jotms-Manvi l le l)ow and OuPlt)nt) only clai ly cJVerage wind speed and frequency
data were available In al I cases we used data from the meteorological
station nearest the modeled e11issrnn source Table 112-1 summarizes the
141
Table 112-1 METEOROLOGICAL DATA BASE
Observation Site Data Period of Pldnt (Distdnce Source Observation Remarks
from Plant)
A11 i ed Che11 i Ca 1 Lennox SCA~MD 1Y71-1974 Averaged hourly readings available ~ 1 Segundo (3 111iles)
S ff~r-- Uemical Long l3each SCA()MD 1962-1974 Averaged hourmiddoty leadir19s avai1able offsite Carson (3 miles) storage tanks appruxi1ntL~ly 5 miies from
Figure 114-3 Cumulative Population Exposure to Carcinogenic Polycyclic Aromatic Hydrocarbons From Coke Ovens at Kaiser Steel Corporation Steel Mill Fontana
CV)
0 -I X-V) V) a _J
s 0
C 0 micro co s micro C a u
l ~
14
1 --
J- t_bulli
_ C
uw m 0
1 -0 a bullbull micro
microdeg ()
0 micro tbullmiddot C QJ V)
0
X LLJ
C
middot4middot~middot C 0 micro rt1
-- - 0 ~ 0
0
n __ _______bull~ -______ --al_ - ~_
I bull
I I - bull
middot 1
-~1
l i l ~l
~ 0 I 1
Li bull I t_~I
i bull -1 _1-J
0fmiddotbullmiddot-middot+-middott-t-r+-t-+-+middotr+-t--+-1--t+ t f-t-J-t--r-+tmiddotmiddot~~bull+-t--1~~bull-+-1 Id middot t 4 E1 (
I ncrementa 1 Concentration (~gm3)
Figure 11 4-4 Cumulative Population Exposure to Asbestos From Johns-Manville Plant Stockton
I
)
--~ ~w t _ ---~- ----------~ --____~-=~ M
~
bull I bull I I I I ii i Q t I i E e- I Ii tl I I I 6 I a I I J I f t t bull I G fl I
a I e 1 1 a 1 1 1 1 iii bull bull 1 1 bull 1 1 bull 1 1 bull bull bull bull 1 1i bull bull 1 1 bull a bull 1 1 r ~ middot u ~ ~ I0
+H~ ~ ~~~~-~~~~-rmiddot-~~~~~~~~~-~--~~-~~-~t~~~~~-~J C
~1_ c bull 1 t~ bullbull bullabmiddot 11 l bullmiddot II bullbull C- I u~
0 rl
gtlt I I I- 20 () ()
CJ -1 ~_J
J Cbull s
I If I I I I0
C bull I Ii I I I I I I I I0 1E
bull-+l I t I I ro s Lower+l
I I I G I iii I bull
() 14 I I IC
u I0 I I II Iu
C
1-_~I fi
CJ ~ I-- (j) CJ I G I I p +l
ro +-l I I I1~3~V)
0 +) UDf)Pr u I I 11 I I
CJ 1iII
Vi I 0 I I It- I a X
0 ~
w I I I I bull
Lower C 0
middotr-
- 4+)
(1j
J Q_
0 bull1
ij a 3- ti I
D d I I I I
I I I bull
II B I t
i ] i amp I
1 ei ~ amp 3 ii
I I I I
e ft a a I bull G
I I I bull I
I I bull I I t If
UpPrf
I I
I I I-_L~ lI I I f
I I I I I
- I I
j I I II I -
lI ImiddotbullII
l I
l Li
Incremental Concentration gm3)
Figure 114-5 Cumulative Population Exposure to Perchloroethylene and Carbon Tetrachloride From Dow Pittsburg (Solid Lines) and to Carbon Tetrachloride From DuPont Antioch (Xs) Results For Upper Lower Bounds on Emission Estimates Are Shown
1i-1 - middotbullth~middot t =lttF--i-+-middotr+-r+middot~middoti-middot-~-4-4middoti t -~--is -~middot-rmiddotibull--+ u ~~ i--t4~ t-bull~~311 li-i1middoti- C - JJ _ CL
-- I 1 t middot1 c7 Ibull ) ~ I-middot bulli C bull 1-_ 11 11gt1l11 ~bull -a bull-bullbull t a- II bullbullbull bullI lt_I Iii _- I
Incremehtal Concentration (microgm 3)
Figure 114-7 Cumulative Population Exposure to Carbon Tetrachloride From Allied Chemical El Segundo (Curves Corresond to Upper and Lower Bounds on Emission Rate Estimate)
--- I
~__~-unt2wbull--o1WCUatJ41ia1
16 0amp (V)
0 ~
gtlt
14~11l 1l QJ _J
~
1 middot-0
c t1middotmiddotmiddot 0
micro ro
_ c micro ~ 1euro1pound1
0- QJ u-J c 0 u O =-middot ~-= QJ
c_1 micro ro +- V)
0 micro ~ t1 O QJ 1l 0 a 40gtlt w c 0
-
ro --
micro
20middot l ---a I ~ I a 0
0
pound1r-t k
Figure 114-8
8tWI-S~1111111N~-
a I __ I wi~ i awLi--~~ 2 I I
1--+ I ~--if bull =~~--bullJ-+--f-1-----middot+--+middot-middot--r9c- 1 c- J
bull _ bull -~
Incremental Concentration (microgm 3)
Cumulative Population Exposure to Carbon Tetrachloride and Chlorofonn From Allied Chemical El Segundo Assumingsix-months Operation With Each Feedstock (Curves Correspond to Upper and Lower Bounds on Emission Rate Estimate)
i u---J--bull-+ i i--+ --4--+-imf-bullbullQ~-+-+-+~bullbull-4 171 i R = Fi middot1 1bull1 r1__ bullltaa ~a
Incre0ental Concentration (microgm 3)
Figure 114-9 Cumualtive Population Exposure to Carbon Tetrachloride From Allied Chemical El Segundo During Offloading From Midnight to 6 am middot-middotmiddot------middot--middot
--bull ft ampLA 1111-~lliaIN_WPl1A 1JIU_Mt91 MJ1WWlaquoPI-W1111~ --------__----Et1
M gt~middotI
0 -I
-X
~ t~ bullM
V) middot-middot ~ V)
QJ _J
f 0
0 -
C -4 0middot~middotmicro re f micro C a u
10 Cdeg u 0 3 0middotmiddotmiddotbull~ -0 a
+-gt ro
+J V)
+-gt 0
2E1-0 QJ V)
0 a X
LLJ
C 0 1pound1--middot+-gt ro
-- Q_ 0 0
I I
~ -~middot bull I -bull-_ bull I
middot lbullgtbull___I I
frac14 llk I
I bull
-~
I
-1_ I
Ibull II__ JI
I~
(1 f---1-+ bull-r_--f bull middot~-bull +-- bull i -bull 1 -1-+--+bull-+-t-+bull--t1 ~ _ 4middot 6 E 1t1 1middot2
1 3 5 I middot~ _ 11
Incremental Concentration (ngm3)
Figure 11 4-10 Cumu1ative Population Exposure to Ethylene Dichloride From Stauffer Plant Carson
_~ ~ t M
C X E k1~
-
- c- 1-1 ~ C middot-middot -_J
I--- -J 4t1middotbullmiddotmiddota
D
-+-shy --middot liiit1J
r ~
ltL
t-1 _ V
0 -+-1
middotr -bull ~1
X
C
+- bull1 ~1 atr l -=
~~lll(k~IHmiddot~~-Wa~middot~nL1middot~~~ta KF u n P middot~dE-tveRS ~-tl~~~-u-~~~tmiddot_-~~~~liil-~-~~~ tc r
I bull
bull -r~l-yen--~1l~a l fl __ A4o bullbull bullbullbull ~ 111bull--middotbull
8 NJ industrial sites 28 samples in industrial areas with benzene consumption facilities Torrance CA Tuscaloosa AL National Forest 1000 sarip Ies-Torontt) Azusa CA El Monte CA Long Beach CA Upland CA Los Angeles CA Riverside CA
bull Assumes all year operation on either feedstock If plant operates 50 on each feed the population exposed to greater than 50 increment over background goes to 16025 - 19377 and 4044 - 16025 for 100
Ambient concentrations of benzene vdry over one order of magnitude in the literature and therefore make tnis calculation questionable For Kaiser therefore the carcinogenic PAH assessment was used to evaluate incremental population exposure above background
+ The partition of emissions from Oow are 78 carbon tetrachloride and 22 perchloroethylene by weight
173
1142 Comparison of Incremental and l3ackground Concentratiors
Those sites whicn elevate background concentrations greater than SO~~
City Kentucky 6600 The latter two are associated wi t11 EOC pl ants Data
taken in senFice s~ations ano lrctffic areas ir various cities range from -~
300-3640 ngm~ As previous~J mentioned the Stauffer plant is discontinuing
operations These two sources should be examined as part of any possible
start--up permitting activity
11422 Kaiser Steel Corporation
The Kaiser steel plant polycyclic aromatic hydrocarbon (PAH)
emissions cffmiddotise from the coke oven omicroerations Comfdrison of the cumulative
population exposure Figure 114-3 and Table 114-4 with the specified
background levels for urban areas illustrates the breadth of the exposure
distribution The five known PAH carcinogens that were isolated in the coke
oven emissions were quantified in tile ambient air of Los Angeles by Gordon
1Y76
Benzo[a~pyrene 046 ngm J
t3enzo~e]pyrene osio Benzaanthracene U lo
Chrysene U60
lndeno Cl23-cd~pyrene 134
Althouyn these concentrations ctr~ low co1npared with 1 number of other cities
cited in the literature and repres2nt a very I imited data base tt1e predicted
concentrations from the Kaiser plant ~enerally exceed these levels by d
174
significant margin ie greater t11an 30000 are predict2d to be exposed to
4redter than l()-lt t 1tbull 1111tlit~nt background of 3j nqm 3 It ic 1 iklly tlat
lHrllerte exmicroo~urt 11 Ilie drld dre also 11evatPd ovr r11nbient I1owever since
backyround concer1tr1tions of t)enzlmiddotne show large variation no population
calculation was specified
11 4 bull 2 bull 3 Dmv
Carl1on tetrdcnloride releases from Dow constitute approximately 78
of tl1e total CT plus perc emissions in Table 114-6 and were found to elevate
urban background concentrations greater than 50 in five census tracts Emissions are predominately from storage and check tank working and brething remiddot1 eases
1142~4 Allied Chemical
The Al lied plant was modeled several ways since the plant can
operrttt~ wiUl chlorotor111 dr cc1rbon tetrctlt hloridt fo(~d The c1bull-i pr~middot11ntlid in
ldlgtlt~o ll4- drld llil-B represent dflrlUil 01Hrdtion with eitt1er teed Partial
yedr operation with each feed can be scaled from the individual annual modes
and is presented for equal hltllf year operation in Table 114-9 As with the
Du Pont plant carbon tetrachloride emissions arise from feed tank working
loss fable 114-10 illustrates peak exposures predicted to arise during an
off loading cycle As expected concentrations in that six hour period far
~xceed annual average values Insufficient data were available to contrast
levels with backyround transient concentrations
175
l~O
AUHLAlHJ~ CONTROL n CHNil)UES
Alternative control approacnes for the most significant emission
sources dinmig ~he various pl1ants n ees fitie(J n the follo11 ric suJsections
EmpiiasVi nas Deen placed in dedlinq ~riU1 Uiose sources of 91reatest absolute
inaynHud2 (lbyr) and those const1 ut 0ing trie largest increm121middot1t middot] bulli~he
bdc kgrnund con cent ration of Hie em~ tted subs tcince i~o ctt tention Wi 11 be ~i ven
to the s~iondc1tY sources witt1middotn cK1 microant or to trlL case of J011ns-Manvi I le
since -UH imiddotajor source is less tran LOO 1byr and is not microrejicted to raise
bctckgro 1wd exposure levels to the ~~ntr21 population Furt12rmore a11
emission sources dt Kaiser Ste 1 1iC(1 v1 1~n directly dedlt -1ith in this
program r1re elltlteci ro t11e cc1lt1bull ovcmiddotr nperatmiddotions These faci 1ities are to be
closed du1rm and all primary reel 11i~ oH~rdtions discontinmiddot 2d Kaiser
forecasters have predicted further deterioration of the plant economics and
the phased closure has been accelerdted for primary steel making operations
Note that this closure is essentially irreversible cince differential
exmicroansion and contraction of the coke oven structur1 occurs in the cooliny
process and it would be improbable that ovens could be reheated without
extensive reuuildiny at major expense
12l STORAGE TANK EMISSIONS - fAUFFrn IHJW OUPUNf ANU ALLIED
At c1ll four sites emissions from storage tanks constitute either the
primary or near dominant (Stauffer) source of carcinogen release andor
genero1i iJOpulation exposure Curreritly the tanks of interest at each site are
permitted by the local Air Quality IJistricts hJwever they do not require
emission ontrol systems for vdrious reasons Tl1e estimated releases grounds
for exemption and other pertinent information are given in Table 121-1
In order to amicromicroreciate tth~ practicdl dlternatives for emission
controls the provisions of SCAWMU Rile 463 ar~ listed below which specify the
acceptable alternatives for tdnks r4uirin9 controls ie tanks 11aving
capacities greater than 39630 yal with suostances of true vapor pressure
176
Table 121-1
Site
Stduffer Uff-Site Tanks (J) (Sdn Pedro)
Allied (El Seyunrlo) - KdJ
1bulllctLer1 d I reld iturctjc
-J -J
Dow (Pittsbury)-Product Check Tanks (4)
Uow - microroduct sturag~
DuPont (Antioch) - Raw Materidl Feed Stordge
Approx Capacity~
6lUS x 1g (c) 84 X 10 (1)
lt3Y630
18000 each
70000L) 3UUOUO
30000 working 42636 liquid full
STORAGE TANK
Substance
Ethylene Uichloride
Carbon Tetrctci1 I Jr i J1
Perchloreshythylene and Carbon Tetrashychloride
CT perc
Carbon Tetracr1 l ori de
SUMMARIES
Emission Mode
Tank breattli ng
Tank v10rk i ng Tank breatttiny
PERC-~JOrk i ng PERC-b r2a t ~ i ng CT-working CT-breattling
Tank breathing Tank breatt1i ng
Tank workin9 Tank breathing
Vamicroor Pressure
014 at 70 F psi
1 63 psi at 68 F
16d psi at 68 F(CT) UJ9 psi at 68 F(perc)
168 i)Si dt 68 F
Grounds for Exemption
Exempt from control re4uirements to SCAQNJ Rule 463 since vapor microressur-2 dues nut iXC~ec
l~ io at sturdJ~ conditions
txe11pt TlJil fule -+b3 since tank capaL ~J ~ s u~der 39630 gal (lSO 11 )
Exempt from Regulation Rule v~~vu u~~~~~shytank capacity is under 39630 ga 1 andor substances are not classified as organic 1i4uicts
Exempt f ro1n Ru l ~ 85300 because subsLctnce is not classified as dn organic li4uid accordin to Regulation-Rule 8120
exceeding l~ microsi at storage conditions
~ floating roof tanks
~ fixed roof tanks with an internal floating type cover
bull a vapor recovery system with vapor collection and retllrn (or disposal) proce5sinJ lXctbullerlin~ 95 efficiency
ln a cllemicai plant a typical vapor recovery systein (KR Evans
SCAQMU) migm consist of a collection inanifold to d recovery system (suctl as a
vapor sphere) to a compression syst2m dnd subsequently to absorption and
recovery systems The absorpt~on st~n c~nsisting possibly of (rubber
stripper or activated carbon
In dealing with speciti( carcinogenic substances such as vinyl
chloride hiyhly efficient vapor control system perf0rmance has sometimes been
stipulated necessitating incineration systems
Fer th~ cases of corccrn realistic alternltltives are as follows
Stauffer Off-Site TanKs - Tnere are a number of options which can 1~arkedly 3
reduce ei1issicns from these tanks from th~ calculaL~d value of over 20 x 10
lbyr One alternative is to consolidat~ the material in tne three tanks
One of the large tdnks can contain the currently stured material and reduce 3emissions to apmicroroximately 13 0 x 10 lbyr Anotl1er alternative is to
transfer dll E0C to a single floating root tank Various types of such tanks
exist dlld are reviewed in the EPA report Lrganic Cht~mical Manufacturing Volume
J Storage Fugitive dnd Secondary Sources EPA-450J-80-025 eg internal and
external floating roofs and a variety of seal desiyn configurcttions Generalshy
ly such designs would be expected to reduce emissions to the order of
one-fourth to one-fifth the current level Cost of 1nstalliny ct contact
single seal internal floating roof wds estimated by BB Lumquist Pentrex
for EPA as $27770 in 1Y8U dollars for 7U ft diameter tdnk Cost to build an
external floating roof tank was estimated by G Stilt of Pittsburg Oes Moines
for EPA as ~14UUUO for D=67 H=4Uft Anottler com111on approach is to utilize
cdroon ctdsorption This works we11 witn nonpolar hydrocarbons dS VvC is
removed from the vapor phase A basic syte111 consists ot tvo carbon beds and
a regeneration system Regeneration is typically performed with steam or
vacuum Figure 121-1 illustrattS rhe systems (Basd2kis 1~110) Steam raises
tne VOC vctpor microressure The resulting sti~am-VOC mixture is condensed and
173
AC-TJA~D CA-e0-l
-----~---
LOW- ffiESiJ~C ~T~H----
PiltOCESS COOL-J~ At-JO oLOWER DRY t-J~ 2gtLOWE~
COgtJDENSER
J
COOLIN(a - --------
WATi=-A
RECOVeRE SOLVENT
WASTE WATER
Fi l 1 1 middot gure middot - Activated-c arbon Adsorption System
179
rout2d to c ~epordtor decanted and returned to storage In vacuum regenerashy
tion VOC vapor is desorbed by pu1 ling a vact1um on the carbon bed then condens-
ed a1d returned System efficiency ~ ~ est ii1a ted at 96 f(duc middot i m from fixed
roof levels (EPA 4503-80-025)
H~fri 9ctated vent concicr1sers dre une of the most cu1mon emission
reduction prJcesses -for conttol ~ ng hx2-d-- storage tai vUC Figure
12 1-- 2 i l 1ust rates a u n i t ( Er i scmiddot n 19HU ) eff i c i en cy of middot middot~ ~ c middotlt middotmiddot _y i s rate ct
bet1Jee~middoti 60-90 for the vapor pressure range of concern For such a large tank
the capital costs would be high ~igure 121-3 illustrat~s EPA estimates
(EPA l ~~SOb) for the condenser section Condenser sys tern a ea mu l d be in
excess of 1000 ftL
It should be noted t -at Jtc1 uf fer Uiemi cc1 l has arH1ounced the c 1 osu re
of its VCPVC plart PrevfoJsly tt12) had planned to purge tlle off-site tanks
of EUC Since any future po~sible micro1ant stdrt-umicro will necessitate a compreshy
hensive SCAQMD review this ltoument can assist in evaluating proposed control
measures
Allied and DuPont Feed Tanks - In bott1 of these cases the nore significant
quantity of emissions arise from tank workinltJ ie during the off-loading
activity rather than tank breathing Contnl ~easures taken for working
emissions are thus of primary concern Ther~fore no detailed discussion will
be provided on the alternatives for control gtf bredthiny emissions Additionshy
ally both tanks are in the range of 20U00 gallons which is a capacity where
floating roof tanks are almost nonexistent Out of 670 floating roof tanks
surveyd by EPA less then 15 were smaller than 30000 gallons in capacity
Therefore the utilization of any floating roof concept and seal combination
will not be considered
Commmerc i a11 y it is estimated by Ou Pont that carbon tetrachloride
feed costs approximately $017lb (E Taylor personal communication)
Ttierefore less than i2500 is lost to l)uPont annud l ly as a result of working
l eve l em i s s fo ns and 1es s fr om Al l i ed bull Thus i t is u n l i kel y that any
appreciable economic incentive exists to dev1~lop ci vapor recovery system
However a Cdndidate system could be pattern~d after the chloroform
feed-storage unit currently at Allied This is a dedicated vapor balance
system Chloroform is off-lndded from tank cars and stored in a closed
unvented tdnk As the 5tordye tank is filled the c1ir spdce d1spldced is fed
Figure 121-3 Installed Capital Cost vs Condenser Area for Various Materials of Cunstruction for a Complete Condenser Section
182
back to the tank car Also in this case the storage tank is not vented in its
normal breathing mode and is part of a closed feed system to the reactor
Vapor recovery system alternatives which are particularly effective
in loading and handling include r~frigeration adsorption andor absorption
Control efficienci~- 1re est1mateJ in the rlt1nge of 90-95 (EPA 1980b) and of
course depend on the speci tic su1)Stance and equipment used Carbon adsorpshy
tion systems were discussed above The smallest capacity carbon adsorptionmiddot
system priced by EPA (EPA 1980b) has 2 vertical beds of carbon (900 lbs - 4ft
diameter by 3 ft depth) witl1 an ilstalled cctpital cost of $135000 based on
December 1979 dollars
Dow Tanks - Dow has two pair of p~oduct check tanks which alternately are
filled dnd off-loaded Emhsions from th~sl tanks were calculated based upon
operating cycle (3 day fi 11) satubull-dtion vapor pressure and physicdl
chardcteristics Direct hec1d-spa1e testing was planned however it could not
be accomodated because of r 1stri c ed access due to unscheduled mai ntendnce on
the field test day
Dow has indicated that chey are studying the option of installing a
vapor controlrecovery systt~m in 1hese and their largerproduct storage tanks
The extent of their engi neri ng ind assessment work is unknown as are their
current plans It may be possibk to incorporate a vapor balance design into
the system whereby the disp1aced apor is trdnsferred to another process point
within tl1e system Alterna1ively a vapor r1middotcovery system such as carbon
adsmiddotorption is fedsible Hot-1ever since emissions for t11e check tanks are
primarily due to working loss tht~ use of a conservation vent or an adjustment
of its operational differential p~essure would be ineffective for tr1e
reduction of the bulk of such emissions Furthermore since check tank size
is re1at i ve 1y sma 11 no cons i derctt I on was given to conversion to a fl oat i ng
roof configuration for those tanks Conversion would be possible for
the large storage tanks
12~ WASTEWATE~ EMISSIONS - STAUFFER
Emissions of me throug11 ~-a~)te-1dter discharge are the largest single
source identified at the plctnt sites The discharge limit was set at 25 ppm
by the Los Angeles County Sc1nitation iistrict and was estdblished with the
occupational limit in mind of 50 ppm JVer an 8 hour work shift dt the District
18~
tredtrient center~ T11ere hctd been suine di~cuss1on between pdrti~s of possibly
raisiny tne discharge limit since it is felt by Stauffer that dilution of the
stream is sufficient to d 11 ow it Itmiddot shou Id he noted t10-1ever tna NIUSH has
recommended the permissible exposure limit be reduced to 5 ppm averaged over a
work period of 10 nours microer ddy 40 ours per Ieek witt1 a cei l iny level of 15
pmicroin dVeraged over a 15-mi nute microet~ v(J
It is not certain at 1~hrt rdte me 1rill be reledscd from the
wastewater stream At the microlant discharge points wastewater is both hot and
aerated thus favoring release~ No measurements have been ta~en downstream of
the plant after considerable dilut1on has occured The distance to the water
treatment pa11t is approximately 5 kin at 1i1ic~1 point anerobic diyestion is
conducted lt is presumed that a11 ELlC will be released before final ocean
di scnar~Ji
A possible emi ss i D~ cont ro process for nmiddotduct ion of the EDC in the
effluent is by ~recess adjust~Pnts or additio~s For example process
modifications to the EUC stripping stage coula dramatically reduce discharge
lev~ls A control alternative is the use of octivdted carnon or XA0-2 resin
to recover EOC in tne discharge stream Tests of Gulf South Researd1
Institute on XA0-2 and activated carbon (Coco 1980) show high recovery yields
for nonpolar organic carcinogens unaer a range of concentrations Viable
suggested alternatives oy Srnitn included regeneration of Ue trapping
materials dnu even consideration of burning tt1e concentrate carbon media (at
greater than 1000 pmicrom)
Control approaches to reduce emissions from so called secondary
sources in general and waste water emissions in particular fa 11 into four
categories - waste source control resour(e recovery alternative disposal and
add-on controls Alternative control pro(~esses which were considered but
appear to be inappropriate to this case i11clude chtmiddotmica means eg
neutralization precipitation coagulatior1 and chemical oxidation thermal
destruction of tne unconcentrated ~dste s1ream is ii1practical biological
treatment eg aerdtion and biomass-wasteater contcct ~ienerally relates to
the treatment of soluole degrddable organics in the concentration range
bet~~een U01 and l~~ terminal storagt eg lanctfilliny urface impoundment
and deep well injection dre either indppl 1cable or 1mmicroractical Therefore in
summary the poss i b I e contra l approac ties fur this ca e inc 1ude the improvement
r
184
of semicrodration efficiencies in steam stripmicroing tl1e internal recycle of ~aste
stredms ctnd the ddsorption b activdted carbon The design configuration
~fficiency and cost )bviously depend upon numerous microlarit specific factors and
their determination vJOuld rec-Jire detailed analyses
n1e ~~dsrewctter systbull-m of a model cl1emical production plant based
upon the average properties uf a composite of 30 chemicals was evaluated for
EPA by IT Enviroscience (EPA 1980b) Included prominantly among the 30 was
tUC with the hignest uncontrolled secondary emission wastewater release rate
ie ~ percent of tt1e production and 34 perc~nt of the emission Cost and
impact analyses were evaluated for alternative control systems to reduce
secondary voe emissions from wastewater Four systems were considered a
carbon adsorption system (eAS) for recovery of the voe from the wastewater a
cover to reduce secondary VOC emissions from the wastewater clarifier a cover
for the clarifier plus a carbon desorption system and a cover for the
clarifier plus ct eAS system usinu a fume incinerator The scale of the model
system was greatly in excess of the Stauffer plant thus further making
detailed comparison i mpract i cal bull However emission reduction factors ~-1ere 6diven as ~9 and cost effectiveness per 10 g reduction genera1middot1y ranged
between $450 to 1733 These factors would likely grossly underestimate the
system cost if scaled down to the range of the Stauffer plant ie the order
of 34600 lb annual discharge
The alternative approaches of improved steam stripping efficiency
and internal recycling of the stripper iischarge stream with a reduction in
makeup re4uirements could decreas~ net ~DC wastewater content release
123 CONTROLS FUR SECONDARY LEAD S~1tLTER STACK EMISSIONS
Given our findiny of low (16 kgyr) ~missions of arsenic from the
reverberatory furndce at RSR it middotJOuld ippear that the arsenic content of themiddot
lead feedstock is low andor that the microants system for reducing lead
emissions is dlso quite effective for arsenic ~SR it will be recalled from
Section 32 uses a quenchinu cnariber ctnd Dayiwuse filters to remove
pctrticuldte matter and a carbonate scruooer to remove sulfur dioxide
There are no federal new source microertormance standards for lead
eini ssions per se t10wever the Standdrds of Pbullrformance for Secondary Lead
SmeHers (40 CFR 6Ull2) limit total particuLte emission from a blast furnace
185
or reverberatory furnace to SU mgm 3 According to Augenstein et al (1978)
who reviewed the techno1ogy for control ing lead ~missions from ttwse sources
baghouse filters or wet scrubbers are yenerally used to contro1 particulate
emissions When fabric filters are used to control blast furnace emissions
they are nurma11y preceded by an afterburner which inciner~tes hydrocarbons
that bull~GU Id otnerwi se b 1 ind t ne f o1 ur-i c After-burners a re nor necessary for
reverb2middotmiddotatcry furnace emiss-ion smiddotnce the excess air and temperature are
usually sufficient to oxidize tn2 hydrocarbons
According to Augenstein et ali 11 sr1aker--type baghouse filters are
the 1nost effective means of controlling ead fume emissions from secondary
furnace opErations 11 Collection etficiencies can exceed 99 oercent One
advantage to this control approact is that l~ad oxide dust can be recovered
easily cmd recycled in the smeller Flue gases must be coated to below 300 degF
for dacro~ bags and to below ~00 degF for fiberglass bags (Hi~h et al 1977)
Although wet scrubbers are effecti1e under some circumstances in
controlling lead emissions it is more difficult to recover the lead oxide for
recycling In addition sulfur dioxide presi~nt in the flue gases becomes
oxidized to sulfuric acid and can cause corr1sion problems For this reason
sodium carbonate or other basic reagents ar~ added to the scrubber solution
Although it concerns a yold smelter an approach described by
Marchant dnd Meek (1980) provides an exampli~ of an arsenic control
alternative which might be apmicrolicatgtle to secmdary lead smelters At the
Campbe1 1 Red lake Gold Smelter in Balmerton Ontarion Canada Smelter gases
are first passed throuyh a hot electrostatic precipitator (ESP) The ESP is
heated so that the arsenic (which is principally in the form of As ) remains2o3 gaseous and is not yet collected This exclusion of arsenic allo~~s the ESP to
recover particulate gold rnore easily The ESP exhaust is then quenched with
ambient air to condense the arsenic trioxide Baghouse filters then remove
the arsenic along with other particulate matter
124 CUNT~OLS FUR STEEL MILL EMISSIONS
Given the imminent and irreversibli~ cessdtion of coking activities
at Ka 15 er Stee 1 Corpora t i on a rev I e~ o r t e c Imo 1 o gi es for ( on t r o 1 l i ng
~missions from the coke ovens rnd he Clt1ke byproduct recovimiddotry pldnt WdS not
deemed to be necessary Tnis is t1e only primary steel mill in California
1Pi6
KEFUENCE)
Al lf~n Jr Ll~ llJBU1 J 111odel to 1middotstimite hi1drdous (bullmissions from coke oven Juurs pr~pdrelt1 DY 1-kSecircn Tr1dnye7nrffut fur U~ Environmental Protection Agency Uffi ce of Air ~ua 1i ty and Standards fltesearch Tri ang I e Park Nortn Cdrolina KTl17362-UL
Jl len Jr CC 1~8Ub tstination of chdr in(i emissions for by-product coke oven microrepared by fltesedrch Tr1anglt Insc1tute for US nvironmental Protection 1-yency Uf f ice of Air l)ua l i ty and Standards Ke search Triangle Park North Cdrolina RTI17362-0
Allen Jr CC 198Uc 11 Environmeritdl c1ssess1ent of coke by-product recovery plants Proceedings First Symposium on Iron c1nd Steel Pollution Abatement Technology Ctncayo Illrno1s--Tin1ctober - 1 November 1979 EPA-600-9-80-012 pp 7~-dd
Augenstein ur1 T Cornin et al 1978 Con_trol techniques for lead air emissions prepared by PEDCU Envircnmental Inc for US Environmental Protection Agency Research Triangle Park North Carolina EPA-4502-77-012-B ( NT IS P88U-197551)
jarber WC 1977 Interim air pcllution a~sessment for eight t1igh-volume industrial organic chemicals 11 US Environin~ntal Protection Agency memorandum to Air and Hazardous Materials Uivision Kegions I III-X and to Uirector Environmental middotPrograms Uivision Region II
t3drrett KE WL Margard et al 1977 Sampling and analysis of coke-oven door emissions Prepared by ~dttelle-Columbus Laboratories for US Env 1 ronmenta I Protection Agen y Industri a I Environmental Research Laboratory Research Triangle Park North Carolina EPA-b002-77-213
t3 a r ret t flt bull r bull and P bull K bull ~~ebb 1Y78 bull 11 E f f e ct i venes s of a wet el e ct r o stat i c precipitator for controlling PUM emissions from coke oven door leakage Presented at 71st Meetiny of the Air Pollution Control Association Houston Texas 25-29 June 1978
l3asdekis HS 1980 Carbon Adsorption Control Uevice Evaluation IT Enviroscience Inc report in prepdration for US Environmental Protection Agency ESEU
8ee fltW et al 1974 Coke oven charging emission control test program Vol I c1nd II US Environmental Protection Jgency EPA-6502-74-062
~eimer RG 1978 Air pollution emission test Analysis of polynuclear aromatic hydrocarbons from coke oven effluents Prepared by TRW Uefense and Smicrodce Systems Groumicro for US tnvirorunental Protection Agency Office of Air wuality Planning and Standards Kes~arch Triangle Park North Carolina EMB Kemicroort 8-CKU-12
187
Braman RS 1976 Application of the arsine evolution methods to environmental analyses presented at the International Conference on Environmental Arsenic Ft Lauderdale Florida 5-8 October
Bratina J E 1979 11 Fabric filter applications on coke oven pushing operations J Air Poll Cont Assoc 29(9) lll6-920
tiuonicore AJ 1980 Environmental assessment of coke quench towers in proceedings First Symposium on Iron and Steel Pollution Abltlternent Technology Chicago~ Uirnois JO October - f N~vernber 1979 EPA-6009-~1b-62~ pp 112-142
t3usse A D and dR Zimmerrao1 1973 User 1 s guide for tdegh~ Climatological Uisper~ion Model US Environmental Protection Agency Resea~ch Triangle Park Nortl Carolina) EPA-R4--73~0024
California Air Resources 8oard 1976 Coke oven emissions miscellaneous emissions and their control at Kaiser Steel Corporations Fontana Steel making facihty Enforcement l3ranch) S2crcmento California L amp E-76-11
Calvert~ CA 1976 Envir __ Sci_ i( __ Technol 10256
Coco L~i 2t aL~ 1979~ 11 02c10pmert of trec1tment and control technology for 11refractory pet rochemi co was tt=1s r middotl 1d l Report Gu 1f South Research In st it ute
to U S tnv i ronmenta l Protection Agency EPA-ti002- 79-080
Colucci~ LM and CR Begemi 1971 Palynuclear aromatic hydrocarbons and other pollutants in Los Angeles air In Proceedings of the International Clean Jir Congress Vol 2 Academic Press pp 28-35
Cooper JA and JG Watson Jr 1980 Receptor oriented methods of air partkulate source apportionment 11 J Air Pollution Control Assoc 30(10) 1116-1125
Coutant R W JS McNulty and RD Grammar 1975 Final report on determi 11at ion of trace elements in a combustion sys tern Prepared by Batte11 e Columbus Laboratories for the Electric Power Research Institute EPRI 122-1
11 11Dilling W 1977 lnterphase transfer processes Envir Sci TechnoL 11 4 pp 405-409
Oow1irgl) MP JO Jeffry and AH Laube 1978 11 Reduction of quench tower emissions Presented at the 71st Air Pollution Conotrol Association Conference Houston Texas 25-30 June APCA No 78-92
EPA 1980a Review of Criteria for Vapor Phase Hydrocarbons 11 US Environmental Protection Agency ~eport 6008-80-045
EPA 1980b Organic Chemical Manufacturing Volume J Storage Fugitive and Secondary Source EPA report 450J-80-025
EPA 1981 Compilation of Air Pol lutdnt Emission Fdctors (Including Suppleshyments 1-7) Third Edition Suppl1111ent 12 Utfice of ir l)uality Planning dnd Stdndlt1rds AP-42-ED-3-Supp I - lt
188
Erikson Uli 1~80 11 Control Uevice Evaluation Condensdtion 11 IT Envioscience lnc report in prepdration for US Environmental Protection Agency ESEU
lrtel liL 197Y Quench tower particuldte emissions J Air Poll Cont Assoc 2Y(9) 913-916
Ev a n s K bull P bull 1981 Pe r son a l Commun i cat i on v i th R bull Zi s k i n d bull
Goodwin DR 1980 11 Air pollution emissions standards in Proceedings First Symposium on Iron and Steel Pollution Abatement Technology Chicago 11 lrno1s 30 October - 1 November 1979 EPA-6009-80-012 pp 37-45
Gordon GE 10 Receptor Models Environ Sci Tech 14(7)792-800
Gordon KJ 1976 11 lJistribution of airborne polycyclic aromatic hydrocarbons throuyhout Los Angeles Envir Sci Technol 370-373
Great Lakes Cdrbon Corporation 1977 Study of coke side coke-oven emissions~ Vol 1 Source testing of a stationary coke side enclosure EPA-340l-77-014A
lirimsrud EP and HA Rasmussen 1975 Atmos Envir Y 1010
Hartnan MW 1~80 Source test c1t US Steel Clairton Coke Ovens Clairton P~nnsylvania Prepared by rRW Environmental Engrneerrny D1v1s1on for US lnv1ronrnental Protection Agency Emissions Standards and Engineering l)ivision Research Trianyle Pdrk North Carolina EM~ Report 78-CKU-13
Hendriks Ilt V et a I 1Y7Y Urgani c air emissions from coke quench towers 11
Presented at 2nd Arrnual M(etiny ot the ir IJol lution Control Associdtiont Cir1cinndti Utlio June 1~7lJ l-dmicroer NU 9-391
Higt1 MU ME Lukey dnd TA Li Puma llJ77 Inspection manual for secondary lead smelters premicrolt1red b Enyineeriny Science lnc for US lnvironmentdl Protection Ayency Office 0f ~nforc~nent EPA-3401-77-001
Holzer G et al 197 11 Collection amp analyses of trace organic emissions from natura I sources 11 Journal of Cnromatogra flhY 142 pp 755-764
Science Applications lnc Final lemicroort to US Environmental Protection Agency tPA-560lJ-80-UU2
J bull Mi l n e l lJ 81 Person a I Cornm un i cat i on wi t t1 bull Zi ski nd bull
Kemner WF 1979 Cost tffectiven2ss model for pollution control at coking facilities Premicroarect oy PEDCo Environmental lnc for US Environmental Protection Agency Industrial Environmental Research Laboratory Research Tridnyle Park Nortll Carolind EPA-GU02-7l-1U5 (NTS PB 8U-118706)
Kessler T AG Sharkey Jr and kA Friedel 1973 Analysis of trace ele111ents in coal by smicroark-source mass spectrometry US 8ureau of Mines Report ot I n v es t 1 gd t 1on s No bull 77 14 bull
189
Kolak NP J Hyde and R Forrest~r 19 1 9 Particulate source contributions in tt1e Niltlgara Frontierffi Prepared t)_y Nr2bulll York State IJepartrnent of Env i ronm~ntc i Con serv d ti on for US En vi r inment ct l Protect ion Agency Region 11 EPA-9024-79-006
Mackay U and PJ Leinonen llt175 Hate of Evamicroordcion of low - solubility contaminants from vJater bodies to c1tmospmre 11 Envir Sci Tecnnol 9 13 pp 1178-libU
Magee tiv HJ Hctll and GM Vc~rlt_a Jr 1973 Potential Doimiddotiutants in fossi1 f1Jel Prepared by ESSO fbullt~scarcr middotnd Engineeriny Co tor US Enviromihm~ct Protection Agency Ufl-R2-7 --2l1~L (NTIS Pt3 as u~9)
Marcriant middotG0H ard RL Meek 1~gu Eva uation of technology for control of arsenic tmissfons at the CdlmpJel Hd Lakmiddot G~ld Smelter prepared for the US
1Env i ronmentd l Protection Agenc)-l-l(lU-sfrT l lnv i ronmenta l kesedrCiI Laboratory Cincirrn2ti Uh70 lYA-60U2-8U--141 (NTIS 8 oU-21Y36J)
Margler L$ HA Ziskind M8 lcgozen (Ind M Axelrod 1979 11 An inventory of care i nog)n i c substances re i ea sect into ne ambient air of California Vol une I - Scn~enin~ and id2nt~fication o-f r~1rci1ogens of greatest concern Science J- pp 1i ca middott~ L J s In c bull Fi nd l kcmicro~ rt ~ i I - U 6 C) - U- 5 0 4 to U1 e Ca I i ~- o r n i a Ai r kesources 130ard
Milne J 198L Persond1 C~mmunicdtmiddotion middot1itt1 R Ziskind
llurchio JC WrC Cooper and A ue Leo11 1970 Asbestos fibers in ambient air of California 11 Cal Horn i ct Air lesourc2s l3oa rd ARl3 4-0S4- l EHS Report 73-2 (March)
National Academy of Sciences 1Y72 Part1culate polycyclic organic matter Washington~ DC pmicro 4-ll
Nati ona I Kesedrcn Counc i I 197b Arsenic prepi1 red by Subcommittee on Arsenic Co111mittee on 1bull1edical ctnd t)ioloyicai Effects of Environmental Pollutants washin~ton uc
Oliver JF and JT Lane 1979 11 Control of visible emissions at CFampl 1 s coke plant-Pueblo Colorado J Air PolL Cont Assoc 29(9) 920-925
Pe I I i z z a r i E bull 0 bull anct J bull E bull l$ u n c h 1s 19 1 nbi en t ai r ca r c i n o gen i c v a po r s Improved samplin~ and analysis tectrnology EPA Contract 68-02-2764
Pellizzari E U 1977 11 Tle 11easure11ent of carcinogenic vamicroors in ambient middotatmospheres EPA-b007-77-J55
Phelps RG 1Y79 lSI-CPA-oatLlle c-ilte oven door sealing proyram J Air Pol 1 Cont Assoc 29(J) JlJ8--JlL
Porter RA 1976 Uismicroersion ~yuation s lucions Dy calculator I guide for air pollution engineers ctnd sci12nthts S cond edition rexas Air Control tioard ~MT1S P~ 262 lJ~)
Redburn T 1980 Kaiser battles aginst l~gacy of woes 11 Los Angeles Timesmiddot (7 Semicroternber irno) Part VI pp 111
Research Triangle lnstittJte 1977 11 Qudntification of benzene in 15D -1bullnoient air samples 11 for US Environmental 11rotection Agency Office of Air Quality Planning a11d Stitndards
Rinkus SJ and MS Legator 1979 Chemical characterization of 465 known or suspected carcinogens and their correlation with mutagnic activity in the Salmonella typhimurium system Cancer Research 393289-3318
Koberts R M 1980 11 An inventory of arc i nogen i c substances r~ leased nto the ambient air of California Tasks 11 and IV KVB Final Report KVB 26900-836 to Science Applications Inc
Rogozen MB LW Margler P Mankiemiddott1icz cind M Axelrod 1978 Water pollution control for coal slurry pipeiines Prepared by Science Appl1cat10ns Inc for US Department of Energy (NTIS SAI-068-79-516
Rogozen MB DF Hausknecht and RA Ziskind 1976 Methodology for ranking elements in fossil fuels according to their potential health impact Science Applications Inc Final Report 260-77-539 to the Electric Power ~esearch Institute
Russel 1 JW and LA Shadoff The sdmpl ing and determination of halocarbons in ambient air usin~ ~rncentration on porous polymers
Siebert PC CA Craig and Eb Coffey 1978 Prelimary assessment of the sources control and population to airoorne polycyclic organic matter as indicated by benzo(d)pyrene Prepare( by lnergy and Environmental Analyses Inc for US Environmental Protection Agency Office of Air Quality Planning and Standards Resedrch Triangle Park North Carolina
Simmonds PG et amiddot1 1974 Distribution of atmospheric halocarbons in the air over Los Angeles basin Atmos Envir Vol 8 pp 209-216
Singh HB LJ Salas and RE Stiles Distribution of Selected Gaseous Oryanic Mutagens and Suspected Carcinoyens in Ambient Air in Proceedings Annual Meeting Air Pollution Controrl Association 82-651 June 20-25 1982 New Url eans
Smith WM 1979 Evaluation of cok oven emissions in Yearbook of the American Iron and Steel Institute pp 163-179
SRI International 1980 Asessment oi human exposures to atmospheric benzene SJ Mara and SS Lee finai report to US Environmental Protection Agency EPA-4503-78-031
Taback HJ 1978 Control uf Hydrocarbon emissions from stationary sources in the California South Codst Air Basir Prepared by KVB Inc for the Cal1fornid Air Resources Bod-rdmiddotKVB No 5804-714
191
Trenholm J-f dnd LL )eek 1Y78 11 Assessment of t1azardous organic emissions fror1 slot type coke oven Datteries unpublished paper US Environmental l)rotection Ayency Ernssion Standdrds and Engineeriny Jivision
Turner U~ J970 Workbook of atmusphe~ic dispersion estiimiddot1Jt2s IJS Environmental Protection Agency k~sedrch Tri1ngle Park North Carolina
US Bureau of Mines lYl f-inercls yearbook 197U Vol 11 p 4~3
Van Os dE I l U bull iJ bull lL Ma rs l and et ct 1Y7~1 o En i r onmt~ nta I a s s es smen t of co k e Dy-product recovery plants prepareu by ReseMcn Trictngle lnstitute for US Environmental Protection AgerKy 11cJustrial Environ11ental R~search Laboratory Kesearcll Tri2n~e Idrk NorU1 Caroline EPA-6U02-7J-Ul6 (NTfS PB 293-278)
Westbrool( C W 1979a Level l ass~ssment of uncontrolled sinter plant emissions Prepared by fesearch Triangle Institute for US Environmenta1 Protection Agcncy Industria1 tnvirnnmental fltesearc-1 Laboratory kesearch Trianyle Park North Carolind EPA-bOU2-7Y-112
Westbroollt C~~ 197ltJb Level l css~ssment of uncontro1led l)-BOP emissions Prepared by Researcn Triangle institute for US Environmental Protection Agency Industrial Environmental Hesearch Laboratory Research Triangle Park ~Orth Carolina EPA-6002-79-190
Wetherold rLb LP Provost and CD S111ith 198U Assessment of atmospneric emissions from microetroleu111 refinin~ Volume 3 Appendix l3 Radian Corp Fina Report to US tnv i ron111enta I Protection Agency EPA-600 2-80-07 5c
Ziskind RA OF S1ilith JL Hdhn etnd G Smicroivey (1980) Determinants of Cancer and Carctiovasculdr Uisease Mortality in Asbestos Mining counties of California SAi Report 068-dl-514 l May
192
APPENDIX A
Rapid Screening and Identification of Airborne Carcinogens of Greatest Concern in California
Lawrence W Margler Michael B Rogozen Richard A Ziskind and Robert Reynolds Science Applications Inc Los Angeles California
This paper describes a method for establishing a priority list of airborne carcinogens within a
state jurisdiction In this case it was necessary to identify from among hundreds of potential
candidates the live to ten materials of greatest potential concern In California as airborne
carcinogens
Because no previous Inventory of carcinogens In California existed published lists rankings
and assessments of national scope were used to Identify candidates By systematic manipulation
and comparison of these data sources 47 materials of some notoriety were chosen for closer
scrutiny This selection was pared to 22 candidates largely by eliminating those which had
very little production and use In California (Substances primarily used as pesticides were
excluded from the scope of this study) The remc1ining candidates were then ranked by additive
and multiplicative algorithms and by a panel of experts The results of these rankings were
combined to produce a single selection of 11 priority candidates In alphabetical order they
ethylene dichloride N-nllrosoamlncs perchloroethylene and polycyclk aromatic hydroiarbons
In continuing studies a baseline emissions inventory Is being prepared and a source testing
program is being designed
In recent years concern has grown over su~pected arcinug-ens New Jersey seshythe possibility that certain materials lected ten volatile organic compounds released to the atmosphere through inshy and five htbullavy metals to be examined dustrial and commercial activity may be further and is currently measuring responsible for a significant portion of ambient atmospheric concentrations of the incidence of cancer in the general these substances in a variety of areas In population This concern is manifested the second year of the study the state at the federal level in the National h1s incre1sed the volatile organics Emissiltm Standards for Hazardous Air stt1died to 20 and begun measuring Pollutants (NESHAP) which limit heavier orianics associated with parshyemissions of the known carcinogens asshy ticulatcs2 In California a very different bestos beryllium and vinyl chloride 1 approach-was taken
Only a few states including New Jersey and California have begun efshy Overview of Californias Approach forts to identify airborne carcinogens of concern to the general public for the The California Air Resources Board purpose of setting state emission regushy (( ARB) is sponsoring a three-stage lations for these substances After reshy st 11d~bull of airbornC rmissions of carcinoshyviewing national use data for known and g1middot11s from anthropogenic atlivities The
November 1979 Volume 29 No 11
first stage which is the subject of this paper was to identify roughly five to ten materials which of the hundreds of known or suspected airborne carcinoshygens are most likely to be of greatest concern to Californias general populashytion Also of interest were those which in order to satisfy occupational health and safety regulations might be transshyferred from the workplace air to the outside atmosphere The second stage which is now underway includes pinshypointing of emission sources for each of the carcinogens of importance quantishyfiltation of emissions nnd design of source-testing methods A subsequent stage will consist of source testing and measuring public exposures to those substances for which data are unavailshyable The basis for regulatory action if appropriate will include the results of this program and other related reshysearch
Screening of Candidate Carcinogens
The National Institute for Occupashytional Safety and Health (NIOSH) lists 1905 chemicals which have reported neoplastigenic or carcinogenic effects and 510 which have otherwise received attention for their neoplastigenic poshytential1 The need to select five to ten materials from such a large number of potential candidates dictated that we devise a way to rapidly eliminate from further consideration the vast majority of the substances Given the paucity of published data on most of the candidate substances the screening method was designed to make best use of readilymiddot
C1111yrich1 19i9-Air Pullulion Control A~~ociatlon
1153
A- I
----- ---- ----- ---------------------available information The screening process was as follows (1) eight comshyplications of known and suspected carshycinogens were reviewed and those subshystances which were not used in Califorshynia were highly unstable in air or were very doubtfully carcinogenic were eliminated (2) after more detailed inshyformation was obtained for the reshymaining 25 substances candidates were rated by two different analytical methshyods (3) an expert panel was convened to review dossiers on the candidates and to independeitly nmk them (-i) from the eight to deven substances ranked highest by all three approaches eleven were selected for the emission identifishycation and source-testing design stages of the CAHB effort
lnitiI $cre~ng ot Potential Candidate Carcingtgens
65 compnunds middotwe ire selected from 642 industrial ur~nnic air pollutants comshypiled by MrTRE Corporation for the US Envircmnental Protection Agency (USEPA) in U1at study pollutants were scored by multiplying four explicshyitly defined rating factors annual US production fraction of production lost to the environment volaflity and toxshyicity To adapt this york to our pmpose we first selected the 114 substances listed as being carcinogeni 1~ or neoplasshytigenic Then the scores of each of these compounds under the criteria annual US production fraction of producshytion lost volatility and carcinogeshynicity were multiplied together Seshylected for further consideration were those substances which had a product score above 50 Another 15 substances listed as being carcinogenic but lacking information for one of the other rating factors were also selected This list of 65 was then compar~d with seven other lists of corcinogeris1
-middot 11 Materials comshy
mon to the reduced MITRE list and at least one of the other lists were chosen for further consideration Added as candidates were those suhstances which are regulated as occupational carcino-
gens by the Occupational Safety and Health Administration (OSHA) and certain inorganic carcinogens Finally substances were added which in our judgment should be investigated but had been eliminated at this point Exshyamples of these are bis(chloromethyl)shyether epichlorohydrin and hydrazine
Next the refined list which now contained 47 substances or chemical groups was pared further hy another rapid screening process Eliminated
were all candidates (1) whose production andor use in California was very low (under 10~ lbyr) and wns not thought likely to pose a risk to a localized popushylation (2) which are very unstable in 1ir or CH whilh should not on t lw hnsis of CUtrltbullnL cidlnre liebull considtbullnmiddotd r11rci-
1154
Tahllt I Suh tances n-iewed in letail
Candidate Substances
Arsenic Inorganic lead Ashestos Alkyl lead Benzene Maieic anhydride
nogenic The result of the initial the rating under that criterion and the screening was a list of 22 candidate mashy corresponding criterion weight The terials which is presented in Table I overall rating for pollutant i is then the
sum of the scores under all the crishyteriaRanking Candidates by AddiUve ancll
Mulllpllcatlve AigorUhms The additive approach has several virtues the main one being that it forces
Many screening or ranking systems the user to make all assumptions exshyfaH into one oft wo cat~middotgories additive plicit In the process of setting up such and multiplicative Some systems are a a ranking system new insights into the combination of the two while others problem under consideration may be combine nn ol1jective appronc 11 with g1ined Once the system is set up it is suhjlctive cvnluition of the remiddot ults 1l rdatively easy to use Where data for The L2 substances surviving th1middot initial scoring pollutants are unavailable arshyscreening were ranked independtmiddotntly by tificial scales can be constructed to the tvo approaches If the same subshy quantify subjective factors Finally the stance rated high~middot under both systems sensitivity of the results to the systems its importance to California was judged subjective aspects may be measured For to be more likely than if it had scored example one can determine the effect of highly in only one method changing criteria weights upon the final
Additive Approach In the additive pollutant ranking Similarly an appreshyapproach the user identifies one or more ciation may be gained of the significance criteria and rates each alternatiw subshy of the range of uncertainty for a particshystance against each criterion while sishy ular required data element by varying _ multaneously deciding the relative imshy rating factor values portance of the criteria Eq (1) shows its A fundamental problem with the apshymathematical formulation proach is that there is no logical basis for
adding the individual scores assignedRating for pollutant i = f W1 R11 under the criteria other than the asshy
rbullI sumption that this simulates or even
(1) improves upon the users thought proshyEach criterion or rating factor (1 ) is crss A major operational problem is assined a value for each pollutant and that of weighting the criteria A common each rating factor is weighted (b W1 ) practice is to give all critNia equal -icnrding to its importance reht veto wltgtight hut that is in itstbulllf a st1Ubullm1middotnl lhe otlwr rritnia llw sc11rtbull for 1111 aLnut the 1rcl11tive importance of Ow 1n1 1 undmiddotr nitn1 n j is the prod11middott of n1ttria
Journal of the Air Pollution Control Association
A-2
---------
Multiplicative Approach In the multiplicative approach the rating for each alternative is the product of the rat ihgs under each criterion
Hating for pollutant i --- 11Ill
UJ (~) Jbull I
A multiplicative approach can have ~omc altlvrntnges over additive onrs First in some cases the ratings can be physical parameters such as concenshytrations or volatilities there is then no need to weight the criteria and hence less controversy over subjective judgshyments Second multiplication generally provides a wider range of scores than does addition allowing clearer disshycrimination among alternatives Finally this approach provides results which are more intuitively acceptable As an exshyample of this last point suppose that exposure and harmfulness levels for candidate substance are each converted to values on aO to IO scale and that a certain substance is both extremely toxic and extremely rare An additive approach would give the compound a rating ofO + 10 = 10 which is equivalent to that of a moderately prevalent subshystance (rated say at 5) which is modshyerately harmful (rated also at =-) A multiplicative approach on the other hand would rate the first substance at 0 and the second at 25
Criteria The six criteria used in the additive and multiplicative ranking procedures are defined in Table 11 Asshysignment of values to the Rij was based upon data gathered from published litshyerature personal communications and panel discussion and has been fully documented 13
Because the purpose of this exercise was to determine the relative imporshytance of the suspected candidate carshycinogens R 1 was scaled to the most heavily used candidate substance benshyzene whose annual production and use in California is nearly 109 lb Materials with a use under 105 lbyr would be rated zero for R 1 and rejected Howevshyerbefore rejecting a substance by this criterion we considered whether its emissions could in particular circumshystances result in high exposures to a Joshycalized population
R2 takes into account the fact that the chemical industry is in continual change Substances of concern today may be phased out while the use of others may rise dramatically increasing their imshyporta nee asmiddot poll utan ts Information on developments which could likely result in a change in the growth rate was facshytored into the choice of a value for R2- As an example asbestos consumption in California has been stable in recent years However the pending phaseout of asbestos in motor vehicle friction materials will hasten the decline in asshybestos consumption hence we assigned a value of 1 for R2bull
November 1979 Volume 29 No 11
Ideally one cou1 1 use pouu1c11u
sion as a criterion However in this case emissions were to l e estimated in detail only for the five t1 ten carcinogens fishynally ~elected Thmiddotrtfor(bull a 11w1st1n of tt II iim pol cint ial vus 1~1middotd 1atbulld llpon
knowledge of the s11bstances manufacshyture and USP for R The higlwst rating went to substanc middots which are v idely used especially in consumer products A slightly lower rating went to subshystances which an~ routinely emitted from industrial processes during proshyduction and use Some materials are employed in such a way that emissions are quite low even though tight emission control may not be required by law Materials in this cntegory were assigned a value of two for H3bull Substances which under federal or st ate regulations may not be discharged to the exterior envishyronment but which could be dischJrged by accident received the lowest rating
Each candidate was evaluated (n the basis of its 1ropen-ity to decompbullgtSe in ambient air Materials with half-lives greater than eight hours were considered moderately to highly stable and rated five for R4 bull Low to moderate stability was assigned to substances with halfshylives between zero and eight hours Compounds known to exist in air for only a few minutes would be rated zero
Ta hie II Dcfinitinns of the criteria used
R1 Prbullbullsent use in California
100 of max (109 lbyr) 5 10 1)f max (108 lbyr) 4
1 of max (107 lbvr) 3 01 of max (106 lbyr) 2
001 of max (105 lbyr) l lt001 of max (lt105 lbyr) O
R2 Growth in California use
+ 20 5 +Oolti to +20 4 Positive growth to 0 3 Stahle or unknown 2 DPcline 1 B1middotin~ phased out 0
R Emision potential
Widespread use in ltonsumer products 5 R1middotlatively p1or con rot over emissions 4 Rdatively good control over emissions 2 Tihtly c1int rolled 1
R4 8tabilit~- in ambient Air4 tvlnderale t1 high st 1bility (l 1 middot2 gt 8 hr) 5 Lw to modmiddotrate s1 bility (t1 l -0-8 hr) 3 Unstahle(t2~febull minutes) 0
R Dispcision Potential
Emitted lar~ely as apor or iine 5 particulnte
Emit led largely as rnarse particulate
Rlt Evidence f Carcinu~enici ty
Known or suspected human carcinogen 5 Known mammalian carcinogtmiddotn 4 ~usptctcd 111ammalan carcino~en or 3
known m mmalin muta~tmiddotn Ames ltmiddotst p sitive 2 Prpoundgtcursor 01 co-car inogen
a t11 is the half-liftmiddot
A-3
tion state or anion associations may change in the atmosphere metals do not degrade and were considered stable Asbestos is likewise itahlc Many of the clcc-ompo~ilion rtbullnctions oi orinnic molecules are mediated by light Such substances if released at night would have several hours to disperse in the surrounding area
A rapid way of assessing the relative potential of different substances to spread from a release point is to note their physical state under normal amshybient conditions Accordingly we scored materials emitted as vapors or fine particulates the highest for R5 and coarse particulates the lowest Intershymediate values are possible for varying amounts of fine and coarse particulate emissions from the same source or from different sources
There is as yet no widely agreed upon measure of the relative potenciemiddots of carcinogens although some ranking systems have been proposed14 Extrapshyolating data from in vitro techniques such as the Ames bacterial mutagenicity test and from laboratory animal studies to humans is problematic Therefore a less quantitative measure of the carcishynogenic potential of each candidate substance was used The candidates receiving the highest scores for Rs would be those for which there is strong evishydence of carcinogenesis in humans Exshyamples are asbestos which is implicated in mesothelioma vinyl chloride which has been identified as the agent of liver cancer in exposed workers and bisshy(chloroinethyl)ether shown by epideshymiological studies to cause lung cancer in resin workers The next highest rated substances are those for which human carcinogenicity is unknown but which have produced cancer in one or more mammalian species in laboratory tests Next are those which have not been shown to be carcinogens but which have proven to be mutagenic in test animals Substances for which the only knowlshyedge of carcinogenic potential is a posishytive Ames test (producing mutations in histidine-requiring strains of Salmoshynella) are rated 2 Finally substances which are implicated only as precursors or co-carcinogens would be rated lowest
Substances unequivocally associated with carcinogenesis were considered as carcinogens in this study Conditions of emission and exposure including the presence of co-carcinogens were facshytored into the evaluation of each candishydate where possible Carcinogenic subshystances derived from the metabolism of a precursor were considered as carcinoshygens However ubiquitous substances which have been hypothesized to be precursors (eg secondary amines nishytrous acid and nitric oxide which combine under certain circumstances to
1155
form N-nitrosoamines) were not conshysidered because of uncertainties in the importance of their link to the carcinoshygenic compound and the practical conshysiderations demanded by the scope of the study
It was beyond the scope of this study to jud~e the validity and interpietation of the experimental ad epidemiological evidence upon which the carcinogenicity of candidate substances has been esshytablished We accepted the conclusions about carcinogenicity drawn by the Inshyternational Agency for Research on Cancer or the National Cancer Institute and did not consider dosage or route of administratioll of tested substances However considerations of test validity did enter into the subjective evaluation by the panel of experts
Weight lr1 the sdditive ranking scheme each rating criterion is weighted according to its limportance relative to the other criteria Little precedent exists for assigning these weights In our judgment and as generally agreed by the panel of experts (see below) R 1 RJ and R6 are more important than the other criteria Evidence of carcinogeshynicity was considered to be the most important criterion of aB so W 15 was assigned a valtze of 3 W 1 aid W 3 were set at 2 and W W-i and W5 were set at 1 In order to discern the potential senshysitivity of the rankings to the weight assignments the candidates were also ranked using equaily weighted crishyteria
Ranking Candidates by Panel o~ Experts
Anine-member panel of experienced governmental industrial and academic scientists whose disciplines include~ -organic and physical chemistry indusshytrial hygiene toxicology epidemiology and iegulatory control of toxic sub~ stances was convened to provide addishytional data for our ranking algorithms
to discuss our candidate --ubstanc -sand rejections to suggest pos-ibie ne suushystances for consiceration and Lt rank the candidates indepen11mtly ( f our own ranking Tvo v-ee ilts he fore the meeting pa-d uember-s were jven one-to three-page dossiers on eac I canshydidaie substance
At th~ start of the meeting- before any group discussion the pand was asked to rate each candidate suhtance with a score from Oto 5 Next cmiddotach candidate was discussed at length e provided ~n ovrview and summarized critical issues identifi2d up to that Pbullgtnnt Through materials brought to thtmiddot meetin and their persoril experience panel memshyber Veimiddote able to provide much useful information on the candidates and adshyditional insight in to our ~a ting criteria
At the encl of the two-d1y sc~sion the pa-iel ugain -ated the cai 1didates
Flnal SelediDn
Tabe III show- the highest-scoring substance a~ det(rraine1 by the 1ddishytive and mlltiplictive approaches and by th~ panel in ~e additive approach 21 single r1nking as obuined by avershyaging the two ran kings resulting from using equal and uneguai weights The ran kings ltif most candidi tes were unafshyfected hut carbo11 tetralt hloride chloshyroform chromium and inorganic lead changed more tban thmiddotee positions_ Because uncertainties in the data base predude imputing signifiance to small differences in the Lnal ordering the lists in Table III are prtmiddotsented in alphabetishycal order However it is of interest to point out that benzene consistently ranked hihest
Becausimiddot some cindidates had equal rating scores we ltould nut choose exshyactly ten candidat ~s from the additive and multiplicative rankings Instead the
top nine and eleven were selected from the two exercises respectively We also considered the ten substances scored highest hy the panel at the end of the session The final consensus selection cornsistt-d of the 11 candidate substances appearing on at least two of the three lisL- For the substances included in the cun~ensus ranking a baseline emissions inventory is being conducted and a source testing program is being deshysigned
1ejected Substances
Some comments about certain subshySiF1ces not appearing on the final list are in order inasmuch as they include known carcinogens and compounds which hae received considerable atshyllttion in recent years as occupational c11c111ogens
rouisionally Rejected Substances Appended to the consensus ranking (T1ble III) were vinyl chloride gasoline and engine exhausts tobacco smoke and pesticides No further action by the CRB is recommended at this time for vinyl chloride because it is already subject tu the USEPA emissions stanshydard a CARB ambient air quality strndard and an OSHA standard
Gasoline and tobacco smoke were appcndlmiddotd to this list because each ocshycurs very widely and contains several of the candidate substances reviewed in this study some of which are in the final listing For example gasoline contains benzene ethylene dibromide ethylene dilmiddothloride and alkyl lead compounds the last three being in leaded grades only Both gasoline and diesel combusshytion products include PAHs Tobacco smoke contains among other neoplasshytigenic substances nitrosamines PAHs nickel arsenic cadmium and other heavv metals Manv individuals are inshyvolu~tarily and i~ many situations
Table HI Hi~hest ranked candidates from each rnnkin~ methodbull
Highly ranked but no inventory
l I ighesl consensus recommended Additive Multiplicativ(~ Pan I of eqwrts rnnkmg at this time
1156 Journal of the Air Pollution Control Association
A-4
lirtualy unavoidably exposed to t11-k1ltco smoke Because the sources of gasoline its combustion products and tobacco smoke emissions are well known no specific action was recomshymended for these materials during the tbullmissions innnlory and -ource testing design stages of the present study It was considered importint ho-vemiddoter to draw attrntion to the general publics exposhysure tot hrse substances PrsticidltS are listed for the same reason thou~h ad(bullmiddot tailed examination of pesticides w1s beyond the scope of this study Many pesticides are widely used and some of them are known to be carcinogenic
Other Rejected Substances Acryshylonitrile and vinylidene chloride were placed in a provisionally rejected group because of the panels suspicion that imports of these compounds to Cnlifornia from Japan may be appreshyciable yet are hard to substantiate Should such imports be verified in the future these two compounds would take on greater importance Formaldehyde was also provisionally rejected because the preponderance of evidence indicates that it is not carcinogenic and that bisshy(chloromethyl)ether is not formed from formaldehyde in appreciable quintities in industrial environments 15 The ocshycupationally controlled carcinogens ethyleneimine and beta-propiolactone were rejected in part because of their reactivity in air At the time of this study DBCP a pesticide was no longer heing middotproduced in California and was therefore rejected from further considshyeration
Occupational Regulations and Community Exposure
A question of interest to the CARB was whether the regulation of acknowlshyedged occupational carcinogens adshyverselv affects the ambient air outside the v~rkplace Our general findings can be illustrated by the example of asshybestos
Asbestos is a very widely used mateshyrial for which no ambient air standard exists Concentrations in the workplace are limited to an eight-hour timtmiddotshyweighted average concentration of two fiberscm=1 of air and a ceiling concenshytration of ten fihcrscm 1bull In meeting this standard exhausting air containin~ asshybestos to the ambient air is not reshystricted except by the USEPAs reshyquirement that there shall he no visible emissions containing asbestos piut ides from s11eh foci lit i1bulls (bullxd11din~ hrak(bull shops 1 Considning that undc-r certain conditions 101 asbestos fiherscm1 could ht Pmitted without being visihl( 11 this standard mamiddot allow considerable asshybestos emissicns It is unlikely however that emissions would actually approach these levels First since the OSHA standard cannot generally be achieved
November 1979 Volume 29 No 11
by ventilation n ine the generation or asbestos partid in the workplace must be greatly reltlu ed Second the air is usually filtered 11gt prevent recirculation of asbestos to 1 workplace Asbestos waste must bed 1sposed of in sealed imshypermeable bag~ or containers Thus under current onupational regulations the amhient air 1enerally appCars to be afford(bulld greatn protection than it would without s11ch regulations 1
Conclusion
The screenin and ranilting methodshyology presented in this paper proved to be a feasible approach to establishing a priority list of Jirborne carcinc)gens in California We ftel that it is an efficient means of focu~ing further efforts on emissi()ns inventories source testing and ambient measurement for it not only identifie all the carcinogens of potential concern but it also permits the state regulatory agency to direct its reshysearch resourCtmiddots toward those subshystances of partilmiddotular interest within its jurisdiction
References l National Elllission Standards for Hazshy
ardous Air P llutants A Compilation as of April 1 19 3 PEDCO Environmenshytal Inc for US Environmental Proshytection Ag1middotncy EPA3401-78008 (NTIS No I B 288 205) 19i6
S Gray Ne Jersey Department of Enshyvironmental l 1rotection Trenton NJ private com111unication 1979
1 H E Chrislllsen E J Fairchild and R J Lewis Sr middotmiddotSuspected Carcinogens A Suhfile of th1middot NIOSH Regii-try of Toxic Effects of Chemical Substances 2nd Ed National Institute for Occupational Safety and HPalth NIOSH i7-149
4 B 8 Fuller J Hushon 1 Kornreich R Ouellette L Thomas and P Valker Scoring of rganic air pollut1nts for US Environrnental Protection Agency Mitre Corpt ration T1middotchnical Report MTR-6248 I 9i6
S F R Lozan11 Toxicants Selected for Priority En- 1ronmental Assessment corresponde11ce with EPA Region VI Administrat r 19i7
1gt Survey of inyl Chloride Levels in the Vicinity of Keysor-Century Saugus California US Environmental Proshytection AJet middoty National Enforcement lnvestiiatim Center Denn-rand San Francisco E A-n02-77-017a 19i8
0 Bardin l middotstimon~middot in Hearings on Helationshi1 Bet wfen Cancer and the Environmei 1 US llltuse of Hepreshysentatives bull th Congress 2nd Session Committee (In Interstate and Foreicn Commerce ~crial No 94-141 1976 pp 7~i0
middot T H Ma11middoth Carcinogens in the workplace lwrE to start d1middotaning up Sc-imiddotrmmiddot l17 1-1110) lfi8 (197)
I CH ar1111 Classif1r11t1on of the Exshyposure Haza d Prtbullstgtnled Ii~middot Thirt~middot-Six Clwmirals 011 I llixllmmiddots lternorandum from Clllm1middot I Offi(middote of thtgt Assmmiddotiate lgt1rector for 1 middotarcinoltnesis Division of Cancer Cau~ l and Prevention lo SushyJwrmiddoti-or~middot ( 0 Auditor Gtmiddotneral Acshycounting Oft lmiddote19ii1
I0 Chemicals ith Sufficient Evidence of ( arcinogenit ty in Experimental Ani-
A-5
ma1s Jnlernauona1 Agency 1or 1c-curcn on Cancer IAHC Working Group Reshyport Lyon France 1978
11 Threshold Limit Values for Chemical Substances and Physical Agents in the Workroom Environment with Intended Changes for 1977 American Conference of Governmental Industrial Hygienists 1977
12 M H Rogozen D 1 Hausknecht and R A Zisk ind 1976 A Methodology for Ranking Trace Elements in Fossil Fuels According to Their Potential Health Impact by Science Applications Inc for Electric Power Research Institute 1976
1t L Margler R Ziskind M Rogozen and M Axelrod An Inventory of Carcinoshygenic Substances Released into the Amshybient Air of California Vol I Final Reshyport Screening and Identification of Carcinogens of Greatest Concern by Science Applications Inc for California Air Resources Board 1979
14 T H Maugh Estimating potency of carcinogens is an inexact science Science 202 (4363) 38(1978)
15 C C Yao and G C Miller Research Study on ais(Chloromethyl)Ether Forshymation and Detection in Selected Work Environments by the Bendix Corposhyration for National Institute for Occushypational Safety and Health Cincinnati Ohio Contract No 210-75-0056 1979
16 Occupational Cancer Control Unit Calshyifornia Department of Industrial Relashytions Los Angeles California private communications with staff members 19i9
This paper was developed while Dr Margler was a Staff SriPntist in the Energy-Environment Systems Division of Science Applications Inc (SAIi 1801 Avenue of the Stars Suite 1205 Los Angeles CA 90067 and Mr Reynolds was a Project Manager with the California Air Reshysources Board Sacramento CA 95812 Dr ltar~ler is currently emshyployed as DirEgtrtnr of Bnvironmental Scitbulllltcs for WinzlN nnd Kelly Conshysultini EnintNi t1 Third Stretbullt EurekH CA 9gtgt01 and Mr Heynolds is Air Pollution Control Diretbulltor for the Lake County Air Pollution Conshytrol District 55 N Forhes Street Lakeport CA 95-13 Dr Rogozpn is a Staff Scientist and Dr Ziskind is Deputy Manager of the Energy-Enshyvironment Systems Division of SAi
1157
Appendix f~
_l)ispobullition of Miscellaneous Sites
Gould Inc Ver_n)_~ St~condary Lead S~_n_elter
The emissions of arsenic from the four large secondary lead smelt~rs
in Cali forni) J1~1 ~ bull~tiilnted in the program This estimate was based upon a
uniform fugitive emission f1cJr nt WPll supported by measurement inforriation
Therefore source monitoring was r2cq11)nded 1nd Gould Inc Vernon being
~he largest was singled out It was however proposed to monitor both Gouli
and RSR Corp (Quemetco) in the City of Industry since analysis of the plants
revealed significant differences in plant equipment and engineering The
latter being more typical of 1 nodern facility
As a r2sult Jf pr-~-middott -iiscussions and plant inspections we became
aware that Gould was actively in the process 1)f constructing a new facility
which would completely replace the existing one We have monitored progress
on the new site and concluded it would he inappropriate to utilize program
resources to conduct field tests at Gould ~rr1issi 1rns from the new Gould
faci 1ity wi 11 be bas~rl 11p11) test results from RSR
PG ampE - Pittsburg PG ampE - Salinas a~d So Cal Edison - Long Beach Oil Fuel Power Plants
It was app ro p ri ate tJ cons i dP r the (~mission of arsenic from power
plants during the initial study stage inc~ imiddotrace quantities in the fuel oil
are known to be ernitted Because of the population distribution in tile
vicinity of three plants and some 11realistically conservative estimat2s of
erlissiJn c-1ctors tht~ facilities appearc~d among the top seventeen stationary
sources of potential c1Jn-r-n HovJ~ver as a result of a reexamination of
emission data it was concluded that an error ~dj ~een made in the material
balance of arsenic - resulting in tn emission factor equivalent of a 300
release ~io lit~rature was found to justify 1reater than a 30 transfer
function Thus we estimated at the outside th~ emission factor should be
reduced from 013 lb per 1000 lb to approximately 0013 lb and resulting
ars9nic 2missions for the entire state w~r~ cJ11servatively estimated to be
1760 lbs (from 17600) divided amongst 311 the states power plants Clearly
then the four secondary lPad smelters estimat~ of S9410 lbs of arsenic(
B-1
emissions per year makes consideration of poer pl int ernissio1s if ffsenic a
very low priority
In a literdture review by SAI - Methodokgy __f_or Ranking Trace
Elements in Fossil Fuels Accordi~9__t_~_their Potential Health Iripact (Roqozen
1976) - emissions estimates f0r 15 trac~ substances were researched for c11l
and fuel oil pomiddot1er plant c 1J1v---r-sion Source content combustion pr-)~~
transfer functions and contrcl methodology were co1~idered to develop e~ission
factors The output to input ~atio computed and utilized n -~i-i~ study for
arsenic was 002 to 03 (ie tra1sfer function betw~en 2 11d JO~) reflecti~q
the wide variety of fuels processses and controls nationwiJe The upper end
reflecti1g both high arsenic coals and poor e11iss10n contro 1 devices Clearly
neither of those conditions accurtely apply to th2 thr~~ si~~~ r1nrl thus
substantiate the rlecision not to perform emissions neasur-i-~n~s
Calaveras Asbestos Company - Copperopo l middotis Asbestos Mini n(i and Mi 11 i nq
In the past (late sixties and early seventies) ~h2 ~3laveras
Asbestos Company came under heavy criti-is1 1fter imiddot1sp~ction rieasurements
revealed serious problems However significant reduction of e1nissi0ns have
occured prompted by NESHAP regulation and occupcttional standards SAI
inspected the site in f)~c~1b-2r 1979 under an EPA contract An missions
inventory was published under that work (Ztskind 1980) and the bottom line
conclusion is that currently no significant ernissioi1 are gteing released as a
r~sult of hl-isting which would reach the publ le i 1le 0pen pit is some 900
feet deep with blasting at the bottom 0ver 80 percent of emissions in the
CARB Emission Inventory System were attributej to pit blasting In fact at
its current depth blasted 1aterial does not reach the mine surface with tile
explosive implacement used Additionally the site is more remote than might
be realized The ~ituation is vastly different today than in the past and
currently attention should be paid to the issues of occupational exposure and
breakdown of ventilation syste111 controls in t1e inilling operationsmiddot He
recorrrnended that no furtw~ corsideration be Jiven to this site for the
purposes of tl1is w)JJil ie identification )f silt~middot1ificant releases vhich
might he responsible for causi1J middot1 fmiddot middotgtbullgt middot 1= 1(11 1bull11 tinn Pxposure
8-2
Vctri0J5 Refineries
It was est=tblished early in the anrilysis prmiddotJr111 V1at gtenzene
emissions from refinery operations coulrl in th~ 1guregate constitute a
significrnt source Since there ilre a large rrnrnber 0f refineri~s in
California (46 in Los Angeles County itself -it t(~ i 1~ middot)f r-1ltariination) with a
wide variety of types sizes ages etc their eva l uat 11) 1 i =r~sent -3~ i)7
monumental task It was noted immediately t 13t ~~it hi n the total scope of the
study the design and conduct )f c r~f i r1~ y testing program which would develop
a complete benzene emissions in1~ntory for 01e site was impractical let alone
to d1aracteri ze emissions fro111 three ie those listed among the 17 rnost
s i g n if i ca n t potent i a l st a t i on a r y sou r c e em it 1-_ l rs bull
The three refineries were singled out for special attention in the
previol1s phase ~ecause they uriiquely had components which process materials
containing 10 or more per~nt hy middot1-bulli1 11t )-~112~ne (46 FR 1165 Jan 5 1981 pg
1491 ) Est i mates by E P A ( F e d bull r l Rr-~ g i s t ~ r 1q81 ) i n d i cate that 90 p e r cent o r
more of the total benzene fu 1Jitive emissions arise from such components
Therefore attention is appropriately focused on the three henzene production
consumption refineries Chevron (Richmond) Arco (Crson) and Chevron (El
Segundo) SAI staff considered the possibliJ )f -1 testing program at one of
these refineries and cone l url 1middotd it would not 1)e cost-effect i v2 for o number of
reasons
California is a minor producer and consumer of benzene with
approximately 15 of the 114 hill ion pounds produced and
consumed nation1lly in 1977 This can 1)e riY1pared with the fact
that California has approximately 10 of th~ ~ations population
Benzene exposur~ to the generil nJmiddot1lation has been partitioned
among the various sources Aproximately 90 of exposure is
estimated (SRI 1978) to be caused hy gasoline distribution
activities and 1ehiclEmiddot emissions in 1 ir~an areas with nearly all
of the balance )Y benene hrndl ing operations (refineries and
chemical plants) In the case middotlf California this percent1g~ Jill
be even JT1ore di middotpruJor-tionate b~cause of its greater s1are of the
population and ~esser share of the henzene ha~dling Furthermore
since the three refin 1bullry cites ire heavily urbanized no new rural
population cegmmiddotnt is beinq exposed
B-3
Californias most heavily urbanized Air Quality Management
Di st r i cts have a l ready adopted t en zen e f ugit i v e em i s s i on
staridards comparable and vith S( 1ine features VJr stringent than
the proposed national (bullmission lttandarrl (4f-i FR 1165 Jan 5
1981) Thus the concmiddotlusion rleri1ed above (Le refinery emissions
are secondary cause of exposure to the urb~n microopu~ation) is
further reinforced since H is irojected that re1eases from
components in benzene service (gt10 benzene) will be reduced by
73 by the proposed F~deral Standards T~~ ~istrict rules (eg
South Coast Air Quality Management District 456 and 4fi6l) are
not restricted to berz~ne per senor to only components in
benzene service The rjLs 1lso i-lclude flanges in addition to
the components ca i led uut in the propose nat i 01a l standard
The EPA estimated that if the proposed emission standard were adopted the
1~aximum annual benzene concertr0Vioi1 -~01middot 1 plant would be 36 ppb at a
distance of 0 1 kilomeL~~ avay Cor1paring this ground level concentration
with the general urban background i1 California of 19 ppb shows the latt2r t1
dominate In recognition of the secondary im0ortance of these thr~~ i~~s i~
was decided to utilize program resources to d~velop field data at other sites
1111here little emissions informati1)i1 1as cll-1aila1)le
8-4
APPENlJIX C
US tn v 1 ronrnen ta I Protection Agency
middotJc k
METHOD 108 - UETERMINATION OF PA~TICULATE ANlJ GASEOUS ARSENIC EMISSION
FRUM NONFE~RUUS SMELTERS
1 Applicaoil ity and Principle
11 Apmicroli1ability This rnetnod amicropliltS to tt1e determination of
i nor~an i c d rsen i c ( s) e1n is sions trom non f errou Sm(~ I ters dnd other sources as
smicroecified in the reuulations
12 Principle Particulate and gaseous As emissions are withdrawn
isokinetically from the source and collected on a glass mat filter and in
water The collected As is ther1 analyzed by means of atomic aosorption
spectrophotometry
2 Apparatus
21 Sa111pliny Train A schematic of tt1e sampling train is shown in
fiyure 421 it is similar to tt1e Method 5 train of 40 CRF 60 Appendix A
Tne sampliny trdin consists of tt1e followiny components
Note H1i s and a11 subse~uent nferences to ottler methods refer to the Methods in 40 CFflt 60 Appendix I
3 Annual Book of AST11 Stdndarctbull Part 31 Water Atmospheric
Analysis American Society for T~sting and 1-1aterials Philadelphia PA
1974 micromicro 40-4~
C-15
APPENDIX D
~f-~__d_y_ e s f o r Pr o c e s s i ng a n d An a l y s i s of PA H i n
Co~e Oven Emissions from Kaiser Steel
Global Geochemistry Corporation
10 Samples expected
l 1 Connective tubes between section welded to oven port cover and remainder of sampling system - These are to be stainless steel tubing disconnected and capped off for each sample They will be about 10-15 long
1 2 Impingers in series first two water-filled third dry - The center tubes will be removed and wrapped in clean foil and the impingers stoppered as is for shipping One set for each sample Protect from light during shipping
20 Extraction -procedure
Sample fractions are to be covered with foil or in opaque containers with inert gas flush during storage in freezer Solvents used in extraction will be BampJ glass distilled pesticide grade low residue) Working bench area will be screened with amber plastic and light bulbs
2 l Connective tubes will be rinsed with dichloromethane by partial filling capping tilting several times and draining into a beaker This will be repeated until the washings are colorless but at least three times
22 Impinger contents will be transferred to a separatory funnel The impingers and center tubes will be rinsed inlo the funnel using a measured amount of dichloroshymethane in portions The funnel is shaken well and the extract drawn The rinsing and shaking is repeated with two more portions of solvent The extracts and washshyings are combined over Na 2 S0 4 bull
A separate portion of the water used for impingers is extracted in the same way
For each liter of water 100 ml dichloromethane will be used per extraction ie 300 ml total
Internal standard is added to the combined exshytracts at this p0int
D-1
_rmiddot
30
40
Concentration of extracts
Bulk solvent is removed on a rotary evaporator in a water bath at not over 40deg stopp~ng short of dryness The res~dual volume should be just less than the f i n a l v o l ti 11 e o wl1 i ch the s o l uti on i s to be d i l u t e d (accurately) The object is a final solution with roughly five mgml solids concentration in dichloromethane After the vol w1 e ~ s made up to the mar k i n a vol umet r i c flask an aliquot is removed and evaporated on a tared disk in a nitrogen stream and weighed to determine yields The remaining solution is stored in a ial or ampoule sealed with a Teflon-lined septum A split for SAI QA may be made at this point (see Section 43 also)
Gravity column c1eanup chromatogram
4 1 Into a 10 min 1 d column with co2rse sinter and Teflon stopcoc~ 6 g 120 mesh silica gel (activated at 120deg) 1s sluiry packed in pentane cooling the column by evaporation from a tissue wrapped around it and wet with acetone) This minimizes vapor gaps in the column A cap of 3g sodium sulshyfate is added to the top The column must not dry out before the chromatogram is complete
42 A sample of the extract concentrate is taken This is to be based on experience of expected PAH conshytents but initially would be 5-10 mg This is exchanged with pentane by adding 10 ml portions of pentane and evaporating over about 100 mg silica to small volume Three to five portions of pentane are used The final residue is rinsed onto the column with pentane
4 3 The chromatogram is developed with four solvent steps
l 25 ml pentane 2 l 0 ml 2 0 ~~ dichloromethane in pentane
II II ii II3 l 0 ml 50 u II II4 l 0 ml 5 0 ~~ methanol
The column volume is approximately 8 cc This is allowed for while collecting fractions During the first chromatogram the graduated cylinder collector is observed for visible solvent changes and a more precise estimate of the column volume
T h e f o u r f r a c t i o n ~ a r e e v a p o r a t e d i n a n 1 t r o g e n stream at not oveimiddot 40 to l 0 ml then stored under nitrogen in the freezer The principal constituents
0-2
of the fractions will be
l Aliphatics 2 Lower molecular weight PAH
11 113 Higher 11
4 Polar materials
Fractions 2 and 3 are to be analyzed by HPLC the others retained for future interest or in case of an incomplete or defective separation Splits for SAI QA may also be made from these fractions
50 HPLC Analysis
5 l The column is a reverse phase partition column C18 bonded to microparticle silica 4 x 250 mm The solvent system is a gradient from 6040 acetonitrile water to 100 acetonitrile initial slope setting l 5min The flow rate is one mlmin the injecshytion sample volume one to ten microl depending on conshycentration Detectors are UV absorbance and fluorescence in series The absorbance is set routinely at 296 nm but may be varied in certain instances to aid in identification of peaks The fluorescence filtars are narrow pass excitation at 360 nm and cut-off emission above ~10 nm Both d e t e c to rs a re rec o rd e d s i 1i ul ta n e o u s l y
52 Reference calibration is based initially on indishyvidual chromatograms of single compounds and roushytinely interspersed chromatograms of multicomponent reference mixtures The proposed calibration stanshydards are the following at minimum
Phenathrene Pyrene Chrysene Perylene Benzo(a)pyrene Benzo (ghi)pcrylene I n d e n o ( l 2 ~ c d ) p y r e n e Coronene
The internal standards (added in Section 22) are 910-Dimethylanthracene
9-Phenylanthracene 9 10-Diphenylanthracene
60 Peak Identification
Significant peaks not recognizable as present in the reference standard will be investigated by alternate methods for identification
D-3
6 bull I Stopped-flow examination of UV absorbance for maxima
by manual wavelength varmiddotiation allows checking against known reference spectra (published or measured) This is usually sufficient confirm a component already suspected to be present
62 Collection of the fraction contairino the unknown peak allows later determination o~ t~e full absorbance andor the fluorescence spectrum for comparison with reference compounds
63 If not already run on GCMS the sample can be sent to SAI for that purpose possibly allowing the comshyponent to be identified Since the HPLC elution order is not the same as the GC retention sequence it may be necessary in affibiguous cases to collect the HPLC fraction and run it separately by GCMS
The~e may be numerous minor constituents not identified The decision on how many of these to track down and identify will be made in consultation with CARB adv~sors a~d with SAI since at some point the returns ute not worth further expense
_64 IdeAtified peaks are quantified by comparing their areas to the areas of the known concentrations of reference compounds A recovery factor based on the added internal standard nearest in elution order to the sample component is then applied the amount of internal standard originally added to the sample is divided by the amount found in the HPLC analysis This ratio is the multiplier used to correct for losses in concentration and chromatography steps
70 Quality Assurance
71 Transmittal of sample ~plits to SAI provides for external QA This should be done with 15 (or one in seven) of all samples
72 Blank water extracts are carried through the same analytical procedure (10 of total)
73 The cleanup and HPLC analysis are duplicated on 15 (or one in seven) of all extracts
74 The multicomponent reference is run first every day then elution volumes and peak area for each comshyponent are plotted on a control chdrt to pinpoint developing problems At least evetmiddoty tenth chromato-gram will be this standard It wi 11 also be rerun after any service on the instrument~ such as a column change
D-4
APPENDIX E
Gas chromatograph 1mass spcmiddotctromlt~try protocols used by SAI Trace
Organic Chemistry Laborator1
Aliquots of sampl=s received from Global Geochemistry Corporation
were analyzed by combined c11s chromatography mass spectrometry (GCMS)
Liquid extracts were spiked with an internal standard compound (deuteriumshy
labeled phenanthrene) priot to analysis by gas chromatography mass spectroshy
metrymiddot A Finnigan model 4( 21 Quadrupole GCMS system including an Incas data
system was used for all analysis Liquid samples (1 microliter) were injected
into a 30-miter x 025-mm ID SE-54 fused silica open tubular 9as chromatoshy
graphy column (JampW Scientific) A programmed GC oven rate of a 4 minute hold
at 30degc followed by a 40degCmin temerature increase to 160degc and a s0 cmin
temperature increase to 270degc was used for sample analysis The detector
system was a quadrupole mass spectrometer operated in the electron ionization
mode at 70eV The quadrupole mass filter was scanned from 35 to 475 atomic
mass units in 095 seconds A hold time of 005 seconds was used to stabilize
the electronics prior tote next scan Data were continuously acquired using
a Finnigan Incas Data system which writes time intensity mass spectral data
to a magnetic disc The mass spectrometer was operated under computer control
during acquisition Figures 33-7 through 33-9 show the total ionization
chromatograms for each of the three samples that were analyzed
Following analyses the data files were searched for priority
pollutant polynuclear aromatic hydrocarbons (PNAs) In addition to the
quantitative data reduction used for priority pollutant PNAs a survey analysis
was performed on all significant GC peaks (ie signal to noise of greater
than 101) The survey analysis was based upon a computerized library search of
individual background subtracted peak maxima scans for qualifying GC peaks
The library used for qualitative identification AJas a combined NIHEPA
National Bureau of Standards library 1hich contains approximately 26000
entries
Tables 33-11 through 3-13 present the results of this survey analysis As expected a series of al~yl Slbstituted PNAs were observed in addition to heterocyc l i c compounds con ta i ni n~ nitrogen iind sulfur moieties Further
investigation of the mass spectre of the nitrogen substituted identifications
confirmed the presence of benzonitrile isoquinoline andor quinoline acridine
carbazole and alkyl substibted ~yridines Ion specific searches for N-nitroso
substituted amines were neg1tive
E-1
APPENDIX F
Detailed Fiber and Mass Concentration Data - Johns Manville
=-- -i l L - [- - =-- -= f- - =-- -l J middot - -=-
i-- L =-- - ~ ~ L ~ ~ =--shy r---- --
- 7 l -=----7- -- - - - -- =-- - - ~ =-- - 1- -~- l
g~~~~ r_-middot~ ~-i [-- L ~ L Ldeg
middot--~ l - --------=-- r- L r- r-
~~~ =-- -- r- = - -r---
a- - - - - _ - -I~=-- L -- l 7 -~--
gtZlt _
- _(_ -lt lt =- = ZJZ~
-
__ _
~ -- __ middot- -~ ~~~ = L ~f~ Z ~ C
middot -
__ ~ lt lt =- _~~ijE
gt = z lt gt
rbullshy__
F-2
SI I HMllLfc J H 1 I OBr I lll-H ANI) NS~ coHcrrrllliTJ ON CUfllllTI VImiddot SIJMMIIY
TIITIL JnI~ SfiNtlEO TOTtl Ill- or FI ITETI TOTI Illlt ~111-11 11 ltllOT FIIACT I Ori 1111 OF ~middotWI MUllllNJbull lOTI 01IIMImiddot 01- JII TOTI FI lilmiddott~ UgtIJH nD
= t nwo 00 so MI CllONH = t I 1JOO SO (M = II 900 SO CM t bull (0000 l1T~
I I JO SO CM 70000 CUil JC Ml-TEHS
= ~lt O V l lWfl~
Manville D-2 Baghouse
723 347 -633
CIIHYSOTI LE JfllII I nou J11BICUOUS NO PTTFllN NON-SBI-STOS tLL FIIIEIIS
FI 111-11 COllflT ltFIIWll-
ITllClmiddotrir r lllrf
2gt()
) ( I f S bull1 bullbull
I 0
l ~-Irbullmiddot
00
fl JI)(~
00
0 OUU~~
00
0 OflO
~60
1110 000~~
FI IWII ( ()[IClmiddotNTHT I ON ltI- I mn ll11 ~41 GI OF FI LTEH) (FIBlmiddotn~ ll-11 CIJB f-11bullTl-H OF IH)
I (2fi71bull+0G (aG(l-+05
( gt027 F+04 OfmiddotHH
OOOOOE+OO OOOOOEtOO
OOOOOF+00 o uuoul~middotH)O
0 O(OOE+OO o 0000I+oo
J bull f 1)07 E+O( fi r71HE+OG
I
w
TcfLI ~~lllllmiddotCE 1lllmiddot lt~O Ctl lFII ~~O UI OF FIiTEi) t q UI 11-11 Cll II fllmiddotTlbullll OF J I H)
I 590)1+07 6 t 90llbull-HH
( HMOE+O 2 40411bull-1 o
0 OIHHlF+OO o 00001+oo
0 OUO(I 1 00 OOOOOJi+00
() 10110~ Ji)
OOOOOE+OO 1 lHHllJ i 00
0 00001middotgt1 00
M ~~~ middotor-ic Irrrn TI or
( PGRAMS 1bull111 ~c UI OF F J LTEH) (PGRAMS 1111 UJI illI Lil ()j Ill)
IUitTII Ml-Ml ltMI tJIOfl~~) ~Tlgt lgtEV
Ml- H Lot c1middoton nr COlmiddotF I II
I) I MWTEll Ml-N ltMI c11or1~~) STI) tHV
MEtl IOC cImiddotor1 r1N CCWF V11
AIt lbullI rt 1bulli I~ (Sil MIC) STIgt 1gt1-V
Ml- N 101 c1-ori r1u COlbulltmiddot V1ll
I 0 I ~if1+07 bull tJf I (11bull-HH
r 1gta 1 1) bull 2 (10()
0 lt700 l 1Hi4~ 2G44
() ~( 1 027B
- l bull 6 I 1bull1middot 0 J l)J 09J~fgt
1a 1071 1~ 09rn
0 2(jlt)fi l ~HVil 4040fi
~ HltF+O~ U (H~61bull+o I
o rooo 0()()00
-0GJOB 0 (000 0 (000
OOfiOO 00000
- )()57
OOGOO 00000
0 0 1)02 00000
- I J 0 0 0 1W 00000
00000 0 ()000 00000 00000 00000
00000 00000 00000 0 (WHO 00000
0()000 0()()()()
0 ()()()) 00000 00000
00000 00000 0000() 00000 00000
00000 00000 00000 00000 00000
00000 00000 00000 00000 00000
00000 00000 00000 00000 00000
00000 00000 00000 00000 00000
00000 00000 0 (WOO 00000 ()0000
( ~1701 l B bullJl~H 0 lt~( I BIJI ~ ~i Ill
o~-~ 0 ~j
- I ult-fc 0 I Hbull)I o bull)abullgt~
12 (1 41 17bullgt7
0 1 l I H) 4 ltmiddotI 1
VOUJME (ClJB MIC)
MFiN ~Tl) 111-V ru-1N 10C Ct-OM MN COE- Vtll
2lHH7 gt HUB
-2 7m 0 060 1)
( IJl)()C)
0 001 00000
-ltj 74a 1gt 0001~ 00000
00000 0()000 00000 00000 00000
00000 00000 00000 00000 00000
ooono 0 1)000 00000 00000 0 ltWOO
2 7 0 1J lJ I J J
-2J+H () ()~(
B 77gtB
lt ~
lt
C
J c
~ lf
_
ii
(]j (fJ
J I 0
a1 c r--- l)-- Clltr (Y) r- ro C)(V) gt deg CN C1 (0 I ~ C M
N --- r--
w J
s rJ gt-
c
~
LJ
t-
~ - -lt-[-bull
0- ~ l L L l~ -18~~
L a- - ~ I-
-I - ---
-
-- -- -- - - ~l
-1_ - - --
--~--- [-middot-~--- t-- -- [- -_ 3J ~I r-
z g ---
l
t--[-
l l c +-shyc
~ Ci - - I
-1--r-middot~- ~
L l- L ~
L
II V
C
c
t--
-= -- - - ---- - _
middot= - middot-= l J l~ - ~ ~ L - ~ L ~ [--
II II II II II II II
_
---
--
z
- I
==== lt _ --middot --~ 7 _
- ~~- _
z ~ ~ =lt ~~~ti3
f
gt =2 lt_
2 z _ lt=lt i~iS
-- z ~ lt
~~~t3
~lt~ =--~~~sect~ _ -- -middotImiddot_
L -_ -
F-4
(lJ II) I
~ Or--M
ltlJCsr-M ~ O) bullbullbullbull ~ lCMIO bull- ci gt (= N lC I M ~ 0 N
_
J J
---r--co - - ~l r-j = ~i
~~ 2-
88 rJ--_-ltI -
lt ~lt
z
Cl J ~ _ e-l ~~
- ~ middot~ e-- l ~ l~ [- - ~ -~=----~
_ L
L
--o---
-o-~shyc L~ ~ - 0 - Ll-C d o~-
11middot II 11 11 II II II
-_ 0- _
--
-- lt== ~~ ltZ z-== -~~=E~8C~l~-=rI - lt ~ fJ
ltZ lt---=lt===~~~- _ ltltlt=gt ~~=~~~~~ -
~ ~ lt -=z~ z _ -shy middot--z =-
-middotmiddotmiddotmiddot middot-middot -----
r bull bull _
- _ - ~-middot __ _
~ _
middot- - --- middotmiddot
_
- l
~=~r -~= ==
lt =~~~ti
F-5
FIHJ-ll Nil MAS~ CONCFNTllJTlOi SJ I SMllLE I l) Al OJ CUMJIATI JIbull t-llHMHY
TOTL iHf- SCNNFD lOTiL Illmiddot OF FILTFTI TOJI Ill- JSlllmiddot~ll I I OIIOT FlltCT [ ON 1111middot OImiddot ~1-M iIHlllllJNE TO 1 1 I ()lIJMImiddot ()I t I H TOTI V 1111-11~ COUNTED
= 1080000 SQ MI~RlozNS = 11 900 SQ CM = 1 l bull )00 SO CM = 1 00000 liHlS
11 IH) so CM 2G70000 CUBIC METfiHS
1 ) fi FI JllIIS
Man vi l1 e D-2 Baghouse
723 0 347 -633
Cl Ill Y-OJ l LE MIPII I BOLF Jf1BI(UOUS NO PATTEnN NON-ARHFSTOS ALL FI BFllf~
F 11l-ll f()llrlr ( I 1IFH-- l 17 G o 00 00 00 19fi
Tl fL P1 Sl~JNNFD Tl lI ltl- OF 1bullrn~n H 11I HF ~IJlbullI) lll 1 ll(IT 1-ILCTON IL CW ~U-1 HErlllllNE
THl1 (11llrJI~ OF 1IH Tt1T-d 11111middotH COIJrfflI)
= 697~oo so MICRONS l l 1)00 so CM t I JOO t-o CM
= 1 00(00 lHT~ = I l 1)0 ~o GI
ant()Oo cun1r NETEns = 7 0 I l lWllS
Manville (upwind)
CIIIIYsOllY AMPII I nou JMfl[CUOUS NO PA1vrrHN NON-A~~HFSTOS ALL FIBFHS
v1mI1 Cttllffl l 1-- I II I fl~ 1 () () oo l 0 00 40 70
1middot1111unmiddot ToT1L f J BFllS o ooo~ ooow- t OTT~ ( 01rn 57 I 1~m too ooo~
1middot 1111--11 1or111rnnT I ON I Imiddot I 111 II~~ 1111 ~o CM OF f I LTFTl) 1 11111 1middot111 CUil middot11bullnn Illmiddot J l ll)
O OOOlJEmiddotH)O UUUUOL+OO
0 00001-1-00 0 00011 lbull-Imiddot00
G 1 I nn+OG 1 HJ llmiddot~+OG
00001+oo 0001101-1-00
r n~+H+uG l 7llHmiddotlE+Ofl
l bull l 1)1-n+O(i I J ~ 1)(11-middotl()i
T11T I ~Ull FiCImiddot 1 ll f ( ~ l lrl ll11 tn Ii OF F 1ITFll)
f ill I I I i 11 I U l ( i I Ill J
0 OOl)Ol ll() () U()0 I Imiddot()()
0 UOOOlmiddotgt1-00 O OOlgtUlbullgt100
l 0 1) ) I 1 lJ J ~ tl( 1H-HH
0 UUUOlmiddotH)I) o 00001-+oo
~ lt1Hlt1(E+OG 7(H(Ml-middotHM
0 OOOOJbullH)O 0 0000lbullgtHlO
77 I
N ~
rl gt- 10NCJfffll1T I ON
r orR AMS ITll ~q lfI (II- FIT n~ll) - LIi LIIll 111middotTU OI 1lll)(PGRAMS II i1 Ill MlmiddotN
bull I t bull I 1 f I ~ middotr1 Ill lmiddotl 1- [~ IOC ( Hl1I f r~ Ull-F ll
0 0000 Imiddot bull-110 0 0000 Ibull HO
0 ltHgtOO 000110 t) (HJ()()
0()()()()
0()000
llOOOOF+Oo 0 00001bull-100
0 ()000 OOtlOO 00000 0 (WOO 00000
0 (((7
0 ~0-i -0 middotlbullribull)bulll-
0 ltl l lt 0 oHi
00000 00000 00000 0(0HO 0000()
077[0 0 -1117
-0 ~bull)I)()
0 7middot1middot1 (i () L~bull)7
0 7~B( 0 I I
-o alt17n 0 (1)~1 0 17lt1 I
11 I 11ITlt I j i 1l 1 111bull I
f 11middotgt N gt I ii ii I V 111 ii IOC tHlfl flf1 Clll-tmiddotmiddot VAIi
00000 0()000 0000() 00000 O 0000
041000 ()0000 () 0000 o 0(1()0
0000()
0 llfl(i7 OltHJ
-~ lt-11 I 0 07 I 1 () I)lt)
0 (H)ll()
00000 00000 0 ()()()() 0 0(100
0 0 Omiddotlmiddot l
-c07 1)
0 I 1 I( 0 1101
0 l l j--
OOGli -~bullI) ( (J
0 1lt11 I 077G7
I 1 (71 rJI C)
tll N -r11 111-V rmiddotllfl trn CI llf I 111bull1 COIT rll
() ())(l(J
00000 0 ()(1()()
0000() () (1()(10
00000 00000 00000 0 (I()(()
00000
0 ~ I Ifdeggt 0 I Imiddot~ I
- I H 1)7
O I GOO I I i ) l)
0 0001) 00000 00000 00000 00000
0 1 1)17 o~ViH
-1 0 1)gtG 01lbullH 0 7 I till
0 l l lt G 0 ~r1
- J bull J1 1) I 0 17 I bull at-lt
lllllrmiddot lt Clll ii IC)
~I Imiddotmiddot N ~middot1 D 1)1bullI f-1 Imiddotgt f IOC ClltHI rm uw-- i ll
O(lOOO 00000 00000 () ()(J()(J
(l(t(()()
0()()00 00000 00000 (1(1(1(()
00000
000fi7 OOOGO
-G Jll7l 0011G 1 H20
00000 00000 00000 000()() 00000
0 0 I bulllCi 001w
-4 iHd 00101 I ~700
OOIOH 0 0 I Ilt
-(1 j l)~G
000(i 1tHH
--
--
L c~
U c t c
(
J
J -
C
C
c g 0 0
+ + e--c~ =~ - r~ C- l~
--
+ +---
- -C 0 o- -c--1
I
lt t w
lt
C
ta J ~ ltCi
lt ~
a---~ -- c gt C 0 rtl 5 -
VJ C f-rJJ c
[J_
ltI ~ C z
z _ ~
~ lte 0 ~
C
tJ
e
C
~ c C
L N
~~ C 0
C
l~ 1 0
t CI ~ ----~ -
i-
cc cc + + t=~= g~ --
-~ C + + t~ ---[w (~ lC
-
ccc ~t c5 ~ ee 00000
co--oc I
- - - - -_____
------ _ - _ _--__ ---__ ---- _ - _
ooocc
rr ~ c-~ E lt
C
-
~ C g L-
gt o + + ~~ ( L C- lN o -l
t l - -I
~ - L~ [ti e- [
l _ 1
- - ~_ C
gg -- -- ~ -0000-tr--- ____
----C __
- __ - _ --ggggg
- -
C
+ + -- --- -~_shy-----_ _ -----___ ooo~
-----__ ___ _ - - - _ - _ _- - - - ___ 0~000
--__ ---_ __ -- --c5 ooco
C cc
1 II II 11 11 II II
z
-
r ~ c
z
~ t lt i C z __ -- -~tt ~ middot~ ~
-middot middotmiddotmiddot _ - -----=___
_ =~ - _ -=shy ~
-- -~== 2~= ~ () V)
_middot ~ c( c(
0 er ~- ~ ~ lt 0 0___
-
- amp lt~ ~~~t8
F-21
) ) ) )
TOT1L AllE SCiNrmn TIIT1 L ll lbull1 tW FI 1lFll TIIL1 11bulli 1~lllbullll lllIIOT lIIJClION -11- OV TUI Mlbullrllll1Nft TOT1 I VOIIHIIbull 01- I Ill TOTmiddot I Imiddot I I 11-11~~ COi l NTlbulln
_LI__l_ ___l 4 bull P 7 bull P A R l n A 4 R 1 n 4 4 ___ 7 A ___ _RA____ 1 A 1A 8 1 SL__ 1 R 5 1 ~ 3 _1] ________ 111~____JJ__) 1 ~---~-~ 4 9 1 ----1___11
A0 1A4 JR 17 lAR A 41 4C 174 A ll~ JnR ll7 9 5 R12
----- --------- - ---- ---- -- middot-middot --- ---- -- ----middot-- ----middot ----~ 7 c ~ bull n 1 1 2 l 9 -s 1 -i s s ~ c c__ bull A____02 1 1 2 1 r n 1 c 0 r-- l ~ 1 o 1 0 1 t 1 o c- ----~--- -Pl Ol OP lA2 11A 61 c 0 ~ cq03 l 5 4 l 4 4 1 i r 7 -i r) c q 1 deg 1n
qA RO 107 ~_12_7 SA c 7 A04 _l _ s_ _________l 7 l RA____ 7 ______l _o 4 __ l nn_________ 0 _7________ R_l __ o ~ _____ J(__7 11__r ___ _t-__ o ____ ln ___ h_ ______ An____ c-_n __05 ]lA 140 170 7cA __2n1 qc- 7q 1n1
c 2c 7 lR 1 O 406 11 c A 0 4 n 3 4 7 l 4 R ~ 7 7 5
___ ___~ c Z bull A ~R_ ---2_h 2 bull bull 4 e R bull C07 l_ H l_ ri ------- R1 --- c ----------- AQ _middotbull 4_p____ r 7 ___1[_2_ _r--_______ _ A____________ )_3-____ _____ __t 9 ______ ~_____ _3 __ --~___()_______ __I__08 C7 lR9 9A 1n~ ll7 S AS 84 ~Q 33 11 11 1n R 3A 3309 7RJ 1R lA 14-1 17fI 77 4R 7(_
---~- L~ 1 7 -~ 1 7 4 1 l 1 9 S 3 9 C) 4 9 A ~ 9 q10 __ 7 c _ ___ _ 2 RA l A ____J__6_R 1 C 4 7 P C 1 rq I
-----~~a-R___-=4bull2_1 e __ middot-middot middot bulls __ - __ IL 9 1 ~_1_ ___3 7 40__19 1 ~~- gt q6_ (- l Q 144 43 _37 _ A f ---middot ~ bull r bull q ~ 9 IL ~ 1 _0 -~ bull 1 A 3420 l r -- I- I q4 l qo l c c 0
Valve-Reactor Heat Exchanger 200 009 2222 I I I 285 I 93 52
All measurement are for 100 CT streams
1--- w 0
air smicroace Saturation concentratiun at 20degc is amicroproxiriately 12 Measured
values were bY and 74 These values r~sulted from gas chromatographic
dn a I y s e s ot t rd n s t er re c1 sa111 p I e s fr() m t n e 1U U I i t e r Ted l a r b a y t i 11 e i nt e9r a t ed
sample ror tt1e tank car displdcement of 2008 ft 3 the total CT emitted based
on tile ctverctye is LUU8 x 007L = 144 ft 3bull
Tnen for a density of p =PMRT
P = (1)(1gt4) 64 gL = 04lbft 3
(082)(293)
Tl1erefore tne displacement weiyl1t of CT microer ()ff-load is equal to 144 ftJ x U4
l b ft J = ~ 7 bull 6 1 b bull F o r 2 5 U off - l o ad i n gs an n u c1 1I y we have 14 400 l b s bull Th i s
compares with UuPont I s rnectsure1nents of 9 3 vapor content arid 18800 lbyr
emissions Saturation vapor pressure concentration emission would yield
2400Ulo
l)reathing loss from the tank can be calculated from AP-42 as
Lt3(1bday) = 619 X 10-5 M( p ) 068 D 173 H 051 T 05F CKcmicro 147-P
5 0 68 173 = (619 X 10- )(154( 173 ) bull (155)
147-i73
(lj) 051 (26) 05 (1)(1)(1)
= S 63 lbday
3where a working range of zuu x 10 o 400 X 103 lbs were used to determine an
average l i quid nei grit Jn nu a 11 y e111 i s s ions would be 2057 lb Therefore
working losses dorrii nate tt1e totil elissions from tne microlant
Summary
The upper and lovJer bounos of caroori tetracriloride emission ~vere
determined as Ll467 lbyear (upper bound on fugitives DuPont measurement of
storage tank work i ny erni ss ions nur 1a I tanK oreattli ng) and lbgt 15 loyear
(lower bound on fuyitives s1 111eas--1re111ent of storage tank working emissions
no ri n a I tan k breatt1 i ny ) bull
13]
100
SOUfCE TESTS - STAUFFER CHEMICAL COMPANY
101 SITE OVtRVItw
Summary
Stauffer is located in the Carson area of Los Angeles County It is
a manufacturer of polyvinyl chloride from tl1e monomer (VCM) ll11hich is produced
on sit2 Stauffer is the only California producer of ethylene dichloride
(EUC) which is converted to VCM The VCM is regulated by several standards
including Cal OSHA EPA (emission standard) and CARB (ambient)
During Phase I the plant was inspected and possible emission
sources were identified Th~se cans~sted of EDC storage ta11ks fugitive
emissions from valves flanges and pumps process vat2r content gas
incinerator effluents and th~ loading of outbound tankers
Since the completion of Phase I several events have transpired which
a ffectebullj the test program to deterni ne foe i 1i ty emissions These wer2
~ the completion of a vent ~ldS incineration system tied to all significant EDC storage tdnk breathers
$ the completion of a comprehensive study by SAi staff to develop a nationwide material balance for EOC
bull inactivity in EOC importation for the plant and elimination of EDC exportation from the plant
Based 11pon this input and the plant inspection of February 3 1981
it was expected that a plant emission factor for EDC can be determined with
relatively little uncertainty It is further expected that si 1Jnificant
atmospheric release of EOC du~ to plant operation may not occur at the plant
itself but rather offsite These vmuld drise fro11 two sources - the process
water discharge from the plant and severcil off site EDC storaye tanks
Facilities Description
Ethylene dichloride is being produced at Stauffer by two processes
direct chlorination and oxychlorindtion In the former EDC is produced by
d i rec t ch I o r i nd t i on uf et 11y l t~ n e i n the pr 1~senc e of cHl Fe C 1 J crl t d lys t I n th e
latter process EUC is microroduced by the oxychl)rindtion of ethylene with
hydroyen chloride and oxyuen in tne prescrlCe of a cdtdlyst typically Cuc1 bull2
132
lhe microldrrt fluriti1~s tuC after tne microrimar y reactions by separation and digt ti l I dt i ur1 fgtr-ocJuLL 1 s ctured ctncl umiddot 2d ctS foed for VCM production All
sourclS agre~ tl1dt c1ission of EDC is (f concern in its productionstorage
process stages ctnJ O little importance in VCM production steps (J~B 1980)
At the timt of plant inspection some storage of EOC existed at three
ledsed storage tanks located at tne Port of Los lngeles 1-iaterial has not
been 1-Jithdravm from these tanks during the 1dst year and there had been
discussion to consolidate materictl into a single tank
All microrocess components handling EDC storage are now tied to a closed
ventilation incineration syste11 It is expected that virtually all
chlorinated hydrocarbons includiny EDC will be etfectively destroyed since the
system must demonstrdte VCM concentrations are reduced below 1 ppm Firebox
tempera tu re is aoodegF and residence ti me greater than 1 5 seconds under
heaviest flow conditions Note that by way of comparison combustion perforshy
mance datct for PC8s are 99995 destruction at 1832degF and 1 second residence
time and 9Y9Y9YY4 at 2 seconds dnd 2372degF (N Flynn SAi Personal
Communication)
Fugitive emissions from valves and flanges are monitored by the
plant according to the requirements of SCAQMD Rule 4661 Pumps and
compressor follow ~ule 466 Emission and control requirements for vinyl
chloride are specified by ~ules 1005 and 10051 EOC concentrations in
wastewater are monitored several times daily in tne primary EDC steam stripper
stream and once daily in the composite plant outflow stream Daily water
disct1arge limits are 2~ ppm wit11 typical montt1ly averages being in the 8 ppm
vicinity The discharge limit is ~mbodied in the discharge permit (number
5061) vJith tt1e Los Angeles County ~anitation District
Tnere are a number of points at which tt1e rnc can be released to the
dtmospnere and they are similar for both dir~ct chlorination and oxychlorinashy
tion These include the following along with their emissions factors for direct chlorination
A chlorinator vent 2xliJ- 5 IOdSS per unit mass EDC produced
I) I ight end column vent -52x lt)
C dist i I I at i un column vent lXlU-5
u storage tank breatning 7xlu- 5
133
E storaye tankworkiny loss 27x 0-u 5F fugitiv~ emission bxll -
-JLi ildSteJdter l gtX] U
Em~ssion factors are si~~lJ fer oxychlorination processes
Enlission factor vau-5 (~1rbulle Tdilt2~ frcn tne literature (JR13 198U) and
assume 9b iciency for incineration (A-E) It is also assumed above that
EIJC emissions to water are Lf~ uf ~rihsion to citir ard that in wastewater
treatment 100 of EOC discharged tc1 vater is rel easl~d to a i 1middot ~ These emission
factors 1bull1e1middote used to prioritize r-2ieaies iJut were cldrly uude approximations
to tre planL l)ecause of thlt ncine-xt~on system it was anticipated that
fdctors A-E would be ri2dtKcdft 1-middotactor - 1nii~t1t be reduced since a monitoring
program had been in force al20st o~e y2ar Factor G and the off site storage
tanks~ vni~cn are not tied into lt1n ircinerator system rniyht -iominate emissions
Emissmiddotion limits and concomitant regulations embodied in Rule l00S of
the SC-l)MU have necessitctted tlw incinera-cion system Ninety gas chronatograph
probes are located throughout t11e plant including the stack of the primary
incinerator Con cent rations of vc1v1 are reported es sent id I I y be I ow tt1e r~yu la -
tory limit of 10 ppm and in fact below the limit of detection somewhat less
tl1an 01 micropra
lUl MEASUK~MENT APPROACH
Tne objective of tt1e proyram is to determine a pl ant emissions for
EOC It is not accemicrotabl~ to dtmiddotvelop sucll informatmiddot1on basect upon published
indlllstry Jide esti11dtes of plant control efficiencivs Fortunately it was
possible to desiyn a monitoring and calculational progra1n to determine EUC
emissions for tne site and not ctbsorb a disproportionate share of program
resources
Hie re are tnre~ modes ot re I ease from tt12 micro Iant dnd d fourt11
offsite
bull Post-Process 1nci11er-dtion
Processes A-E of ecc ion lll 1 dre a 11 1ented into the pl ant
incinerdtio11 syste11 rt1e conceritrcttion of VCfmiddotl contir1uously iectsured in till=
stctck cts SJdS output remicroresents d rectsondble UjJper li11it to diJply tor EDC
con cent rat i ons s i nce i ts ef f i c i ency of i nc i nera t i on 1s d t 1 east c l1 u a I to tr1 at
134
ot VCibulll Tlerefore knowledge of the system dirflow rnd VCM concentrcttion are
tt1e necessary and sufficient conditions for determining the bounds of EDC
reledse fhe pldnt expn~sse I wi 11 inyne~)s to provide tnese data in order to
support the calculations
bull Fugitive Emisions - Valves Flanges Seals and Other Sources
Fugitive emission from the valves pumps and flanges can be
determined more precisely tnan was anticipated since Stauffer has completed a
comprehensive leak inventory of all such components in compliance with Rule
466 466 1 and 1005 The Inventory delineates all leaks uncovered by their
three man crew throughout tne year and indicates the screening level in ppmv
and component identity In consultation with Stauffer we will be able to
identify the total number of components associated with EDC handling systems
( 700) the distribution ot substance composition streams the distribution
and incidence of leaks by hardware component type and leak rate We will
utilize this information to provide historical data to compliment our Foxboro
OVA sampling at the site dased upon this monitoring we predict an EDC mass
emission rate for the fugitive releases The plant has agreed to provide the
necessary information We propose to meet witt plant personnel prior to the
start of monitoring and finlize the sdmpliny strategy to accomplish 100
coverage of the lines of hiuhest EDC composition The sampling approach and
calculation of mass emission are described in Section 72
bull ~~as tewater
From examination of the emission factors derived from published
literature (see Section 10~) it is clear that EDC release from wastewater to
air is potentially several 11rders of magnitude higher than any other plant
source l1ased upon microl ant ml~a sured concentrcJ ti ons of 8 ppm EDC in water a more
realistic emission factor wgtuld b 24xlo-4 1-iassunit mass EOC produced
Clearly this could still be the dominant source Furthermore the release
point would be expected to middotgte locctted between the plant and the sanitary
district treatment site whi1 his~ kmfrom the plant at 24501 S Figueroa We
believe it is necessary to ndemicroendently confirm the average 24 hour EDC
concentration in the dischar-qe valer by obt1ininy the refrigerated composite
sample We t1ave identified ttie I 1nes of int~rest and received agrt~ement to
sample and analyze for EDC in the stream ~e propose to draw a duplicate
sample from the compositor nd ani1lyze for its rnc content This will be
135
compared with Stauffers para11eI tnalysis It is not reliable to obtain
direct readings of EDC by survey instrument in the air above the water flow since concentrations at crny single point will be i 1 the near dmbient range
Emissions will be cc1lcJLtecl 1_sing the plants vulumetric daily flm
and ass~m~ng that complete degass40~ of the effluent will eventually occur
This assumption is based umicroon the relative volatility uf EDC and tne distances
involved However it should be noted that no direct experimental or
monitoring data is available with which to confirn this Th2 composition of
the discharge stream is unknomiddotwn since it merges into a larqe multisource flow
Conversation with the LA County Sanitation l)istrict (J f1ilne) reveals the
stream to be both exposed and covered Vents exist where me measurements
could De made to assess gross leaka~e at key points
We will obtain samp1es of several plant process discharge streams in
40 ml bottles witn no head space Transit time fr0m sample collection to
analysis will be minimized and wi l I not exceed 24 nours This time frame is
conservat i v e al though ED C i s a v o l ct ti l e mater i al ar1 ct vJi 11 undergo
concentration degradation and outgdssing lnalysi) will b(~ performed in tt1e
SAI Trace Environmental Chemistry Laborcttory utilizing pur 1Je and trap analysis
FID gas chromatography A trial a11alysmiddotis was conducted and the EOC
characteristic peak was distinctive down to the ppb level Therefore the
analysis should easily confirm concentrations in t1e 8 ppm range
Offsite Storage ranks
Three EOC leased storage tanks are located offsite at the Port
of Los Angeles and are owned by annther firm Annual emissions from the tanks
were very roughly estimated based un AP-42 JRB 1980) as 44 kkgyear
based on preliminary estimates of stored quantites
Considering tne magnitude of this source these storage tanks must
be investigated We will gather information about their configuration and
control i n order to ca I cu l ate their e111 i s s i on s bull
136
lU3 UEfERM I NAmiddot 10~~ OF MISS IONS
1031 Incinerator
Vinyl chl 11ride co11centration is monitored dt approximately 90
locations throughoul the pl trlt (Lanyner 19Bl) including the incinerator
output Conc~ntrations are reported by Stauffer dS less than 0l ppm at a
flow rate between 10000 ant 12000 SCFM Although no direct monitoring of
EDC is conducted it is posible to conservatively bound the concentration of
EDC at U 1 pp111 Tl1at conce 1ltrat ion is a suitable ci10i ce s i nee it represents a
conservative bound on tile VCM levels measured near the stack output
Furthermore the molar volume will be taken as 224 L rather than the higher
value it would nave because of the sl i~thtly elevated temperature at the
detector location
Using 12000 SCFM one has thE annual emission of EDC as 12 x 103
SCFM x 283 LSCF x 526x105 minyr x lmole x 97gmole x 10-7vv 224L
Therefore incinerator emissions of EOG are thought to be bounded by 773 kg or
170 1byear
10J2 Fugitive Emissions
Seven hundred (7UlJ) sources were surveyed comprising nearly 100 of
EDC service All accessibl~ plant areas with streams containing greater than
1 EUC were screened except for a small number of devices located in areas
11lere active maintenance was being condJCted It is believed unnecessary to
p2rf-gtrm any emission factor adjustment since it is estimated that gredter than
95 of the requisite components were screened
Three leaks above an arbitrary OVA reading threshold of 20 ppm were
parameters and mass emission numbers fhe upper bound leak rate was
determined to be close to twice the nominal leak rate which accounted for the
response fltlctor of amicroproximamiddotely 05 by Radirn for TLV detection of EDC
( 13 rown 1980) bull
Stauffer found and reported approximately 10 leaks in the EDC
service during 9 months previous to the pLrnt testing Assuming no111inal leak
137
) )
Table 103-1
STAUFFER CHEMICAL MASS EMISSIONS PREDICTION FROM OVA SCREENING VALUES
OVA SAI Actual Upper Device Response Response Cone Source Leak Rate Bound Location (ppm) Rae tor (ppm) a b Se I Re Function lbyr lbyr
Gate Valve Unit P1559A
Valve Heavy Ends Column Unit 14439
Compressor Seal EDC Storage Unit Tl062
I-- w co
300 11 273 -035 096 0 17 153 D 6
11 I500 114 439 II 241 C 1
r III II II o-is lUI bull2000 vl 1 1818 A 9~
All emissions 100 EDC (12-Dichloroethane)
11
values in the nei~hborhood of 2000 lbs~ total could be emitted annually
Therefore it appears ttiat major reductions in the mass emissions were achieved
by the company run inspection prugram and the fugitive emission source has now
become of seconqary importance as a fraction of total plant emissions
lUJJ Wastewater Discharge
Wastewater samples were collected in duplicate from four sites
within the plant for determination of ethylene dichloride (EDC) concentration
The samples were collected in EPP standard 40 ml VOA vials on July 1 1981
Analyses were performed using standard purge and trap techniques coupled with flame ionization detection gas chromatography The results are given in the
table below
Sample description EDC conc~ntration range (microJml)
EDC concentration average ( 1-19ml)
Approximate fl ow ( gpm)
PVC Interceptor Box 82 - 108 95 200
EDC Stripper number Chlorination area C1404
2
011 - 014 012 25
Final collection site for pH adjustmentP663 349 - 3B2 366 350
Final discharge site sanitary sewer 6 - 254 158 500
The water in the PVC lnterc1~ptor ~ox was warm and represents washings from the
PVC reactor which travels t 1J the Interceptor Box in a concrete drainage ditch
Water from the EDC ctripper was vPry t10t and because of its heat was difficult to collect The ~1eat ot trie wat11 r may in pdrt account for tne
relatively low concentratio11 of ElC here as the EDC would out-gas from the hot
~vater more readily The fir1ctl collection site tor pH adjustment is a large
concrete container middotvith mixers in it located just prior to the final discharge
site Water at the final dischdrg~ site is being constantly aerated due to
the speed that it flows thru~yn the concrete drainage trough This aeration
could account for tne relatively large range in the rnc conotent found here
The average EDC concentraticmiddot1 at tne final discharge site is middotbelow the daily
discharye requirement of 25 19ml EDC
139
Plant oersonnel indicdted monthly iverage readings are typically on
the order of 8 ppm but daily averages can re 1ch greater than 20 microgml In
addition LA County Sanitation Uistrict staff (J Milne 1981) indicated that
an on-site impoundment pond can become laden with EOC during certain abnormal
operating periods and discharge Vdriances arl requested by Stauffer This
mi 9t1t Jccur on the order of O1c0 aCii ecH c1 1d t1erefore is no-c expected to
signifcantly impact averag~ cnnJJI discharg ~ values lt s not expected at
this t1ime tEit evaporative eriss uns from this pond are significant except
infrequently during upset conditions and spi 11 control operations Program
staff ~~ 1er2 una111Jare of any periods ~h 1~n EDC cimtent in the ponds could be
appreciable and therefore no sampling was performed
Utilizing the average uf the two final discharge concentration
1readinJs (158 micrograrnsm~ 1 and 1G 1Ji gpm fl)W one has 34600 lbyear of EDC
released into the sanitary sEwer For the pJrposes of calculating population
exposures from plant releases it will be assJmed that the EDC is locally
emitted from the wastewater streams Tl1ere ire no monitoring data available
with which to develop a more accurate release profile
However emissions of me fron the plant wastewater discharge can be
calculated according to the method of Mcmiddotckay (197S) as modified by Dilling
(1977) It should be noted that this mLst be considered an estimate since
conditions of fl ow and the presence of other substances will influence
emission rates
Using Uilling (1977) for non~erated flow first recalculate
Henry 1 s law constant (dimensionless-my of chmiddot~mical per liter of air divided by
mg of chemical per liter of water) dS
1604 P M 1 Wl
TS l
)
wnere
H Henrys law constant dimensionlessl
Pi tile compounds pur1~ componen- vapor pressure in mm Hg at T
Mwi the 11olecular weiglt
T tile absol ~te temperature of he wastewater in K
Si tt1e compounds ~uluhility in myliter at T
Then for UJC
(l6u4)(7~)(9Y)H = = U0447 with 1
(2lJ4)(8690)
LIii bullul1Jl)i I 1 1y 1 middot UL i11 wt1Ler yiven -1L L0degC is 8u4JU UHI tile vapl)r pressure
1 4 psi ( 7 2mm) at 7 0 degF bull
Tl1e overall liquid mass-transfer coefficient Kil is given by Dilling (1975)
as
(2211)(06) in mhr
-f-0-4~ (M _) 12+ 100 Wl-H-
1
=U108111tH
Then from Mackay the percent desorption is given by
c 1 = exp (-Kil tL) where
c 0
c = the concentrd ti on at time t of EOC 1
co = tile initial concentration
L = the Ii quid depth (Ill)
t = tt1e retention time (in hr) of the liquid in the wastewater
system
Retention time is bc1s~d 11pon a flow velocity range of 3 ftsec as estimated by
the Los Angeles County Sdnitatii)middot1 Dibulltrict (1J Milne) over a middotdistance of
ctmicroprox i 1nate I y S km to the Joi n_t Wdter Po 11 ut ion Contra l Pl ant Sewer fl ow
depth throughout the entire route have been estimated by J Milne as typically
1 foot 030m) to the Davison Pump Plant and 3 feet (09m) to the Joint
Tredtment Plant distances assumed to bt~ 1 mile and 2 miles respectively
Trdnsit tiine l)eco111es between 05 tnd LU hours sequentially Then computing
the net emission reduction as the product of each leg
Therefore 26 of the EOC i erni t ted bdween the pl ant and the Joint Water
Pollution Control Plant Rtgtsfdence ti111e and conditions at the plant account
141
for additional releases and residudl content is forther emitted between the
plant cnd the ocean discharge poiiit ~-Herever the flow is r2tained aerated
or shallow emissions will be accentuated
Offsite Storage Tank Emissions
There are three storage tanks leased by Stauffer for EDC storage
1ocated at 22nd and Gaffey Sts ~ bull San Pedro These tan ks a ~e used for long
term storage rather than providing feed on a routine basis Therefore
breathing loss rather than v1orkir 1J lass is )f concern An tanks are white
two being 67 ft in diameter while t~2 third is 57 ft All are 40 ft 3 inches
high and have capacities of e~thi l0~10000 (2) or 840000 gallons The
tanks are cone roof type and are not tied into vapor recovery systems The
South Coast Air Quality r1dna~J~ment u~ str4 ct has based their estimate of
emissions on the specification of 12 foot vapor level Using the AP-42
formula for breathing loss with the vapor pressure of EDC at 20degc and an
average diurnal temperature variation of 26degF one hds for each of the two
larger tanks
l) 173 H bull51 T bull 5 LlI = (619 x 10-S) M( _P_ tH F CKcp 147-P
68 173 51 5 = (619 X 10-S) (9119~ 116 bull (67) ( I 5) (26) (1) (1)
14 7-1 16)
Thus for the two larger tanks LB= 472 lbday
For the third tank 0=57 and = 178 lbdayL8 The total emissions become 23724 lbyear ~ince he plant is in the process
of shutdown it is not clear what the liquid levels currently are Note that
if the material in the three tanks were combined into one totamiddot1 -~1nissions
would be reduced markedly Exposure to a population from this site was
determined separately since it is lbullgtcated over six miles fron tl1e plant
142
110
ASSESSMENT OF PUPULATlUN EXPUSUR~
111 OVEKVIEW
A major program goal wa~ to compare emissions from the various
sources and identify and rank any 11 tlot spots in California where the general
population was exposed to elevated concentrations of carcinogens A simple
Gaussian dispersion model was therefore used to obtain order-of-magnitude estimates of exposure of the genermiddotal population surrounding each source
Since this was essentially a screening study use of more sophisticated models
was not appropriate
It cannot be emphcsized too stronyly tnat most California residents
are exposed to emissions of hazardous substances from a variety of natural and
rnanmade sources Urban dwellers dre typically exposed to greater concentrashy
tions than rural residents however all are subjected to so called 11 background 11 levels from multiple sources In order to place stationary
source exposures in perspective the typical ambient levels of each substance
were identified from the literature and compdred with the concentrations due
to the emissions from each ~lant Exposures were thus expressed both as
absolute quantities and as increments above background
Comparison of plants presents a further difficulty in that various
substances are being consict~red No attempt was made to evaluate the relative
importance of exposure to two different substances such as chloroform versus carbon tetrachloride other than by ambient concentration
112 DATA SOURCES
1121 Meteorologicdl Llata
The dispersion model to be descritgted in Section 113 required input
of annual average wind speed and frequency of occurrence of wind from each
compass direction These data were obtained for most of the sites from the
South Codst Air wuality Management District (SCAQMD) In other cases (Kaiser
Jotms-Manvi l le l)ow and OuPlt)nt) only clai ly cJVerage wind speed and frequency
data were available In al I cases we used data from the meteorological
station nearest the modeled e11issrnn source Table 112-1 summarizes the
141
Table 112-1 METEOROLOGICAL DATA BASE
Observation Site Data Period of Pldnt (Distdnce Source Observation Remarks
from Plant)
A11 i ed Che11 i Ca 1 Lennox SCA~MD 1Y71-1974 Averaged hourly readings available ~ 1 Segundo (3 111iles)
S ff~r-- Uemical Long l3each SCA()MD 1962-1974 Averaged hourmiddoty leadir19s avai1able offsite Carson (3 miles) storage tanks appruxi1ntL~ly 5 miies from
Figure 114-3 Cumulative Population Exposure to Carcinogenic Polycyclic Aromatic Hydrocarbons From Coke Ovens at Kaiser Steel Corporation Steel Mill Fontana
CV)
0 -I X-V) V) a _J
s 0
C 0 micro co s micro C a u
l ~
14
1 --
J- t_bulli
_ C
uw m 0
1 -0 a bullbull micro
microdeg ()
0 micro tbullmiddot C QJ V)
0
X LLJ
C
middot4middot~middot C 0 micro rt1
-- - 0 ~ 0
0
n __ _______bull~ -______ --al_ - ~_
I bull
I I - bull
middot 1
-~1
l i l ~l
~ 0 I 1
Li bull I t_~I
i bull -1 _1-J
0fmiddotbullmiddot-middot+-middott-t-r+-t-+-+middotr+-t--+-1--t+ t f-t-J-t--r-+tmiddotmiddot~~bull+-t--1~~bull-+-1 Id middot t 4 E1 (
I ncrementa 1 Concentration (~gm3)
Figure 11 4-4 Cumulative Population Exposure to Asbestos From Johns-Manville Plant Stockton
I
)
--~ ~w t _ ---~- ----------~ --____~-=~ M
~
bull I bull I I I I ii i Q t I i E e- I Ii tl I I I 6 I a I I J I f t t bull I G fl I
a I e 1 1 a 1 1 1 1 iii bull bull 1 1 bull 1 1 bull 1 1 bull bull bull bull 1 1i bull bull 1 1 bull a bull 1 1 r ~ middot u ~ ~ I0
+H~ ~ ~~~~-~~~~-rmiddot-~~~~~~~~~-~--~~-~~-~t~~~~~-~J C
~1_ c bull 1 t~ bullbull bullabmiddot 11 l bullmiddot II bullbull C- I u~
0 rl
gtlt I I I- 20 () ()
CJ -1 ~_J
J Cbull s
I If I I I I0
C bull I Ii I I I I I I I I0 1E
bull-+l I t I I ro s Lower+l
I I I G I iii I bull
() 14 I I IC
u I0 I I II Iu
C
1-_~I fi
CJ ~ I-- (j) CJ I G I I p +l
ro +-l I I I1~3~V)
0 +) UDf)Pr u I I 11 I I
CJ 1iII
Vi I 0 I I It- I a X
0 ~
w I I I I bull
Lower C 0
middotr-
- 4+)
(1j
J Q_
0 bull1
ij a 3- ti I
D d I I I I
I I I bull
II B I t
i ] i amp I
1 ei ~ amp 3 ii
I I I I
e ft a a I bull G
I I I bull I
I I bull I I t If
UpPrf
I I
I I I-_L~ lI I I f
I I I I I
- I I
j I I II I -
lI ImiddotbullII
l I
l Li
Incremental Concentration gm3)
Figure 114-5 Cumulative Population Exposure to Perchloroethylene and Carbon Tetrachloride From Dow Pittsburg (Solid Lines) and to Carbon Tetrachloride From DuPont Antioch (Xs) Results For Upper Lower Bounds on Emission Estimates Are Shown
1i-1 - middotbullth~middot t =lttF--i-+-middotr+-r+middot~middoti-middot-~-4-4middoti t -~--is -~middot-rmiddotibull--+ u ~~ i--t4~ t-bull~~311 li-i1middoti- C - JJ _ CL
-- I 1 t middot1 c7 Ibull ) ~ I-middot bulli C bull 1-_ 11 11gt1l11 ~bull -a bull-bullbull t a- II bullbullbull bullI lt_I Iii _- I
Incremehtal Concentration (microgm 3)
Figure 114-7 Cumulative Population Exposure to Carbon Tetrachloride From Allied Chemical El Segundo (Curves Corresond to Upper and Lower Bounds on Emission Rate Estimate)
--- I
~__~-unt2wbull--o1WCUatJ41ia1
16 0amp (V)
0 ~
gtlt
14~11l 1l QJ _J
~
1 middot-0
c t1middotmiddotmiddot 0
micro ro
_ c micro ~ 1euro1pound1
0- QJ u-J c 0 u O =-middot ~-= QJ
c_1 micro ro +- V)
0 micro ~ t1 O QJ 1l 0 a 40gtlt w c 0
-
ro --
micro
20middot l ---a I ~ I a 0
0
pound1r-t k
Figure 114-8
8tWI-S~1111111N~-
a I __ I wi~ i awLi--~~ 2 I I
1--+ I ~--if bull =~~--bullJ-+--f-1-----middot+--+middot-middot--r9c- 1 c- J
bull _ bull -~
Incremental Concentration (microgm 3)
Cumulative Population Exposure to Carbon Tetrachloride and Chlorofonn From Allied Chemical El Segundo Assumingsix-months Operation With Each Feedstock (Curves Correspond to Upper and Lower Bounds on Emission Rate Estimate)
i u---J--bull-+ i i--+ --4--+-imf-bullbullQ~-+-+-+~bullbull-4 171 i R = Fi middot1 1bull1 r1__ bullltaa ~a
Incre0ental Concentration (microgm 3)
Figure 114-9 Cumualtive Population Exposure to Carbon Tetrachloride From Allied Chemical El Segundo During Offloading From Midnight to 6 am middot-middotmiddot------middot--middot
--bull ft ampLA 1111-~lliaIN_WPl1A 1JIU_Mt91 MJ1WWlaquoPI-W1111~ --------__----Et1
M gt~middotI
0 -I
-X
~ t~ bullM
V) middot-middot ~ V)
QJ _J
f 0
0 -
C -4 0middot~middotmicro re f micro C a u
10 Cdeg u 0 3 0middotmiddotmiddotbull~ -0 a
+-gt ro
+J V)
+-gt 0
2E1-0 QJ V)
0 a X
LLJ
C 0 1pound1--middot+-gt ro
-- Q_ 0 0
I I
~ -~middot bull I -bull-_ bull I
middot lbullgtbull___I I
frac14 llk I
I bull
-~
I
-1_ I
Ibull II__ JI
I~
(1 f---1-+ bull-r_--f bull middot~-bull +-- bull i -bull 1 -1-+--+bull-+-t-+bull--t1 ~ _ 4middot 6 E 1t1 1middot2
1 3 5 I middot~ _ 11
Incremental Concentration (ngm3)
Figure 11 4-10 Cumu1ative Population Exposure to Ethylene Dichloride From Stauffer Plant Carson
_~ ~ t M
C X E k1~
-
- c- 1-1 ~ C middot-middot -_J
I--- -J 4t1middotbullmiddotmiddota
D
-+-shy --middot liiit1J
r ~
ltL
t-1 _ V
0 -+-1
middotr -bull ~1
X
C
+- bull1 ~1 atr l -=
~~lll(k~IHmiddot~~-Wa~middot~nL1middot~~~ta KF u n P middot~dE-tveRS ~-tl~~~-u-~~~tmiddot_-~~~~liil-~-~~~ tc r
I bull
bull -r~l-yen--~1l~a l fl __ A4o bullbull bullbullbull ~ 111bull--middotbull
8 NJ industrial sites 28 samples in industrial areas with benzene consumption facilities Torrance CA Tuscaloosa AL National Forest 1000 sarip Ies-Torontt) Azusa CA El Monte CA Long Beach CA Upland CA Los Angeles CA Riverside CA
bull Assumes all year operation on either feedstock If plant operates 50 on each feed the population exposed to greater than 50 increment over background goes to 16025 - 19377 and 4044 - 16025 for 100
Ambient concentrations of benzene vdry over one order of magnitude in the literature and therefore make tnis calculation questionable For Kaiser therefore the carcinogenic PAH assessment was used to evaluate incremental population exposure above background
+ The partition of emissions from Oow are 78 carbon tetrachloride and 22 perchloroethylene by weight
173
1142 Comparison of Incremental and l3ackground Concentratiors
Those sites whicn elevate background concentrations greater than SO~~
City Kentucky 6600 The latter two are associated wi t11 EOC pl ants Data
taken in senFice s~ations ano lrctffic areas ir various cities range from -~
300-3640 ngm~ As previous~J mentioned the Stauffer plant is discontinuing
operations These two sources should be examined as part of any possible
start--up permitting activity
11422 Kaiser Steel Corporation
The Kaiser steel plant polycyclic aromatic hydrocarbon (PAH)
emissions cffmiddotise from the coke oven omicroerations Comfdrison of the cumulative
population exposure Figure 114-3 and Table 114-4 with the specified
background levels for urban areas illustrates the breadth of the exposure
distribution The five known PAH carcinogens that were isolated in the coke
oven emissions were quantified in tile ambient air of Los Angeles by Gordon
1Y76
Benzo[a~pyrene 046 ngm J
t3enzo~e]pyrene osio Benzaanthracene U lo
Chrysene U60
lndeno Cl23-cd~pyrene 134
Althouyn these concentrations ctr~ low co1npared with 1 number of other cities
cited in the literature and repres2nt a very I imited data base tt1e predicted
concentrations from the Kaiser plant ~enerally exceed these levels by d
174
significant margin ie greater t11an 30000 are predict2d to be exposed to
4redter than l()-lt t 1tbull 1111tlit~nt background of 3j nqm 3 It ic 1 iklly tlat
lHrllerte exmicroo~urt 11 Ilie drld dre also 11evatPd ovr r11nbient I1owever since
backyround concer1tr1tions of t)enzlmiddotne show large variation no population
calculation was specified
11 4 bull 2 bull 3 Dmv
Carl1on tetrdcnloride releases from Dow constitute approximately 78
of tl1e total CT plus perc emissions in Table 114-6 and were found to elevate
urban background concentrations greater than 50 in five census tracts Emissions are predominately from storage and check tank working and brething remiddot1 eases
1142~4 Allied Chemical
The Al lied plant was modeled several ways since the plant can
operrttt~ wiUl chlorotor111 dr cc1rbon tetrctlt hloridt fo(~d The c1bull-i pr~middot11ntlid in
ldlgtlt~o ll4- drld llil-B represent dflrlUil 01Hrdtion with eitt1er teed Partial
yedr operation with each feed can be scaled from the individual annual modes
and is presented for equal hltllf year operation in Table 114-9 As with the
Du Pont plant carbon tetrachloride emissions arise from feed tank working
loss fable 114-10 illustrates peak exposures predicted to arise during an
off loading cycle As expected concentrations in that six hour period far
~xceed annual average values Insufficient data were available to contrast
levels with backyround transient concentrations
175
l~O
AUHLAlHJ~ CONTROL n CHNil)UES
Alternative control approacnes for the most significant emission
sources dinmig ~he various pl1ants n ees fitie(J n the follo11 ric suJsections
EmpiiasVi nas Deen placed in dedlinq ~riU1 Uiose sources of 91reatest absolute
inaynHud2 (lbyr) and those const1 ut 0ing trie largest increm121middot1t middot] bulli~he
bdc kgrnund con cent ration of Hie em~ tted subs tcince i~o ctt tention Wi 11 be ~i ven
to the s~iondc1tY sources witt1middotn cK1 microant or to trlL case of J011ns-Manvi I le
since -UH imiddotajor source is less tran LOO 1byr and is not microrejicted to raise
bctckgro 1wd exposure levels to the ~~ntr21 population Furt12rmore a11
emission sources dt Kaiser Ste 1 1iC(1 v1 1~n directly dedlt -1ith in this
program r1re elltlteci ro t11e cc1lt1bull ovcmiddotr nperatmiddotions These faci 1ities are to be
closed du1rm and all primary reel 11i~ oH~rdtions discontinmiddot 2d Kaiser
forecasters have predicted further deterioration of the plant economics and
the phased closure has been accelerdted for primary steel making operations
Note that this closure is essentially irreversible cince differential
exmicroansion and contraction of the coke oven structur1 occurs in the cooliny
process and it would be improbable that ovens could be reheated without
extensive reuuildiny at major expense
12l STORAGE TANK EMISSIONS - fAUFFrn IHJW OUPUNf ANU ALLIED
At c1ll four sites emissions from storage tanks constitute either the
primary or near dominant (Stauffer) source of carcinogen release andor
genero1i iJOpulation exposure Curreritly the tanks of interest at each site are
permitted by the local Air Quality IJistricts hJwever they do not require
emission ontrol systems for vdrious reasons Tl1e estimated releases grounds
for exemption and other pertinent information are given in Table 121-1
In order to amicromicroreciate tth~ practicdl dlternatives for emission
controls the provisions of SCAWMU Rile 463 ar~ listed below which specify the
acceptable alternatives for tdnks r4uirin9 controls ie tanks 11aving
capacities greater than 39630 yal with suostances of true vapor pressure
176
Table 121-1
Site
Stduffer Uff-Site Tanks (J) (Sdn Pedro)
Allied (El Seyunrlo) - KdJ
1bulllctLer1 d I reld iturctjc
-J -J
Dow (Pittsbury)-Product Check Tanks (4)
Uow - microroduct sturag~
DuPont (Antioch) - Raw Materidl Feed Stordge
Approx Capacity~
6lUS x 1g (c) 84 X 10 (1)
lt3Y630
18000 each
70000L) 3UUOUO
30000 working 42636 liquid full
STORAGE TANK
Substance
Ethylene Uichloride
Carbon Tetrctci1 I Jr i J1
Perchloreshythylene and Carbon Tetrashychloride
CT perc
Carbon Tetracr1 l ori de
SUMMARIES
Emission Mode
Tank breattli ng
Tank v10rk i ng Tank breatttiny
PERC-~JOrk i ng PERC-b r2a t ~ i ng CT-working CT-breattling
Tank breathing Tank breatt1i ng
Tank workin9 Tank breathing
Vamicroor Pressure
014 at 70 F psi
1 63 psi at 68 F
16d psi at 68 F(CT) UJ9 psi at 68 F(perc)
168 i)Si dt 68 F
Grounds for Exemption
Exempt from control re4uirements to SCAQNJ Rule 463 since vapor microressur-2 dues nut iXC~ec
l~ io at sturdJ~ conditions
txe11pt TlJil fule -+b3 since tank capaL ~J ~ s u~der 39630 gal (lSO 11 )
Exempt from Regulation Rule v~~vu u~~~~~shytank capacity is under 39630 ga 1 andor substances are not classified as organic 1i4uicts
Exempt f ro1n Ru l ~ 85300 because subsLctnce is not classified as dn organic li4uid accordin to Regulation-Rule 8120
exceeding l~ microsi at storage conditions
~ floating roof tanks
~ fixed roof tanks with an internal floating type cover
bull a vapor recovery system with vapor collection and retllrn (or disposal) proce5sinJ lXctbullerlin~ 95 efficiency
ln a cllemicai plant a typical vapor recovery systein (KR Evans
SCAQMU) migm consist of a collection inanifold to d recovery system (suctl as a
vapor sphere) to a compression syst2m dnd subsequently to absorption and
recovery systems The absorpt~on st~n c~nsisting possibly of (rubber
stripper or activated carbon
In dealing with speciti( carcinogenic substances such as vinyl
chloride hiyhly efficient vapor control system perf0rmance has sometimes been
stipulated necessitating incineration systems
Fer th~ cases of corccrn realistic alternltltives are as follows
Stauffer Off-Site TanKs - Tnere are a number of options which can 1~arkedly 3
reduce ei1issicns from these tanks from th~ calculaL~d value of over 20 x 10
lbyr One alternative is to consolidat~ the material in tne three tanks
One of the large tdnks can contain the currently stured material and reduce 3emissions to apmicroroximately 13 0 x 10 lbyr Anotl1er alternative is to
transfer dll E0C to a single floating root tank Various types of such tanks
exist dlld are reviewed in the EPA report Lrganic Cht~mical Manufacturing Volume
J Storage Fugitive dnd Secondary Sources EPA-450J-80-025 eg internal and
external floating roofs and a variety of seal desiyn configurcttions Generalshy
ly such designs would be expected to reduce emissions to the order of
one-fourth to one-fifth the current level Cost of 1nstalliny ct contact
single seal internal floating roof wds estimated by BB Lumquist Pentrex
for EPA as $27770 in 1Y8U dollars for 7U ft diameter tdnk Cost to build an
external floating roof tank was estimated by G Stilt of Pittsburg Oes Moines
for EPA as ~14UUUO for D=67 H=4Uft Anottler com111on approach is to utilize
cdroon ctdsorption This works we11 witn nonpolar hydrocarbons dS VvC is
removed from the vapor phase A basic syte111 consists ot tvo carbon beds and
a regeneration system Regeneration is typically performed with steam or
vacuum Figure 121-1 illustrattS rhe systems (Basd2kis 1~110) Steam raises
tne VOC vctpor microressure The resulting sti~am-VOC mixture is condensed and
173
AC-TJA~D CA-e0-l
-----~---
LOW- ffiESiJ~C ~T~H----
PiltOCESS COOL-J~ At-JO oLOWER DRY t-J~ 2gtLOWE~
COgtJDENSER
J
COOLIN(a - --------
WATi=-A
RECOVeRE SOLVENT
WASTE WATER
Fi l 1 1 middot gure middot - Activated-c arbon Adsorption System
179
rout2d to c ~epordtor decanted and returned to storage In vacuum regenerashy
tion VOC vapor is desorbed by pu1 ling a vact1um on the carbon bed then condens-
ed a1d returned System efficiency ~ ~ est ii1a ted at 96 f(duc middot i m from fixed
roof levels (EPA 4503-80-025)
H~fri 9ctated vent concicr1sers dre une of the most cu1mon emission
reduction prJcesses -for conttol ~ ng hx2-d-- storage tai vUC Figure
12 1-- 2 i l 1ust rates a u n i t ( Er i scmiddot n 19HU ) eff i c i en cy of middot middot~ ~ c middotlt middotmiddot _y i s rate ct
bet1Jee~middoti 60-90 for the vapor pressure range of concern For such a large tank
the capital costs would be high ~igure 121-3 illustrat~s EPA estimates
(EPA l ~~SOb) for the condenser section Condenser sys tern a ea mu l d be in
excess of 1000 ftL
It should be noted t -at Jtc1 uf fer Uiemi cc1 l has arH1ounced the c 1 osu re
of its VCPVC plart PrevfoJsly tt12) had planned to purge tlle off-site tanks
of EUC Since any future po~sible micro1ant stdrt-umicro will necessitate a compreshy
hensive SCAQMD review this ltoument can assist in evaluating proposed control
measures
Allied and DuPont Feed Tanks - In bott1 of these cases the nore significant
quantity of emissions arise from tank workinltJ ie during the off-loading
activity rather than tank breathing Contnl ~easures taken for working
emissions are thus of primary concern Ther~fore no detailed discussion will
be provided on the alternatives for control gtf bredthiny emissions Additionshy
ally both tanks are in the range of 20U00 gallons which is a capacity where
floating roof tanks are almost nonexistent Out of 670 floating roof tanks
surveyd by EPA less then 15 were smaller than 30000 gallons in capacity
Therefore the utilization of any floating roof concept and seal combination
will not be considered
Commmerc i a11 y it is estimated by Ou Pont that carbon tetrachloride
feed costs approximately $017lb (E Taylor personal communication)
Ttierefore less than i2500 is lost to l)uPont annud l ly as a result of working
l eve l em i s s fo ns and 1es s fr om Al l i ed bull Thus i t is u n l i kel y that any
appreciable economic incentive exists to dev1~lop ci vapor recovery system
However a Cdndidate system could be pattern~d after the chloroform
feed-storage unit currently at Allied This is a dedicated vapor balance
system Chloroform is off-lndded from tank cars and stored in a closed
unvented tdnk As the 5tordye tank is filled the c1ir spdce d1spldced is fed
Figure 121-3 Installed Capital Cost vs Condenser Area for Various Materials of Cunstruction for a Complete Condenser Section
182
back to the tank car Also in this case the storage tank is not vented in its
normal breathing mode and is part of a closed feed system to the reactor
Vapor recovery system alternatives which are particularly effective
in loading and handling include r~frigeration adsorption andor absorption
Control efficienci~- 1re est1mateJ in the rlt1nge of 90-95 (EPA 1980b) and of
course depend on the speci tic su1)Stance and equipment used Carbon adsorpshy
tion systems were discussed above The smallest capacity carbon adsorptionmiddot
system priced by EPA (EPA 1980b) has 2 vertical beds of carbon (900 lbs - 4ft
diameter by 3 ft depth) witl1 an ilstalled cctpital cost of $135000 based on
December 1979 dollars
Dow Tanks - Dow has two pair of p~oduct check tanks which alternately are
filled dnd off-loaded Emhsions from th~sl tanks were calculated based upon
operating cycle (3 day fi 11) satubull-dtion vapor pressure and physicdl
chardcteristics Direct hec1d-spa1e testing was planned however it could not
be accomodated because of r 1stri c ed access due to unscheduled mai ntendnce on
the field test day
Dow has indicated that chey are studying the option of installing a
vapor controlrecovery systt~m in 1hese and their largerproduct storage tanks
The extent of their engi neri ng ind assessment work is unknown as are their
current plans It may be possibk to incorporate a vapor balance design into
the system whereby the disp1aced apor is trdnsferred to another process point
within tl1e system Alterna1ively a vapor r1middotcovery system such as carbon
adsmiddotorption is fedsible Hot-1ever since emissions for t11e check tanks are
primarily due to working loss tht~ use of a conservation vent or an adjustment
of its operational differential p~essure would be ineffective for tr1e
reduction of the bulk of such emissions Furthermore since check tank size
is re1at i ve 1y sma 11 no cons i derctt I on was given to conversion to a fl oat i ng
roof configuration for those tanks Conversion would be possible for
the large storage tanks
12~ WASTEWATE~ EMISSIONS - STAUFFER
Emissions of me throug11 ~-a~)te-1dter discharge are the largest single
source identified at the plctnt sites The discharge limit was set at 25 ppm
by the Los Angeles County Sc1nitation iistrict and was estdblished with the
occupational limit in mind of 50 ppm JVer an 8 hour work shift dt the District
18~
tredtrient center~ T11ere hctd been suine di~cuss1on between pdrti~s of possibly
raisiny tne discharge limit since it is felt by Stauffer that dilution of the
stream is sufficient to d 11 ow it Itmiddot shou Id he noted t10-1ever tna NIUSH has
recommended the permissible exposure limit be reduced to 5 ppm averaged over a
work period of 10 nours microer ddy 40 ours per Ieek witt1 a cei l iny level of 15
pmicroin dVeraged over a 15-mi nute microet~ v(J
It is not certain at 1~hrt rdte me 1rill be reledscd from the
wastewater stream At the microlant discharge points wastewater is both hot and
aerated thus favoring release~ No measurements have been ta~en downstream of
the plant after considerable dilut1on has occured The distance to the water
treatment pa11t is approximately 5 kin at 1i1ic~1 point anerobic diyestion is
conducted lt is presumed that a11 ELlC will be released before final ocean
di scnar~Ji
A possible emi ss i D~ cont ro process for nmiddotduct ion of the EDC in the
effluent is by ~recess adjust~Pnts or additio~s For example process
modifications to the EUC stripping stage coula dramatically reduce discharge
lev~ls A control alternative is the use of octivdted carnon or XA0-2 resin
to recover EOC in tne discharge stream Tests of Gulf South Researd1
Institute on XA0-2 and activated carbon (Coco 1980) show high recovery yields
for nonpolar organic carcinogens unaer a range of concentrations Viable
suggested alternatives oy Srnitn included regeneration of Ue trapping
materials dnu even consideration of burning tt1e concentrate carbon media (at
greater than 1000 pmicrom)
Control approaches to reduce emissions from so called secondary
sources in general and waste water emissions in particular fa 11 into four
categories - waste source control resour(e recovery alternative disposal and
add-on controls Alternative control pro(~esses which were considered but
appear to be inappropriate to this case i11clude chtmiddotmica means eg
neutralization precipitation coagulatior1 and chemical oxidation thermal
destruction of tne unconcentrated ~dste s1ream is ii1practical biological
treatment eg aerdtion and biomass-wasteater contcct ~ienerally relates to
the treatment of soluole degrddable organics in the concentration range
bet~~een U01 and l~~ terminal storagt eg lanctfilliny urface impoundment
and deep well injection dre either indppl 1cable or 1mmicroractical Therefore in
summary the poss i b I e contra l approac ties fur this ca e inc 1ude the improvement
r
184
of semicrodration efficiencies in steam stripmicroing tl1e internal recycle of ~aste
stredms ctnd the ddsorption b activdted carbon The design configuration
~fficiency and cost )bviously depend upon numerous microlarit specific factors and
their determination vJOuld rec-Jire detailed analyses
n1e ~~dsrewctter systbull-m of a model cl1emical production plant based
upon the average properties uf a composite of 30 chemicals was evaluated for
EPA by IT Enviroscience (EPA 1980b) Included prominantly among the 30 was
tUC with the hignest uncontrolled secondary emission wastewater release rate
ie ~ percent of tt1e production and 34 perc~nt of the emission Cost and
impact analyses were evaluated for alternative control systems to reduce
secondary voe emissions from wastewater Four systems were considered a
carbon adsorption system (eAS) for recovery of the voe from the wastewater a
cover to reduce secondary VOC emissions from the wastewater clarifier a cover
for the clarifier plus a carbon desorption system and a cover for the
clarifier plus ct eAS system usinu a fume incinerator The scale of the model
system was greatly in excess of the Stauffer plant thus further making
detailed comparison i mpract i cal bull However emission reduction factors ~-1ere 6diven as ~9 and cost effectiveness per 10 g reduction genera1middot1y ranged
between $450 to 1733 These factors would likely grossly underestimate the
system cost if scaled down to the range of the Stauffer plant ie the order
of 34600 lb annual discharge
The alternative approaches of improved steam stripping efficiency
and internal recycling of the stripper iischarge stream with a reduction in
makeup re4uirements could decreas~ net ~DC wastewater content release
123 CONTROLS FUR SECONDARY LEAD S~1tLTER STACK EMISSIONS
Given our findiny of low (16 kgyr) ~missions of arsenic from the
reverberatory furndce at RSR it middotJOuld ippear that the arsenic content of themiddot
lead feedstock is low andor that the microants system for reducing lead
emissions is dlso quite effective for arsenic ~SR it will be recalled from
Section 32 uses a quenchinu cnariber ctnd Dayiwuse filters to remove
pctrticuldte matter and a carbonate scruooer to remove sulfur dioxide
There are no federal new source microertormance standards for lead
eini ssions per se t10wever the Standdrds of Pbullrformance for Secondary Lead
SmeHers (40 CFR 6Ull2) limit total particuLte emission from a blast furnace
185
or reverberatory furnace to SU mgm 3 According to Augenstein et al (1978)
who reviewed the techno1ogy for control ing lead ~missions from ttwse sources
baghouse filters or wet scrubbers are yenerally used to contro1 particulate
emissions When fabric filters are used to control blast furnace emissions
they are nurma11y preceded by an afterburner which inciner~tes hydrocarbons
that bull~GU Id otnerwi se b 1 ind t ne f o1 ur-i c After-burners a re nor necessary for
reverb2middotmiddotatcry furnace emiss-ion smiddotnce the excess air and temperature are
usually sufficient to oxidize tn2 hydrocarbons
According to Augenstein et ali 11 sr1aker--type baghouse filters are
the 1nost effective means of controlling ead fume emissions from secondary
furnace opErations 11 Collection etficiencies can exceed 99 oercent One
advantage to this control approact is that l~ad oxide dust can be recovered
easily cmd recycled in the smeller Flue gases must be coated to below 300 degF
for dacro~ bags and to below ~00 degF for fiberglass bags (Hi~h et al 1977)
Although wet scrubbers are effecti1e under some circumstances in
controlling lead emissions it is more difficult to recover the lead oxide for
recycling In addition sulfur dioxide presi~nt in the flue gases becomes
oxidized to sulfuric acid and can cause corr1sion problems For this reason
sodium carbonate or other basic reagents ar~ added to the scrubber solution
Although it concerns a yold smelter an approach described by
Marchant dnd Meek (1980) provides an exampli~ of an arsenic control
alternative which might be apmicrolicatgtle to secmdary lead smelters At the
Campbe1 1 Red lake Gold Smelter in Balmerton Ontarion Canada Smelter gases
are first passed throuyh a hot electrostatic precipitator (ESP) The ESP is
heated so that the arsenic (which is principally in the form of As ) remains2o3 gaseous and is not yet collected This exclusion of arsenic allo~~s the ESP to
recover particulate gold rnore easily The ESP exhaust is then quenched with
ambient air to condense the arsenic trioxide Baghouse filters then remove
the arsenic along with other particulate matter
124 CUNT~OLS FUR STEEL MILL EMISSIONS
Given the imminent and irreversibli~ cessdtion of coking activities
at Ka 15 er Stee 1 Corpora t i on a rev I e~ o r t e c Imo 1 o gi es for ( on t r o 1 l i ng
~missions from the coke ovens rnd he Clt1ke byproduct recovimiddotry pldnt WdS not
deemed to be necessary Tnis is t1e only primary steel mill in California
1Pi6
KEFUENCE)
Al lf~n Jr Ll~ llJBU1 J 111odel to 1middotstimite hi1drdous (bullmissions from coke oven Juurs pr~pdrelt1 DY 1-kSecircn Tr1dnye7nrffut fur U~ Environmental Protection Agency Uffi ce of Air ~ua 1i ty and Standards fltesearch Tri ang I e Park Nortn Cdrolina KTl17362-UL
Jl len Jr CC 1~8Ub tstination of chdr in(i emissions for by-product coke oven microrepared by fltesedrch Tr1anglt Insc1tute for US nvironmental Protection 1-yency Uf f ice of Air l)ua l i ty and Standards Ke search Triangle Park North Cdrolina RTI17362-0
Allen Jr CC 198Uc 11 Environmeritdl c1ssess1ent of coke by-product recovery plants Proceedings First Symposium on Iron c1nd Steel Pollution Abatement Technology Ctncayo Illrno1s--Tin1ctober - 1 November 1979 EPA-600-9-80-012 pp 7~-dd
Augenstein ur1 T Cornin et al 1978 Con_trol techniques for lead air emissions prepared by PEDCU Envircnmental Inc for US Environmental Protection Agency Research Triangle Park North Carolina EPA-4502-77-012-B ( NT IS P88U-197551)
jarber WC 1977 Interim air pcllution a~sessment for eight t1igh-volume industrial organic chemicals 11 US Environin~ntal Protection Agency memorandum to Air and Hazardous Materials Uivision Kegions I III-X and to Uirector Environmental middotPrograms Uivision Region II
t3drrett KE WL Margard et al 1977 Sampling and analysis of coke-oven door emissions Prepared by ~dttelle-Columbus Laboratories for US Env 1 ronmenta I Protection Agen y Industri a I Environmental Research Laboratory Research Triangle Park North Carolina EPA-b002-77-213
t3 a r ret t flt bull r bull and P bull K bull ~~ebb 1Y78 bull 11 E f f e ct i venes s of a wet el e ct r o stat i c precipitator for controlling PUM emissions from coke oven door leakage Presented at 71st Meetiny of the Air Pollution Control Association Houston Texas 25-29 June 1978
l3asdekis HS 1980 Carbon Adsorption Control Uevice Evaluation IT Enviroscience Inc report in prepdration for US Environmental Protection Agency ESEU
8ee fltW et al 1974 Coke oven charging emission control test program Vol I c1nd II US Environmental Protection Jgency EPA-6502-74-062
~eimer RG 1978 Air pollution emission test Analysis of polynuclear aromatic hydrocarbons from coke oven effluents Prepared by TRW Uefense and Smicrodce Systems Groumicro for US tnvirorunental Protection Agency Office of Air wuality Planning and Standards Kes~arch Triangle Park North Carolina EMB Kemicroort 8-CKU-12
187
Braman RS 1976 Application of the arsine evolution methods to environmental analyses presented at the International Conference on Environmental Arsenic Ft Lauderdale Florida 5-8 October
Bratina J E 1979 11 Fabric filter applications on coke oven pushing operations J Air Poll Cont Assoc 29(9) lll6-920
tiuonicore AJ 1980 Environmental assessment of coke quench towers in proceedings First Symposium on Iron and Steel Pollution Abltlternent Technology Chicago~ Uirnois JO October - f N~vernber 1979 EPA-6009-~1b-62~ pp 112-142
t3usse A D and dR Zimmerrao1 1973 User 1 s guide for tdegh~ Climatological Uisper~ion Model US Environmental Protection Agency Resea~ch Triangle Park Nortl Carolina) EPA-R4--73~0024
California Air Resources 8oard 1976 Coke oven emissions miscellaneous emissions and their control at Kaiser Steel Corporations Fontana Steel making facihty Enforcement l3ranch) S2crcmento California L amp E-76-11
Calvert~ CA 1976 Envir __ Sci_ i( __ Technol 10256
Coco L~i 2t aL~ 1979~ 11 02c10pmert of trec1tment and control technology for 11refractory pet rochemi co was tt=1s r middotl 1d l Report Gu 1f South Research In st it ute
to U S tnv i ronmenta l Protection Agency EPA-ti002- 79-080
Colucci~ LM and CR Begemi 1971 Palynuclear aromatic hydrocarbons and other pollutants in Los Angeles air In Proceedings of the International Clean Jir Congress Vol 2 Academic Press pp 28-35
Cooper JA and JG Watson Jr 1980 Receptor oriented methods of air partkulate source apportionment 11 J Air Pollution Control Assoc 30(10) 1116-1125
Coutant R W JS McNulty and RD Grammar 1975 Final report on determi 11at ion of trace elements in a combustion sys tern Prepared by Batte11 e Columbus Laboratories for the Electric Power Research Institute EPRI 122-1
11 11Dilling W 1977 lnterphase transfer processes Envir Sci TechnoL 11 4 pp 405-409
Oow1irgl) MP JO Jeffry and AH Laube 1978 11 Reduction of quench tower emissions Presented at the 71st Air Pollution Conotrol Association Conference Houston Texas 25-30 June APCA No 78-92
EPA 1980a Review of Criteria for Vapor Phase Hydrocarbons 11 US Environmental Protection Agency ~eport 6008-80-045
EPA 1980b Organic Chemical Manufacturing Volume J Storage Fugitive and Secondary Source EPA report 450J-80-025
EPA 1981 Compilation of Air Pol lutdnt Emission Fdctors (Including Suppleshyments 1-7) Third Edition Suppl1111ent 12 Utfice of ir l)uality Planning dnd Stdndlt1rds AP-42-ED-3-Supp I - lt
188
Erikson Uli 1~80 11 Control Uevice Evaluation Condensdtion 11 IT Envioscience lnc report in prepdration for US Environmental Protection Agency ESEU
lrtel liL 197Y Quench tower particuldte emissions J Air Poll Cont Assoc 2Y(9) 913-916
Ev a n s K bull P bull 1981 Pe r son a l Commun i cat i on v i th R bull Zi s k i n d bull
Goodwin DR 1980 11 Air pollution emissions standards in Proceedings First Symposium on Iron and Steel Pollution Abatement Technology Chicago 11 lrno1s 30 October - 1 November 1979 EPA-6009-80-012 pp 37-45
Gordon GE 10 Receptor Models Environ Sci Tech 14(7)792-800
Gordon KJ 1976 11 lJistribution of airborne polycyclic aromatic hydrocarbons throuyhout Los Angeles Envir Sci Technol 370-373
Great Lakes Cdrbon Corporation 1977 Study of coke side coke-oven emissions~ Vol 1 Source testing of a stationary coke side enclosure EPA-340l-77-014A
lirimsrud EP and HA Rasmussen 1975 Atmos Envir Y 1010
Hartnan MW 1~80 Source test c1t US Steel Clairton Coke Ovens Clairton P~nnsylvania Prepared by rRW Environmental Engrneerrny D1v1s1on for US lnv1ronrnental Protection Agency Emissions Standards and Engineering l)ivision Research Trianyle Pdrk North Carolina EM~ Report 78-CKU-13
Hendriks Ilt V et a I 1Y7Y Urgani c air emissions from coke quench towers 11
Presented at 2nd Arrnual M(etiny ot the ir IJol lution Control Associdtiont Cir1cinndti Utlio June 1~7lJ l-dmicroer NU 9-391
Higt1 MU ME Lukey dnd TA Li Puma llJ77 Inspection manual for secondary lead smelters premicrolt1red b Enyineeriny Science lnc for US lnvironmentdl Protection Ayency Office 0f ~nforc~nent EPA-3401-77-001
Holzer G et al 197 11 Collection amp analyses of trace organic emissions from natura I sources 11 Journal of Cnromatogra flhY 142 pp 755-764
Science Applications lnc Final lemicroort to US Environmental Protection Agency tPA-560lJ-80-UU2
J bull Mi l n e l lJ 81 Person a I Cornm un i cat i on wi t t1 bull Zi ski nd bull
Kemner WF 1979 Cost tffectiven2ss model for pollution control at coking facilities Premicroarect oy PEDCo Environmental lnc for US Environmental Protection Agency Industrial Environmental Research Laboratory Research Tridnyle Park Nortll Carolind EPA-GU02-7l-1U5 (NTS PB 8U-118706)
Kessler T AG Sharkey Jr and kA Friedel 1973 Analysis of trace ele111ents in coal by smicroark-source mass spectrometry US 8ureau of Mines Report ot I n v es t 1 gd t 1on s No bull 77 14 bull
189
Kolak NP J Hyde and R Forrest~r 19 1 9 Particulate source contributions in tt1e Niltlgara Frontierffi Prepared t)_y Nr2bulll York State IJepartrnent of Env i ronm~ntc i Con serv d ti on for US En vi r inment ct l Protect ion Agency Region 11 EPA-9024-79-006
Mackay U and PJ Leinonen llt175 Hate of Evamicroordcion of low - solubility contaminants from vJater bodies to c1tmospmre 11 Envir Sci Tecnnol 9 13 pp 1178-libU
Magee tiv HJ Hctll and GM Vc~rlt_a Jr 1973 Potential Doimiddotiutants in fossi1 f1Jel Prepared by ESSO fbullt~scarcr middotnd Engineeriny Co tor US Enviromihm~ct Protection Agency Ufl-R2-7 --2l1~L (NTIS Pt3 as u~9)
Marcriant middotG0H ard RL Meek 1~gu Eva uation of technology for control of arsenic tmissfons at the CdlmpJel Hd Lakmiddot G~ld Smelter prepared for the US
1Env i ronmentd l Protection Agenc)-l-l(lU-sfrT l lnv i ronmenta l kesedrCiI Laboratory Cincirrn2ti Uh70 lYA-60U2-8U--141 (NTIS 8 oU-21Y36J)
Margler L$ HA Ziskind M8 lcgozen (Ind M Axelrod 1979 11 An inventory of care i nog)n i c substances re i ea sect into ne ambient air of California Vol une I - Scn~enin~ and id2nt~fication o-f r~1rci1ogens of greatest concern Science J- pp 1i ca middott~ L J s In c bull Fi nd l kcmicro~ rt ~ i I - U 6 C) - U- 5 0 4 to U1 e Ca I i ~- o r n i a Ai r kesources 130ard
Milne J 198L Persond1 C~mmunicdtmiddotion middot1itt1 R Ziskind
llurchio JC WrC Cooper and A ue Leo11 1970 Asbestos fibers in ambient air of California 11 Cal Horn i ct Air lesourc2s l3oa rd ARl3 4-0S4- l EHS Report 73-2 (March)
National Academy of Sciences 1Y72 Part1culate polycyclic organic matter Washington~ DC pmicro 4-ll
Nati ona I Kesedrcn Counc i I 197b Arsenic prepi1 red by Subcommittee on Arsenic Co111mittee on 1bull1edical ctnd t)ioloyicai Effects of Environmental Pollutants washin~ton uc
Oliver JF and JT Lane 1979 11 Control of visible emissions at CFampl 1 s coke plant-Pueblo Colorado J Air PolL Cont Assoc 29(9) 920-925
Pe I I i z z a r i E bull 0 bull anct J bull E bull l$ u n c h 1s 19 1 nbi en t ai r ca r c i n o gen i c v a po r s Improved samplin~ and analysis tectrnology EPA Contract 68-02-2764
Pellizzari E U 1977 11 Tle 11easure11ent of carcinogenic vamicroors in ambient middotatmospheres EPA-b007-77-J55
Phelps RG 1Y79 lSI-CPA-oatLlle c-ilte oven door sealing proyram J Air Pol 1 Cont Assoc 29(J) JlJ8--JlL
Porter RA 1976 Uismicroersion ~yuation s lucions Dy calculator I guide for air pollution engineers ctnd sci12nthts S cond edition rexas Air Control tioard ~MT1S P~ 262 lJ~)
Redburn T 1980 Kaiser battles aginst l~gacy of woes 11 Los Angeles Timesmiddot (7 Semicroternber irno) Part VI pp 111
Research Triangle lnstittJte 1977 11 Qudntification of benzene in 15D -1bullnoient air samples 11 for US Environmental 11rotection Agency Office of Air Quality Planning a11d Stitndards
Rinkus SJ and MS Legator 1979 Chemical characterization of 465 known or suspected carcinogens and their correlation with mutagnic activity in the Salmonella typhimurium system Cancer Research 393289-3318
Koberts R M 1980 11 An inventory of arc i nogen i c substances r~ leased nto the ambient air of California Tasks 11 and IV KVB Final Report KVB 26900-836 to Science Applications Inc
Rogozen MB LW Margler P Mankiemiddott1icz cind M Axelrod 1978 Water pollution control for coal slurry pipeiines Prepared by Science Appl1cat10ns Inc for US Department of Energy (NTIS SAI-068-79-516
Rogozen MB DF Hausknecht and RA Ziskind 1976 Methodology for ranking elements in fossil fuels according to their potential health impact Science Applications Inc Final Report 260-77-539 to the Electric Power ~esearch Institute
Russel 1 JW and LA Shadoff The sdmpl ing and determination of halocarbons in ambient air usin~ ~rncentration on porous polymers
Siebert PC CA Craig and Eb Coffey 1978 Prelimary assessment of the sources control and population to airoorne polycyclic organic matter as indicated by benzo(d)pyrene Prepare( by lnergy and Environmental Analyses Inc for US Environmental Protection Agency Office of Air Quality Planning and Standards Resedrch Triangle Park North Carolina
Simmonds PG et amiddot1 1974 Distribution of atmospheric halocarbons in the air over Los Angeles basin Atmos Envir Vol 8 pp 209-216
Singh HB LJ Salas and RE Stiles Distribution of Selected Gaseous Oryanic Mutagens and Suspected Carcinoyens in Ambient Air in Proceedings Annual Meeting Air Pollution Controrl Association 82-651 June 20-25 1982 New Url eans
Smith WM 1979 Evaluation of cok oven emissions in Yearbook of the American Iron and Steel Institute pp 163-179
SRI International 1980 Asessment oi human exposures to atmospheric benzene SJ Mara and SS Lee finai report to US Environmental Protection Agency EPA-4503-78-031
Taback HJ 1978 Control uf Hydrocarbon emissions from stationary sources in the California South Codst Air Basir Prepared by KVB Inc for the Cal1fornid Air Resources Bod-rdmiddotKVB No 5804-714
191
Trenholm J-f dnd LL )eek 1Y78 11 Assessment of t1azardous organic emissions fror1 slot type coke oven Datteries unpublished paper US Environmental l)rotection Ayency Ernssion Standdrds and Engineeriny Jivision
Turner U~ J970 Workbook of atmusphe~ic dispersion estiimiddot1Jt2s IJS Environmental Protection Agency k~sedrch Tri1ngle Park North Carolina
US Bureau of Mines lYl f-inercls yearbook 197U Vol 11 p 4~3
Van Os dE I l U bull iJ bull lL Ma rs l and et ct 1Y7~1 o En i r onmt~ nta I a s s es smen t of co k e Dy-product recovery plants prepareu by ReseMcn Trictngle lnstitute for US Environmental Protection AgerKy 11cJustrial Environ11ental R~search Laboratory Kesearcll Tri2n~e Idrk NorU1 Caroline EPA-6U02-7J-Ul6 (NTfS PB 293-278)
Westbrool( C W 1979a Level l ass~ssment of uncontrolled sinter plant emissions Prepared by fesearch Triangle Institute for US Environmenta1 Protection Agcncy Industria1 tnvirnnmental fltesearc-1 Laboratory kesearch Trianyle Park North Carolind EPA-bOU2-7Y-112
Westbroollt C~~ 197ltJb Level l css~ssment of uncontro1led l)-BOP emissions Prepared by Researcn Triangle institute for US Environmental Protection Agency Industrial Environmental Hesearch Laboratory Research Triangle Park ~Orth Carolina EPA-6002-79-190
Wetherold rLb LP Provost and CD S111ith 198U Assessment of atmospneric emissions from microetroleu111 refinin~ Volume 3 Appendix l3 Radian Corp Fina Report to US tnv i ron111enta I Protection Agency EPA-600 2-80-07 5c
Ziskind RA OF S1ilith JL Hdhn etnd G Smicroivey (1980) Determinants of Cancer and Carctiovasculdr Uisease Mortality in Asbestos Mining counties of California SAi Report 068-dl-514 l May
192
APPENDIX A
Rapid Screening and Identification of Airborne Carcinogens of Greatest Concern in California
Lawrence W Margler Michael B Rogozen Richard A Ziskind and Robert Reynolds Science Applications Inc Los Angeles California
This paper describes a method for establishing a priority list of airborne carcinogens within a
state jurisdiction In this case it was necessary to identify from among hundreds of potential
candidates the live to ten materials of greatest potential concern In California as airborne
carcinogens
Because no previous Inventory of carcinogens In California existed published lists rankings
and assessments of national scope were used to Identify candidates By systematic manipulation
and comparison of these data sources 47 materials of some notoriety were chosen for closer
scrutiny This selection was pared to 22 candidates largely by eliminating those which had
very little production and use In California (Substances primarily used as pesticides were
excluded from the scope of this study) The remc1ining candidates were then ranked by additive
and multiplicative algorithms and by a panel of experts The results of these rankings were
combined to produce a single selection of 11 priority candidates In alphabetical order they
ethylene dichloride N-nllrosoamlncs perchloroethylene and polycyclk aromatic hydroiarbons
In continuing studies a baseline emissions inventory Is being prepared and a source testing
program is being designed
In recent years concern has grown over su~pected arcinug-ens New Jersey seshythe possibility that certain materials lected ten volatile organic compounds released to the atmosphere through inshy and five htbullavy metals to be examined dustrial and commercial activity may be further and is currently measuring responsible for a significant portion of ambient atmospheric concentrations of the incidence of cancer in the general these substances in a variety of areas In population This concern is manifested the second year of the study the state at the federal level in the National h1s incre1sed the volatile organics Emissiltm Standards for Hazardous Air stt1died to 20 and begun measuring Pollutants (NESHAP) which limit heavier orianics associated with parshyemissions of the known carcinogens asshy ticulatcs2 In California a very different bestos beryllium and vinyl chloride 1 approach-was taken
Only a few states including New Jersey and California have begun efshy Overview of Californias Approach forts to identify airborne carcinogens of concern to the general public for the The California Air Resources Board purpose of setting state emission regushy (( ARB) is sponsoring a three-stage lations for these substances After reshy st 11d~bull of airbornC rmissions of carcinoshyviewing national use data for known and g1middot11s from anthropogenic atlivities The
November 1979 Volume 29 No 11
first stage which is the subject of this paper was to identify roughly five to ten materials which of the hundreds of known or suspected airborne carcinoshygens are most likely to be of greatest concern to Californias general populashytion Also of interest were those which in order to satisfy occupational health and safety regulations might be transshyferred from the workplace air to the outside atmosphere The second stage which is now underway includes pinshypointing of emission sources for each of the carcinogens of importance quantishyfiltation of emissions nnd design of source-testing methods A subsequent stage will consist of source testing and measuring public exposures to those substances for which data are unavailshyable The basis for regulatory action if appropriate will include the results of this program and other related reshysearch
Screening of Candidate Carcinogens
The National Institute for Occupashytional Safety and Health (NIOSH) lists 1905 chemicals which have reported neoplastigenic or carcinogenic effects and 510 which have otherwise received attention for their neoplastigenic poshytential1 The need to select five to ten materials from such a large number of potential candidates dictated that we devise a way to rapidly eliminate from further consideration the vast majority of the substances Given the paucity of published data on most of the candidate substances the screening method was designed to make best use of readilymiddot
C1111yrich1 19i9-Air Pullulion Control A~~ociatlon
1153
A- I
----- ---- ----- ---------------------available information The screening process was as follows (1) eight comshyplications of known and suspected carshycinogens were reviewed and those subshystances which were not used in Califorshynia were highly unstable in air or were very doubtfully carcinogenic were eliminated (2) after more detailed inshyformation was obtained for the reshymaining 25 substances candidates were rated by two different analytical methshyods (3) an expert panel was convened to review dossiers on the candidates and to independeitly nmk them (-i) from the eight to deven substances ranked highest by all three approaches eleven were selected for the emission identifishycation and source-testing design stages of the CAHB effort
lnitiI $cre~ng ot Potential Candidate Carcingtgens
65 compnunds middotwe ire selected from 642 industrial ur~nnic air pollutants comshypiled by MrTRE Corporation for the US Envircmnental Protection Agency (USEPA) in U1at study pollutants were scored by multiplying four explicshyitly defined rating factors annual US production fraction of production lost to the environment volaflity and toxshyicity To adapt this york to our pmpose we first selected the 114 substances listed as being carcinogeni 1~ or neoplasshytigenic Then the scores of each of these compounds under the criteria annual US production fraction of producshytion lost volatility and carcinogeshynicity were multiplied together Seshylected for further consideration were those substances which had a product score above 50 Another 15 substances listed as being carcinogenic but lacking information for one of the other rating factors were also selected This list of 65 was then compar~d with seven other lists of corcinogeris1
-middot 11 Materials comshy
mon to the reduced MITRE list and at least one of the other lists were chosen for further consideration Added as candidates were those suhstances which are regulated as occupational carcino-
gens by the Occupational Safety and Health Administration (OSHA) and certain inorganic carcinogens Finally substances were added which in our judgment should be investigated but had been eliminated at this point Exshyamples of these are bis(chloromethyl)shyether epichlorohydrin and hydrazine
Next the refined list which now contained 47 substances or chemical groups was pared further hy another rapid screening process Eliminated
were all candidates (1) whose production andor use in California was very low (under 10~ lbyr) and wns not thought likely to pose a risk to a localized popushylation (2) which are very unstable in 1ir or CH whilh should not on t lw hnsis of CUtrltbullnL cidlnre liebull considtbullnmiddotd r11rci-
1154
Tahllt I Suh tances n-iewed in letail
Candidate Substances
Arsenic Inorganic lead Ashestos Alkyl lead Benzene Maieic anhydride
nogenic The result of the initial the rating under that criterion and the screening was a list of 22 candidate mashy corresponding criterion weight The terials which is presented in Table I overall rating for pollutant i is then the
sum of the scores under all the crishyteriaRanking Candidates by AddiUve ancll
Mulllpllcatlve AigorUhms The additive approach has several virtues the main one being that it forces
Many screening or ranking systems the user to make all assumptions exshyfaH into one oft wo cat~middotgories additive plicit In the process of setting up such and multiplicative Some systems are a a ranking system new insights into the combination of the two while others problem under consideration may be combine nn ol1jective appronc 11 with g1ined Once the system is set up it is suhjlctive cvnluition of the remiddot ults 1l rdatively easy to use Where data for The L2 substances surviving th1middot initial scoring pollutants are unavailable arshyscreening were ranked independtmiddotntly by tificial scales can be constructed to the tvo approaches If the same subshy quantify subjective factors Finally the stance rated high~middot under both systems sensitivity of the results to the systems its importance to California was judged subjective aspects may be measured For to be more likely than if it had scored example one can determine the effect of highly in only one method changing criteria weights upon the final
Additive Approach In the additive pollutant ranking Similarly an appreshyapproach the user identifies one or more ciation may be gained of the significance criteria and rates each alternatiw subshy of the range of uncertainty for a particshystance against each criterion while sishy ular required data element by varying _ multaneously deciding the relative imshy rating factor values portance of the criteria Eq (1) shows its A fundamental problem with the apshymathematical formulation proach is that there is no logical basis for
adding the individual scores assignedRating for pollutant i = f W1 R11 under the criteria other than the asshy
rbullI sumption that this simulates or even
(1) improves upon the users thought proshyEach criterion or rating factor (1 ) is crss A major operational problem is assined a value for each pollutant and that of weighting the criteria A common each rating factor is weighted (b W1 ) practice is to give all critNia equal -icnrding to its importance reht veto wltgtight hut that is in itstbulllf a st1Ubullm1middotnl lhe otlwr rritnia llw sc11rtbull for 1111 aLnut the 1rcl11tive importance of Ow 1n1 1 undmiddotr nitn1 n j is the prod11middott of n1ttria
Journal of the Air Pollution Control Association
A-2
---------
Multiplicative Approach In the multiplicative approach the rating for each alternative is the product of the rat ihgs under each criterion
Hating for pollutant i --- 11Ill
UJ (~) Jbull I
A multiplicative approach can have ~omc altlvrntnges over additive onrs First in some cases the ratings can be physical parameters such as concenshytrations or volatilities there is then no need to weight the criteria and hence less controversy over subjective judgshyments Second multiplication generally provides a wider range of scores than does addition allowing clearer disshycrimination among alternatives Finally this approach provides results which are more intuitively acceptable As an exshyample of this last point suppose that exposure and harmfulness levels for candidate substance are each converted to values on aO to IO scale and that a certain substance is both extremely toxic and extremely rare An additive approach would give the compound a rating ofO + 10 = 10 which is equivalent to that of a moderately prevalent subshystance (rated say at 5) which is modshyerately harmful (rated also at =-) A multiplicative approach on the other hand would rate the first substance at 0 and the second at 25
Criteria The six criteria used in the additive and multiplicative ranking procedures are defined in Table 11 Asshysignment of values to the Rij was based upon data gathered from published litshyerature personal communications and panel discussion and has been fully documented 13
Because the purpose of this exercise was to determine the relative imporshytance of the suspected candidate carshycinogens R 1 was scaled to the most heavily used candidate substance benshyzene whose annual production and use in California is nearly 109 lb Materials with a use under 105 lbyr would be rated zero for R 1 and rejected Howevshyerbefore rejecting a substance by this criterion we considered whether its emissions could in particular circumshystances result in high exposures to a Joshycalized population
R2 takes into account the fact that the chemical industry is in continual change Substances of concern today may be phased out while the use of others may rise dramatically increasing their imshyporta nee asmiddot poll utan ts Information on developments which could likely result in a change in the growth rate was facshytored into the choice of a value for R2- As an example asbestos consumption in California has been stable in recent years However the pending phaseout of asbestos in motor vehicle friction materials will hasten the decline in asshybestos consumption hence we assigned a value of 1 for R2bull
November 1979 Volume 29 No 11
Ideally one cou1 1 use pouu1c11u
sion as a criterion However in this case emissions were to l e estimated in detail only for the five t1 ten carcinogens fishynally ~elected Thmiddotrtfor(bull a 11w1st1n of tt II iim pol cint ial vus 1~1middotd 1atbulld llpon
knowledge of the s11bstances manufacshyture and USP for R The higlwst rating went to substanc middots which are v idely used especially in consumer products A slightly lower rating went to subshystances which an~ routinely emitted from industrial processes during proshyduction and use Some materials are employed in such a way that emissions are quite low even though tight emission control may not be required by law Materials in this cntegory were assigned a value of two for H3bull Substances which under federal or st ate regulations may not be discharged to the exterior envishyronment but which could be dischJrged by accident received the lowest rating
Each candidate was evaluated (n the basis of its 1ropen-ity to decompbullgtSe in ambient air Materials with half-lives greater than eight hours were considered moderately to highly stable and rated five for R4 bull Low to moderate stability was assigned to substances with halfshylives between zero and eight hours Compounds known to exist in air for only a few minutes would be rated zero
Ta hie II Dcfinitinns of the criteria used
R1 Prbullbullsent use in California
100 of max (109 lbyr) 5 10 1)f max (108 lbyr) 4
1 of max (107 lbvr) 3 01 of max (106 lbyr) 2
001 of max (105 lbyr) l lt001 of max (lt105 lbyr) O
R2 Growth in California use
+ 20 5 +Oolti to +20 4 Positive growth to 0 3 Stahle or unknown 2 DPcline 1 B1middotin~ phased out 0
R Emision potential
Widespread use in ltonsumer products 5 R1middotlatively p1or con rot over emissions 4 Rdatively good control over emissions 2 Tihtly c1int rolled 1
R4 8tabilit~- in ambient Air4 tvlnderale t1 high st 1bility (l 1 middot2 gt 8 hr) 5 Lw to modmiddotrate s1 bility (t1 l -0-8 hr) 3 Unstahle(t2~febull minutes) 0
R Dispcision Potential
Emitted lar~ely as apor or iine 5 particulnte
Emit led largely as rnarse particulate
Rlt Evidence f Carcinu~enici ty
Known or suspected human carcinogen 5 Known mammalian carcinogtmiddotn 4 ~usptctcd 111ammalan carcino~en or 3
known m mmalin muta~tmiddotn Ames ltmiddotst p sitive 2 Prpoundgtcursor 01 co-car inogen
a t11 is the half-liftmiddot
A-3
tion state or anion associations may change in the atmosphere metals do not degrade and were considered stable Asbestos is likewise itahlc Many of the clcc-ompo~ilion rtbullnctions oi orinnic molecules are mediated by light Such substances if released at night would have several hours to disperse in the surrounding area
A rapid way of assessing the relative potential of different substances to spread from a release point is to note their physical state under normal amshybient conditions Accordingly we scored materials emitted as vapors or fine particulates the highest for R5 and coarse particulates the lowest Intershymediate values are possible for varying amounts of fine and coarse particulate emissions from the same source or from different sources
There is as yet no widely agreed upon measure of the relative potenciemiddots of carcinogens although some ranking systems have been proposed14 Extrapshyolating data from in vitro techniques such as the Ames bacterial mutagenicity test and from laboratory animal studies to humans is problematic Therefore a less quantitative measure of the carcishynogenic potential of each candidate substance was used The candidates receiving the highest scores for Rs would be those for which there is strong evishydence of carcinogenesis in humans Exshyamples are asbestos which is implicated in mesothelioma vinyl chloride which has been identified as the agent of liver cancer in exposed workers and bisshy(chloroinethyl)ether shown by epideshymiological studies to cause lung cancer in resin workers The next highest rated substances are those for which human carcinogenicity is unknown but which have produced cancer in one or more mammalian species in laboratory tests Next are those which have not been shown to be carcinogens but which have proven to be mutagenic in test animals Substances for which the only knowlshyedge of carcinogenic potential is a posishytive Ames test (producing mutations in histidine-requiring strains of Salmoshynella) are rated 2 Finally substances which are implicated only as precursors or co-carcinogens would be rated lowest
Substances unequivocally associated with carcinogenesis were considered as carcinogens in this study Conditions of emission and exposure including the presence of co-carcinogens were facshytored into the evaluation of each candishydate where possible Carcinogenic subshystances derived from the metabolism of a precursor were considered as carcinoshygens However ubiquitous substances which have been hypothesized to be precursors (eg secondary amines nishytrous acid and nitric oxide which combine under certain circumstances to
1155
form N-nitrosoamines) were not conshysidered because of uncertainties in the importance of their link to the carcinoshygenic compound and the practical conshysiderations demanded by the scope of the study
It was beyond the scope of this study to jud~e the validity and interpietation of the experimental ad epidemiological evidence upon which the carcinogenicity of candidate substances has been esshytablished We accepted the conclusions about carcinogenicity drawn by the Inshyternational Agency for Research on Cancer or the National Cancer Institute and did not consider dosage or route of administratioll of tested substances However considerations of test validity did enter into the subjective evaluation by the panel of experts
Weight lr1 the sdditive ranking scheme each rating criterion is weighted according to its limportance relative to the other criteria Little precedent exists for assigning these weights In our judgment and as generally agreed by the panel of experts (see below) R 1 RJ and R6 are more important than the other criteria Evidence of carcinogeshynicity was considered to be the most important criterion of aB so W 15 was assigned a valtze of 3 W 1 aid W 3 were set at 2 and W W-i and W5 were set at 1 In order to discern the potential senshysitivity of the rankings to the weight assignments the candidates were also ranked using equaily weighted crishyteria
Ranking Candidates by Panel o~ Experts
Anine-member panel of experienced governmental industrial and academic scientists whose disciplines include~ -organic and physical chemistry indusshytrial hygiene toxicology epidemiology and iegulatory control of toxic sub~ stances was convened to provide addishytional data for our ranking algorithms
to discuss our candidate --ubstanc -sand rejections to suggest pos-ibie ne suushystances for consiceration and Lt rank the candidates indepen11mtly ( f our own ranking Tvo v-ee ilts he fore the meeting pa-d uember-s were jven one-to three-page dossiers on eac I canshydidaie substance
At th~ start of the meeting- before any group discussion the pand was asked to rate each candidate suhtance with a score from Oto 5 Next cmiddotach candidate was discussed at length e provided ~n ovrview and summarized critical issues identifi2d up to that Pbullgtnnt Through materials brought to thtmiddot meetin and their persoril experience panel memshyber Veimiddote able to provide much useful information on the candidates and adshyditional insight in to our ~a ting criteria
At the encl of the two-d1y sc~sion the pa-iel ugain -ated the cai 1didates
Flnal SelediDn
Tabe III show- the highest-scoring substance a~ det(rraine1 by the 1ddishytive and mlltiplictive approaches and by th~ panel in ~e additive approach 21 single r1nking as obuined by avershyaging the two ran kings resulting from using equal and uneguai weights The ran kings ltif most candidi tes were unafshyfected hut carbo11 tetralt hloride chloshyroform chromium and inorganic lead changed more tban thmiddotee positions_ Because uncertainties in the data base predude imputing signifiance to small differences in the Lnal ordering the lists in Table III are prtmiddotsented in alphabetishycal order However it is of interest to point out that benzene consistently ranked hihest
Becausimiddot some cindidates had equal rating scores we ltould nut choose exshyactly ten candidat ~s from the additive and multiplicative rankings Instead the
top nine and eleven were selected from the two exercises respectively We also considered the ten substances scored highest hy the panel at the end of the session The final consensus selection cornsistt-d of the 11 candidate substances appearing on at least two of the three lisL- For the substances included in the cun~ensus ranking a baseline emissions inventory is being conducted and a source testing program is being deshysigned
1ejected Substances
Some comments about certain subshySiF1ces not appearing on the final list are in order inasmuch as they include known carcinogens and compounds which hae received considerable atshyllttion in recent years as occupational c11c111ogens
rouisionally Rejected Substances Appended to the consensus ranking (T1ble III) were vinyl chloride gasoline and engine exhausts tobacco smoke and pesticides No further action by the CRB is recommended at this time for vinyl chloride because it is already subject tu the USEPA emissions stanshydard a CARB ambient air quality strndard and an OSHA standard
Gasoline and tobacco smoke were appcndlmiddotd to this list because each ocshycurs very widely and contains several of the candidate substances reviewed in this study some of which are in the final listing For example gasoline contains benzene ethylene dibromide ethylene dilmiddothloride and alkyl lead compounds the last three being in leaded grades only Both gasoline and diesel combusshytion products include PAHs Tobacco smoke contains among other neoplasshytigenic substances nitrosamines PAHs nickel arsenic cadmium and other heavv metals Manv individuals are inshyvolu~tarily and i~ many situations
Table HI Hi~hest ranked candidates from each rnnkin~ methodbull
Highly ranked but no inventory
l I ighesl consensus recommended Additive Multiplicativ(~ Pan I of eqwrts rnnkmg at this time
1156 Journal of the Air Pollution Control Association
A-4
lirtualy unavoidably exposed to t11-k1ltco smoke Because the sources of gasoline its combustion products and tobacco smoke emissions are well known no specific action was recomshymended for these materials during the tbullmissions innnlory and -ource testing design stages of the present study It was considered importint ho-vemiddoter to draw attrntion to the general publics exposhysure tot hrse substances PrsticidltS are listed for the same reason thou~h ad(bullmiddot tailed examination of pesticides w1s beyond the scope of this study Many pesticides are widely used and some of them are known to be carcinogenic
Other Rejected Substances Acryshylonitrile and vinylidene chloride were placed in a provisionally rejected group because of the panels suspicion that imports of these compounds to Cnlifornia from Japan may be appreshyciable yet are hard to substantiate Should such imports be verified in the future these two compounds would take on greater importance Formaldehyde was also provisionally rejected because the preponderance of evidence indicates that it is not carcinogenic and that bisshy(chloromethyl)ether is not formed from formaldehyde in appreciable quintities in industrial environments 15 The ocshycupationally controlled carcinogens ethyleneimine and beta-propiolactone were rejected in part because of their reactivity in air At the time of this study DBCP a pesticide was no longer heing middotproduced in California and was therefore rejected from further considshyeration
Occupational Regulations and Community Exposure
A question of interest to the CARB was whether the regulation of acknowlshyedged occupational carcinogens adshyverselv affects the ambient air outside the v~rkplace Our general findings can be illustrated by the example of asshybestos
Asbestos is a very widely used mateshyrial for which no ambient air standard exists Concentrations in the workplace are limited to an eight-hour timtmiddotshyweighted average concentration of two fiberscm=1 of air and a ceiling concenshytration of ten fihcrscm 1bull In meeting this standard exhausting air containin~ asshybestos to the ambient air is not reshystricted except by the USEPAs reshyquirement that there shall he no visible emissions containing asbestos piut ides from s11eh foci lit i1bulls (bullxd11din~ hrak(bull shops 1 Considning that undc-r certain conditions 101 asbestos fiherscm1 could ht Pmitted without being visihl( 11 this standard mamiddot allow considerable asshybestos emissicns It is unlikely however that emissions would actually approach these levels First since the OSHA standard cannot generally be achieved
November 1979 Volume 29 No 11
by ventilation n ine the generation or asbestos partid in the workplace must be greatly reltlu ed Second the air is usually filtered 11gt prevent recirculation of asbestos to 1 workplace Asbestos waste must bed 1sposed of in sealed imshypermeable bag~ or containers Thus under current onupational regulations the amhient air 1enerally appCars to be afford(bulld greatn protection than it would without s11ch regulations 1
Conclusion
The screenin and ranilting methodshyology presented in this paper proved to be a feasible approach to establishing a priority list of Jirborne carcinc)gens in California We ftel that it is an efficient means of focu~ing further efforts on emissi()ns inventories source testing and ambient measurement for it not only identifie all the carcinogens of potential concern but it also permits the state regulatory agency to direct its reshysearch resourCtmiddots toward those subshystances of partilmiddotular interest within its jurisdiction
References l National Elllission Standards for Hazshy
ardous Air P llutants A Compilation as of April 1 19 3 PEDCO Environmenshytal Inc for US Environmental Proshytection Ag1middotncy EPA3401-78008 (NTIS No I B 288 205) 19i6
S Gray Ne Jersey Department of Enshyvironmental l 1rotection Trenton NJ private com111unication 1979
1 H E Chrislllsen E J Fairchild and R J Lewis Sr middotmiddotSuspected Carcinogens A Suhfile of th1middot NIOSH Regii-try of Toxic Effects of Chemical Substances 2nd Ed National Institute for Occupational Safety and HPalth NIOSH i7-149
4 B 8 Fuller J Hushon 1 Kornreich R Ouellette L Thomas and P Valker Scoring of rganic air pollut1nts for US Environrnental Protection Agency Mitre Corpt ration T1middotchnical Report MTR-6248 I 9i6
S F R Lozan11 Toxicants Selected for Priority En- 1ronmental Assessment corresponde11ce with EPA Region VI Administrat r 19i7
1gt Survey of inyl Chloride Levels in the Vicinity of Keysor-Century Saugus California US Environmental Proshytection AJet middoty National Enforcement lnvestiiatim Center Denn-rand San Francisco E A-n02-77-017a 19i8
0 Bardin l middotstimon~middot in Hearings on Helationshi1 Bet wfen Cancer and the Environmei 1 US llltuse of Hepreshysentatives bull th Congress 2nd Session Committee (In Interstate and Foreicn Commerce ~crial No 94-141 1976 pp 7~i0
middot T H Ma11middoth Carcinogens in the workplace lwrE to start d1middotaning up Sc-imiddotrmmiddot l17 1-1110) lfi8 (197)
I CH ar1111 Classif1r11t1on of the Exshyposure Haza d Prtbullstgtnled Ii~middot Thirt~middot-Six Clwmirals 011 I llixllmmiddots lternorandum from Clllm1middot I Offi(middote of thtgt Assmmiddotiate lgt1rector for 1 middotarcinoltnesis Division of Cancer Cau~ l and Prevention lo SushyJwrmiddoti-or~middot ( 0 Auditor Gtmiddotneral Acshycounting Oft lmiddote19ii1
I0 Chemicals ith Sufficient Evidence of ( arcinogenit ty in Experimental Ani-
A-5
ma1s Jnlernauona1 Agency 1or 1c-curcn on Cancer IAHC Working Group Reshyport Lyon France 1978
11 Threshold Limit Values for Chemical Substances and Physical Agents in the Workroom Environment with Intended Changes for 1977 American Conference of Governmental Industrial Hygienists 1977
12 M H Rogozen D 1 Hausknecht and R A Zisk ind 1976 A Methodology for Ranking Trace Elements in Fossil Fuels According to Their Potential Health Impact by Science Applications Inc for Electric Power Research Institute 1976
1t L Margler R Ziskind M Rogozen and M Axelrod An Inventory of Carcinoshygenic Substances Released into the Amshybient Air of California Vol I Final Reshyport Screening and Identification of Carcinogens of Greatest Concern by Science Applications Inc for California Air Resources Board 1979
14 T H Maugh Estimating potency of carcinogens is an inexact science Science 202 (4363) 38(1978)
15 C C Yao and G C Miller Research Study on ais(Chloromethyl)Ether Forshymation and Detection in Selected Work Environments by the Bendix Corposhyration for National Institute for Occushypational Safety and Health Cincinnati Ohio Contract No 210-75-0056 1979
16 Occupational Cancer Control Unit Calshyifornia Department of Industrial Relashytions Los Angeles California private communications with staff members 19i9
This paper was developed while Dr Margler was a Staff SriPntist in the Energy-Environment Systems Division of Science Applications Inc (SAIi 1801 Avenue of the Stars Suite 1205 Los Angeles CA 90067 and Mr Reynolds was a Project Manager with the California Air Reshysources Board Sacramento CA 95812 Dr ltar~ler is currently emshyployed as DirEgtrtnr of Bnvironmental Scitbulllltcs for WinzlN nnd Kelly Conshysultini EnintNi t1 Third Stretbullt EurekH CA 9gtgt01 and Mr Heynolds is Air Pollution Control Diretbulltor for the Lake County Air Pollution Conshytrol District 55 N Forhes Street Lakeport CA 95-13 Dr Rogozpn is a Staff Scientist and Dr Ziskind is Deputy Manager of the Energy-Enshyvironment Systems Division of SAi
1157
Appendix f~
_l)ispobullition of Miscellaneous Sites
Gould Inc Ver_n)_~ St~condary Lead S~_n_elter
The emissions of arsenic from the four large secondary lead smelt~rs
in Cali forni) J1~1 ~ bull~tiilnted in the program This estimate was based upon a
uniform fugitive emission f1cJr nt WPll supported by measurement inforriation
Therefore source monitoring was r2cq11)nded 1nd Gould Inc Vernon being
~he largest was singled out It was however proposed to monitor both Gouli
and RSR Corp (Quemetco) in the City of Industry since analysis of the plants
revealed significant differences in plant equipment and engineering The
latter being more typical of 1 nodern facility
As a r2sult Jf pr-~-middott -iiscussions and plant inspections we became
aware that Gould was actively in the process 1)f constructing a new facility
which would completely replace the existing one We have monitored progress
on the new site and concluded it would he inappropriate to utilize program
resources to conduct field tests at Gould ~rr1issi 1rns from the new Gould
faci 1ity wi 11 be bas~rl 11p11) test results from RSR
PG ampE - Pittsburg PG ampE - Salinas a~d So Cal Edison - Long Beach Oil Fuel Power Plants
It was app ro p ri ate tJ cons i dP r the (~mission of arsenic from power
plants during the initial study stage inc~ imiddotrace quantities in the fuel oil
are known to be ernitted Because of the population distribution in tile
vicinity of three plants and some 11realistically conservative estimat2s of
erlissiJn c-1ctors tht~ facilities appearc~d among the top seventeen stationary
sources of potential c1Jn-r-n HovJ~ver as a result of a reexamination of
emission data it was concluded that an error ~dj ~een made in the material
balance of arsenic - resulting in tn emission factor equivalent of a 300
release ~io lit~rature was found to justify 1reater than a 30 transfer
function Thus we estimated at the outside th~ emission factor should be
reduced from 013 lb per 1000 lb to approximately 0013 lb and resulting
ars9nic 2missions for the entire state w~r~ cJ11servatively estimated to be
1760 lbs (from 17600) divided amongst 311 the states power plants Clearly
then the four secondary lPad smelters estimat~ of S9410 lbs of arsenic(
B-1
emissions per year makes consideration of poer pl int ernissio1s if ffsenic a
very low priority
In a literdture review by SAI - Methodokgy __f_or Ranking Trace
Elements in Fossil Fuels Accordi~9__t_~_their Potential Health Iripact (Roqozen
1976) - emissions estimates f0r 15 trac~ substances were researched for c11l
and fuel oil pomiddot1er plant c 1J1v---r-sion Source content combustion pr-)~~
transfer functions and contrcl methodology were co1~idered to develop e~ission
factors The output to input ~atio computed and utilized n -~i-i~ study for
arsenic was 002 to 03 (ie tra1sfer function betw~en 2 11d JO~) reflecti~q
the wide variety of fuels processses and controls nationwiJe The upper end
reflecti1g both high arsenic coals and poor e11iss10n contro 1 devices Clearly
neither of those conditions accurtely apply to th2 thr~~ si~~~ r1nrl thus
substantiate the rlecision not to perform emissions neasur-i-~n~s
Calaveras Asbestos Company - Copperopo l middotis Asbestos Mini n(i and Mi 11 i nq
In the past (late sixties and early seventies) ~h2 ~3laveras
Asbestos Company came under heavy criti-is1 1fter imiddot1sp~ction rieasurements
revealed serious problems However significant reduction of e1nissi0ns have
occured prompted by NESHAP regulation and occupcttional standards SAI
inspected the site in f)~c~1b-2r 1979 under an EPA contract An missions
inventory was published under that work (Ztskind 1980) and the bottom line
conclusion is that currently no significant ernissioi1 are gteing released as a
r~sult of hl-isting which would reach the publ le i 1le 0pen pit is some 900
feet deep with blasting at the bottom 0ver 80 percent of emissions in the
CARB Emission Inventory System were attributej to pit blasting In fact at
its current depth blasted 1aterial does not reach the mine surface with tile
explosive implacement used Additionally the site is more remote than might
be realized The ~ituation is vastly different today than in the past and
currently attention should be paid to the issues of occupational exposure and
breakdown of ventilation syste111 controls in t1e inilling operationsmiddot He
recorrrnended that no furtw~ corsideration be Jiven to this site for the
purposes of tl1is w)JJil ie identification )f silt~middot1ificant releases vhich
might he responsible for causi1J middot1 fmiddot middotgtbullgt middot 1= 1(11 1bull11 tinn Pxposure
8-2
Vctri0J5 Refineries
It was est=tblished early in the anrilysis prmiddotJr111 V1at gtenzene
emissions from refinery operations coulrl in th~ 1guregate constitute a
significrnt source Since there ilre a large rrnrnber 0f refineri~s in
California (46 in Los Angeles County itself -it t(~ i 1~ middot)f r-1ltariination) with a
wide variety of types sizes ages etc their eva l uat 11) 1 i =r~sent -3~ i)7
monumental task It was noted immediately t 13t ~~it hi n the total scope of the
study the design and conduct )f c r~f i r1~ y testing program which would develop
a complete benzene emissions in1~ntory for 01e site was impractical let alone
to d1aracteri ze emissions fro111 three ie those listed among the 17 rnost
s i g n if i ca n t potent i a l st a t i on a r y sou r c e em it 1-_ l rs bull
The three refineries were singled out for special attention in the
previol1s phase ~ecause they uriiquely had components which process materials
containing 10 or more per~nt hy middot1-bulli1 11t )-~112~ne (46 FR 1165 Jan 5 1981 pg
1491 ) Est i mates by E P A ( F e d bull r l Rr-~ g i s t ~ r 1q81 ) i n d i cate that 90 p e r cent o r
more of the total benzene fu 1Jitive emissions arise from such components
Therefore attention is appropriately focused on the three henzene production
consumption refineries Chevron (Richmond) Arco (Crson) and Chevron (El
Segundo) SAI staff considered the possibliJ )f -1 testing program at one of
these refineries and cone l url 1middotd it would not 1)e cost-effect i v2 for o number of
reasons
California is a minor producer and consumer of benzene with
approximately 15 of the 114 hill ion pounds produced and
consumed nation1lly in 1977 This can 1)e riY1pared with the fact
that California has approximately 10 of th~ ~ations population
Benzene exposur~ to the generil nJmiddot1lation has been partitioned
among the various sources Aproximately 90 of exposure is
estimated (SRI 1978) to be caused hy gasoline distribution
activities and 1ehiclEmiddot emissions in 1 ir~an areas with nearly all
of the balance )Y benene hrndl ing operations (refineries and
chemical plants) In the case middotlf California this percent1g~ Jill
be even JT1ore di middotpruJor-tionate b~cause of its greater s1are of the
population and ~esser share of the henzene ha~dling Furthermore
since the three refin 1bullry cites ire heavily urbanized no new rural
population cegmmiddotnt is beinq exposed
B-3
Californias most heavily urbanized Air Quality Management
Di st r i cts have a l ready adopted t en zen e f ugit i v e em i s s i on
staridards comparable and vith S( 1ine features VJr stringent than
the proposed national (bullmission lttandarrl (4f-i FR 1165 Jan 5
1981) Thus the concmiddotlusion rleri1ed above (Le refinery emissions
are secondary cause of exposure to the urb~n microopu~ation) is
further reinforced since H is irojected that re1eases from
components in benzene service (gt10 benzene) will be reduced by
73 by the proposed F~deral Standards T~~ ~istrict rules (eg
South Coast Air Quality Management District 456 and 4fi6l) are
not restricted to berz~ne per senor to only components in
benzene service The rjLs 1lso i-lclude flanges in addition to
the components ca i led uut in the propose nat i 01a l standard
The EPA estimated that if the proposed emission standard were adopted the
1~aximum annual benzene concertr0Vioi1 -~01middot 1 plant would be 36 ppb at a
distance of 0 1 kilomeL~~ avay Cor1paring this ground level concentration
with the general urban background i1 California of 19 ppb shows the latt2r t1
dominate In recognition of the secondary im0ortance of these thr~~ i~~s i~
was decided to utilize program resources to d~velop field data at other sites
1111here little emissions informati1)i1 1as cll-1aila1)le
8-4
APPENlJIX C
US tn v 1 ronrnen ta I Protection Agency
middotJc k
METHOD 108 - UETERMINATION OF PA~TICULATE ANlJ GASEOUS ARSENIC EMISSION
FRUM NONFE~RUUS SMELTERS
1 Applicaoil ity and Principle
11 Apmicroli1ability This rnetnod amicropliltS to tt1e determination of
i nor~an i c d rsen i c ( s) e1n is sions trom non f errou Sm(~ I ters dnd other sources as
smicroecified in the reuulations
12 Principle Particulate and gaseous As emissions are withdrawn
isokinetically from the source and collected on a glass mat filter and in
water The collected As is ther1 analyzed by means of atomic aosorption
spectrophotometry
2 Apparatus
21 Sa111pliny Train A schematic of tt1e sampling train is shown in
fiyure 421 it is similar to tt1e Method 5 train of 40 CRF 60 Appendix A
Tne sampliny trdin consists of tt1e followiny components
Note H1i s and a11 subse~uent nferences to ottler methods refer to the Methods in 40 CFflt 60 Appendix I
3 Annual Book of AST11 Stdndarctbull Part 31 Water Atmospheric
Analysis American Society for T~sting and 1-1aterials Philadelphia PA
1974 micromicro 40-4~
C-15
APPENDIX D
~f-~__d_y_ e s f o r Pr o c e s s i ng a n d An a l y s i s of PA H i n
Co~e Oven Emissions from Kaiser Steel
Global Geochemistry Corporation
10 Samples expected
l 1 Connective tubes between section welded to oven port cover and remainder of sampling system - These are to be stainless steel tubing disconnected and capped off for each sample They will be about 10-15 long
1 2 Impingers in series first two water-filled third dry - The center tubes will be removed and wrapped in clean foil and the impingers stoppered as is for shipping One set for each sample Protect from light during shipping
20 Extraction -procedure
Sample fractions are to be covered with foil or in opaque containers with inert gas flush during storage in freezer Solvents used in extraction will be BampJ glass distilled pesticide grade low residue) Working bench area will be screened with amber plastic and light bulbs
2 l Connective tubes will be rinsed with dichloromethane by partial filling capping tilting several times and draining into a beaker This will be repeated until the washings are colorless but at least three times
22 Impinger contents will be transferred to a separatory funnel The impingers and center tubes will be rinsed inlo the funnel using a measured amount of dichloroshymethane in portions The funnel is shaken well and the extract drawn The rinsing and shaking is repeated with two more portions of solvent The extracts and washshyings are combined over Na 2 S0 4 bull
A separate portion of the water used for impingers is extracted in the same way
For each liter of water 100 ml dichloromethane will be used per extraction ie 300 ml total
Internal standard is added to the combined exshytracts at this p0int
D-1
_rmiddot
30
40
Concentration of extracts
Bulk solvent is removed on a rotary evaporator in a water bath at not over 40deg stopp~ng short of dryness The res~dual volume should be just less than the f i n a l v o l ti 11 e o wl1 i ch the s o l uti on i s to be d i l u t e d (accurately) The object is a final solution with roughly five mgml solids concentration in dichloromethane After the vol w1 e ~ s made up to the mar k i n a vol umet r i c flask an aliquot is removed and evaporated on a tared disk in a nitrogen stream and weighed to determine yields The remaining solution is stored in a ial or ampoule sealed with a Teflon-lined septum A split for SAI QA may be made at this point (see Section 43 also)
Gravity column c1eanup chromatogram
4 1 Into a 10 min 1 d column with co2rse sinter and Teflon stopcoc~ 6 g 120 mesh silica gel (activated at 120deg) 1s sluiry packed in pentane cooling the column by evaporation from a tissue wrapped around it and wet with acetone) This minimizes vapor gaps in the column A cap of 3g sodium sulshyfate is added to the top The column must not dry out before the chromatogram is complete
42 A sample of the extract concentrate is taken This is to be based on experience of expected PAH conshytents but initially would be 5-10 mg This is exchanged with pentane by adding 10 ml portions of pentane and evaporating over about 100 mg silica to small volume Three to five portions of pentane are used The final residue is rinsed onto the column with pentane
4 3 The chromatogram is developed with four solvent steps
l 25 ml pentane 2 l 0 ml 2 0 ~~ dichloromethane in pentane
II II ii II3 l 0 ml 50 u II II4 l 0 ml 5 0 ~~ methanol
The column volume is approximately 8 cc This is allowed for while collecting fractions During the first chromatogram the graduated cylinder collector is observed for visible solvent changes and a more precise estimate of the column volume
T h e f o u r f r a c t i o n ~ a r e e v a p o r a t e d i n a n 1 t r o g e n stream at not oveimiddot 40 to l 0 ml then stored under nitrogen in the freezer The principal constituents
0-2
of the fractions will be
l Aliphatics 2 Lower molecular weight PAH
11 113 Higher 11
4 Polar materials
Fractions 2 and 3 are to be analyzed by HPLC the others retained for future interest or in case of an incomplete or defective separation Splits for SAI QA may also be made from these fractions
50 HPLC Analysis
5 l The column is a reverse phase partition column C18 bonded to microparticle silica 4 x 250 mm The solvent system is a gradient from 6040 acetonitrile water to 100 acetonitrile initial slope setting l 5min The flow rate is one mlmin the injecshytion sample volume one to ten microl depending on conshycentration Detectors are UV absorbance and fluorescence in series The absorbance is set routinely at 296 nm but may be varied in certain instances to aid in identification of peaks The fluorescence filtars are narrow pass excitation at 360 nm and cut-off emission above ~10 nm Both d e t e c to rs a re rec o rd e d s i 1i ul ta n e o u s l y
52 Reference calibration is based initially on indishyvidual chromatograms of single compounds and roushytinely interspersed chromatograms of multicomponent reference mixtures The proposed calibration stanshydards are the following at minimum
Phenathrene Pyrene Chrysene Perylene Benzo(a)pyrene Benzo (ghi)pcrylene I n d e n o ( l 2 ~ c d ) p y r e n e Coronene
The internal standards (added in Section 22) are 910-Dimethylanthracene
9-Phenylanthracene 9 10-Diphenylanthracene
60 Peak Identification
Significant peaks not recognizable as present in the reference standard will be investigated by alternate methods for identification
D-3
6 bull I Stopped-flow examination of UV absorbance for maxima
by manual wavelength varmiddotiation allows checking against known reference spectra (published or measured) This is usually sufficient confirm a component already suspected to be present
62 Collection of the fraction contairino the unknown peak allows later determination o~ t~e full absorbance andor the fluorescence spectrum for comparison with reference compounds
63 If not already run on GCMS the sample can be sent to SAI for that purpose possibly allowing the comshyponent to be identified Since the HPLC elution order is not the same as the GC retention sequence it may be necessary in affibiguous cases to collect the HPLC fraction and run it separately by GCMS
The~e may be numerous minor constituents not identified The decision on how many of these to track down and identify will be made in consultation with CARB adv~sors a~d with SAI since at some point the returns ute not worth further expense
_64 IdeAtified peaks are quantified by comparing their areas to the areas of the known concentrations of reference compounds A recovery factor based on the added internal standard nearest in elution order to the sample component is then applied the amount of internal standard originally added to the sample is divided by the amount found in the HPLC analysis This ratio is the multiplier used to correct for losses in concentration and chromatography steps
70 Quality Assurance
71 Transmittal of sample ~plits to SAI provides for external QA This should be done with 15 (or one in seven) of all samples
72 Blank water extracts are carried through the same analytical procedure (10 of total)
73 The cleanup and HPLC analysis are duplicated on 15 (or one in seven) of all extracts
74 The multicomponent reference is run first every day then elution volumes and peak area for each comshyponent are plotted on a control chdrt to pinpoint developing problems At least evetmiddoty tenth chromato-gram will be this standard It wi 11 also be rerun after any service on the instrument~ such as a column change
D-4
APPENDIX E
Gas chromatograph 1mass spcmiddotctromlt~try protocols used by SAI Trace
Organic Chemistry Laborator1
Aliquots of sampl=s received from Global Geochemistry Corporation
were analyzed by combined c11s chromatography mass spectrometry (GCMS)
Liquid extracts were spiked with an internal standard compound (deuteriumshy
labeled phenanthrene) priot to analysis by gas chromatography mass spectroshy
metrymiddot A Finnigan model 4( 21 Quadrupole GCMS system including an Incas data
system was used for all analysis Liquid samples (1 microliter) were injected
into a 30-miter x 025-mm ID SE-54 fused silica open tubular 9as chromatoshy
graphy column (JampW Scientific) A programmed GC oven rate of a 4 minute hold
at 30degc followed by a 40degCmin temerature increase to 160degc and a s0 cmin
temperature increase to 270degc was used for sample analysis The detector
system was a quadrupole mass spectrometer operated in the electron ionization
mode at 70eV The quadrupole mass filter was scanned from 35 to 475 atomic
mass units in 095 seconds A hold time of 005 seconds was used to stabilize
the electronics prior tote next scan Data were continuously acquired using
a Finnigan Incas Data system which writes time intensity mass spectral data
to a magnetic disc The mass spectrometer was operated under computer control
during acquisition Figures 33-7 through 33-9 show the total ionization
chromatograms for each of the three samples that were analyzed
Following analyses the data files were searched for priority
pollutant polynuclear aromatic hydrocarbons (PNAs) In addition to the
quantitative data reduction used for priority pollutant PNAs a survey analysis
was performed on all significant GC peaks (ie signal to noise of greater
than 101) The survey analysis was based upon a computerized library search of
individual background subtracted peak maxima scans for qualifying GC peaks
The library used for qualitative identification AJas a combined NIHEPA
National Bureau of Standards library 1hich contains approximately 26000
entries
Tables 33-11 through 3-13 present the results of this survey analysis As expected a series of al~yl Slbstituted PNAs were observed in addition to heterocyc l i c compounds con ta i ni n~ nitrogen iind sulfur moieties Further
investigation of the mass spectre of the nitrogen substituted identifications
confirmed the presence of benzonitrile isoquinoline andor quinoline acridine
carbazole and alkyl substibted ~yridines Ion specific searches for N-nitroso
substituted amines were neg1tive
E-1
APPENDIX F
Detailed Fiber and Mass Concentration Data - Johns Manville
=-- -i l L - [- - =-- -= f- - =-- -l J middot - -=-
i-- L =-- - ~ ~ L ~ ~ =--shy r---- --
- 7 l -=----7- -- - - - -- =-- - - ~ =-- - 1- -~- l
g~~~~ r_-middot~ ~-i [-- L ~ L Ldeg
middot--~ l - --------=-- r- L r- r-
~~~ =-- -- r- = - -r---
a- - - - - _ - -I~=-- L -- l 7 -~--
gtZlt _
- _(_ -lt lt =- = ZJZ~
-
__ _
~ -- __ middot- -~ ~~~ = L ~f~ Z ~ C
middot -
__ ~ lt lt =- _~~ijE
gt = z lt gt
rbullshy__
F-2
SI I HMllLfc J H 1 I OBr I lll-H ANI) NS~ coHcrrrllliTJ ON CUfllllTI VImiddot SIJMMIIY
TIITIL JnI~ SfiNtlEO TOTtl Ill- or FI ITETI TOTI Illlt ~111-11 11 ltllOT FIIACT I Ori 1111 OF ~middotWI MUllllNJbull lOTI 01IIMImiddot 01- JII TOTI FI lilmiddott~ UgtIJH nD
= t nwo 00 so MI CllONH = t I 1JOO SO (M = II 900 SO CM t bull (0000 l1T~
I I JO SO CM 70000 CUil JC Ml-TEHS
= ~lt O V l lWfl~
Manville D-2 Baghouse
723 347 -633
CIIHYSOTI LE JfllII I nou J11BICUOUS NO PTTFllN NON-SBI-STOS tLL FIIIEIIS
FI 111-11 COllflT ltFIIWll-
ITllClmiddotrir r lllrf
2gt()
) ( I f S bull1 bullbull
I 0
l ~-Irbullmiddot
00
fl JI)(~
00
0 OUU~~
00
0 OflO
~60
1110 000~~
FI IWII ( ()[IClmiddotNTHT I ON ltI- I mn ll11 ~41 GI OF FI LTEH) (FIBlmiddotn~ ll-11 CIJB f-11bullTl-H OF IH)
I (2fi71bull+0G (aG(l-+05
( gt027 F+04 OfmiddotHH
OOOOOE+OO OOOOOEtOO
OOOOOF+00 o uuoul~middotH)O
0 O(OOE+OO o 0000I+oo
J bull f 1)07 E+O( fi r71HE+OG
I
w
TcfLI ~~lllllmiddotCE 1lllmiddot lt~O Ctl lFII ~~O UI OF FIiTEi) t q UI 11-11 Cll II fllmiddotTlbullll OF J I H)
I 590)1+07 6 t 90llbull-HH
( HMOE+O 2 40411bull-1 o
0 OIHHlF+OO o 00001+oo
0 OUO(I 1 00 OOOOOJi+00
() 10110~ Ji)
OOOOOE+OO 1 lHHllJ i 00
0 00001middotgt1 00
M ~~~ middotor-ic Irrrn TI or
( PGRAMS 1bull111 ~c UI OF F J LTEH) (PGRAMS 1111 UJI illI Lil ()j Ill)
IUitTII Ml-Ml ltMI tJIOfl~~) ~Tlgt lgtEV
Ml- H Lot c1middoton nr COlmiddotF I II
I) I MWTEll Ml-N ltMI c11or1~~) STI) tHV
MEtl IOC cImiddotor1 r1N CCWF V11
AIt lbullI rt 1bulli I~ (Sil MIC) STIgt 1gt1-V
Ml- N 101 c1-ori r1u COlbulltmiddot V1ll
I 0 I ~if1+07 bull tJf I (11bull-HH
r 1gta 1 1) bull 2 (10()
0 lt700 l 1Hi4~ 2G44
() ~( 1 027B
- l bull 6 I 1bull1middot 0 J l)J 09J~fgt
1a 1071 1~ 09rn
0 2(jlt)fi l ~HVil 4040fi
~ HltF+O~ U (H~61bull+o I
o rooo 0()()00
-0GJOB 0 (000 0 (000
OOfiOO 00000
- )()57
OOGOO 00000
0 0 1)02 00000
- I J 0 0 0 1W 00000
00000 0 ()000 00000 00000 00000
00000 00000 00000 0 (WHO 00000
0()000 0()()()()
0 ()()()) 00000 00000
00000 00000 0000() 00000 00000
00000 00000 00000 00000 00000
00000 00000 00000 00000 00000
00000 00000 00000 00000 00000
00000 00000 00000 00000 00000
00000 00000 0 (WOO 00000 ()0000
( ~1701 l B bullJl~H 0 lt~( I BIJI ~ ~i Ill
o~-~ 0 ~j
- I ult-fc 0 I Hbull)I o bull)abullgt~
12 (1 41 17bullgt7
0 1 l I H) 4 ltmiddotI 1
VOUJME (ClJB MIC)
MFiN ~Tl) 111-V ru-1N 10C Ct-OM MN COE- Vtll
2lHH7 gt HUB
-2 7m 0 060 1)
( IJl)()C)
0 001 00000
-ltj 74a 1gt 0001~ 00000
00000 0()000 00000 00000 00000
00000 00000 00000 00000 00000
ooono 0 1)000 00000 00000 0 ltWOO
2 7 0 1J lJ I J J
-2J+H () ()~(
B 77gtB
lt ~
lt
C
J c
~ lf
_
ii
(]j (fJ
J I 0
a1 c r--- l)-- Clltr (Y) r- ro C)(V) gt deg CN C1 (0 I ~ C M
N --- r--
w J
s rJ gt-
c
~
LJ
t-
~ - -lt-[-bull
0- ~ l L L l~ -18~~
L a- - ~ I-
-I - ---
-
-- -- -- - - ~l
-1_ - - --
--~--- [-middot-~--- t-- -- [- -_ 3J ~I r-
z g ---
l
t--[-
l l c +-shyc
~ Ci - - I
-1--r-middot~- ~
L l- L ~
L
II V
C
c
t--
-= -- - - ---- - _
middot= - middot-= l J l~ - ~ ~ L - ~ L ~ [--
II II II II II II II
_
---
--
z
- I
==== lt _ --middot --~ 7 _
- ~~- _
z ~ ~ =lt ~~~ti3
f
gt =2 lt_
2 z _ lt=lt i~iS
-- z ~ lt
~~~t3
~lt~ =--~~~sect~ _ -- -middotImiddot_
L -_ -
F-4
(lJ II) I
~ Or--M
ltlJCsr-M ~ O) bullbullbullbull ~ lCMIO bull- ci gt (= N lC I M ~ 0 N
_
J J
---r--co - - ~l r-j = ~i
~~ 2-
88 rJ--_-ltI -
lt ~lt
z
Cl J ~ _ e-l ~~
- ~ middot~ e-- l ~ l~ [- - ~ -~=----~
_ L
L
--o---
-o-~shyc L~ ~ - 0 - Ll-C d o~-
11middot II 11 11 II II II
-_ 0- _
--
-- lt== ~~ ltZ z-== -~~=E~8C~l~-=rI - lt ~ fJ
ltZ lt---=lt===~~~- _ ltltlt=gt ~~=~~~~~ -
~ ~ lt -=z~ z _ -shy middot--z =-
-middotmiddotmiddotmiddot middot-middot -----
r bull bull _
- _ - ~-middot __ _
~ _
middot- - --- middotmiddot
_
- l
~=~r -~= ==
lt =~~~ti
F-5
FIHJ-ll Nil MAS~ CONCFNTllJTlOi SJ I SMllLE I l) Al OJ CUMJIATI JIbull t-llHMHY
TOTL iHf- SCNNFD lOTiL Illmiddot OF FILTFTI TOJI Ill- JSlllmiddot~ll I I OIIOT FlltCT [ ON 1111middot OImiddot ~1-M iIHlllllJNE TO 1 1 I ()lIJMImiddot ()I t I H TOTI V 1111-11~ COUNTED
= 1080000 SQ MI~RlozNS = 11 900 SQ CM = 1 l bull )00 SO CM = 1 00000 liHlS
11 IH) so CM 2G70000 CUBIC METfiHS
1 ) fi FI JllIIS
Man vi l1 e D-2 Baghouse
723 0 347 -633
Cl Ill Y-OJ l LE MIPII I BOLF Jf1BI(UOUS NO PATTEnN NON-ARHFSTOS ALL FI BFllf~
F 11l-ll f()llrlr ( I 1IFH-- l 17 G o 00 00 00 19fi
Tl fL P1 Sl~JNNFD Tl lI ltl- OF 1bullrn~n H 11I HF ~IJlbullI) lll 1 ll(IT 1-ILCTON IL CW ~U-1 HErlllllNE
THl1 (11llrJI~ OF 1IH Tt1T-d 11111middotH COIJrfflI)
= 697~oo so MICRONS l l 1)00 so CM t I JOO t-o CM
= 1 00(00 lHT~ = I l 1)0 ~o GI
ant()Oo cun1r NETEns = 7 0 I l lWllS
Manville (upwind)
CIIIIYsOllY AMPII I nou JMfl[CUOUS NO PA1vrrHN NON-A~~HFSTOS ALL FIBFHS
v1mI1 Cttllffl l 1-- I II I fl~ 1 () () oo l 0 00 40 70
1middot1111unmiddot ToT1L f J BFllS o ooo~ ooow- t OTT~ ( 01rn 57 I 1~m too ooo~
1middot 1111--11 1or111rnnT I ON I Imiddot I 111 II~~ 1111 ~o CM OF f I LTFTl) 1 11111 1middot111 CUil middot11bullnn Illmiddot J l ll)
O OOOlJEmiddotH)O UUUUOL+OO
0 00001-1-00 0 00011 lbull-Imiddot00
G 1 I nn+OG 1 HJ llmiddot~+OG
00001+oo 0001101-1-00
r n~+H+uG l 7llHmiddotlE+Ofl
l bull l 1)1-n+O(i I J ~ 1)(11-middotl()i
T11T I ~Ull FiCImiddot 1 ll f ( ~ l lrl ll11 tn Ii OF F 1ITFll)
f ill I I I i 11 I U l ( i I Ill J
0 OOl)Ol ll() () U()0 I Imiddot()()
0 UOOOlmiddotgt1-00 O OOlgtUlbullgt100
l 0 1) ) I 1 lJ J ~ tl( 1H-HH
0 UUUOlmiddotH)I) o 00001-+oo
~ lt1Hlt1(E+OG 7(H(Ml-middotHM
0 OOOOJbullH)O 0 0000lbullgtHlO
77 I
N ~
rl gt- 10NCJfffll1T I ON
r orR AMS ITll ~q lfI (II- FIT n~ll) - LIi LIIll 111middotTU OI 1lll)(PGRAMS II i1 Ill MlmiddotN
bull I t bull I 1 f I ~ middotr1 Ill lmiddotl 1- [~ IOC ( Hl1I f r~ Ull-F ll
0 0000 Imiddot bull-110 0 0000 Ibull HO
0 ltHgtOO 000110 t) (HJ()()
0()()()()
0()000
llOOOOF+Oo 0 00001bull-100
0 ()000 OOtlOO 00000 0 (WOO 00000
0 (((7
0 ~0-i -0 middotlbullribull)bulll-
0 ltl l lt 0 oHi
00000 00000 00000 0(0HO 0000()
077[0 0 -1117
-0 ~bull)I)()
0 7middot1middot1 (i () L~bull)7
0 7~B( 0 I I
-o alt17n 0 (1)~1 0 17lt1 I
11 I 11ITlt I j i 1l 1 111bull I
f 11middotgt N gt I ii ii I V 111 ii IOC tHlfl flf1 Clll-tmiddotmiddot VAIi
00000 0()000 0000() 00000 O 0000
041000 ()0000 () 0000 o 0(1()0
0000()
0 llfl(i7 OltHJ
-~ lt-11 I 0 07 I 1 () I)lt)
0 (H)ll()
00000 00000 0 ()()()() 0 0(100
0 0 Omiddotlmiddot l
-c07 1)
0 I 1 I( 0 1101
0 l l j--
OOGli -~bullI) ( (J
0 1lt11 I 077G7
I 1 (71 rJI C)
tll N -r11 111-V rmiddotllfl trn CI llf I 111bull1 COIT rll
() ())(l(J
00000 0 ()(1()()
0000() () (1()(10
00000 00000 00000 0 (I()(()
00000
0 ~ I Ifdeggt 0 I Imiddot~ I
- I H 1)7
O I GOO I I i ) l)
0 0001) 00000 00000 00000 00000
0 1 1)17 o~ViH
-1 0 1)gtG 01lbullH 0 7 I till
0 l l lt G 0 ~r1
- J bull J1 1) I 0 17 I bull at-lt
lllllrmiddot lt Clll ii IC)
~I Imiddotmiddot N ~middot1 D 1)1bullI f-1 Imiddotgt f IOC ClltHI rm uw-- i ll
O(lOOO 00000 00000 () ()(J()(J
(l(t(()()
0()()00 00000 00000 (1(1(1(()
00000
000fi7 OOOGO
-G Jll7l 0011G 1 H20
00000 00000 00000 000()() 00000
0 0 I bulllCi 001w
-4 iHd 00101 I ~700
OOIOH 0 0 I Ilt
-(1 j l)~G
000(i 1tHH
--
--
L c~
U c t c
(
J
J -
C
C
c g 0 0
+ + e--c~ =~ - r~ C- l~
--
+ +---
- -C 0 o- -c--1
I
lt t w
lt
C
ta J ~ ltCi
lt ~
a---~ -- c gt C 0 rtl 5 -
VJ C f-rJJ c
[J_
ltI ~ C z
z _ ~
~ lte 0 ~
C
tJ
e
C
~ c C
L N
~~ C 0
C
l~ 1 0
t CI ~ ----~ -
i-
cc cc + + t=~= g~ --
-~ C + + t~ ---[w (~ lC
-
ccc ~t c5 ~ ee 00000
co--oc I
- - - - -_____
------ _ - _ _--__ ---__ ---- _ - _
ooocc
rr ~ c-~ E lt
C
-
~ C g L-
gt o + + ~~ ( L C- lN o -l
t l - -I
~ - L~ [ti e- [
l _ 1
- - ~_ C
gg -- -- ~ -0000-tr--- ____
----C __
- __ - _ --ggggg
- -
C
+ + -- --- -~_shy-----_ _ -----___ ooo~
-----__ ___ _ - - - _ - _ _- - - - ___ 0~000
--__ ---_ __ -- --c5 ooco
C cc
1 II II 11 11 II II
z
-
r ~ c
z
~ t lt i C z __ -- -~tt ~ middot~ ~
-middot middotmiddotmiddot _ - -----=___
_ =~ - _ -=shy ~
-- -~== 2~= ~ () V)
_middot ~ c( c(
0 er ~- ~ ~ lt 0 0___
-
- amp lt~ ~~~t8
F-21
) ) ) )
TOT1L AllE SCiNrmn TIIT1 L ll lbull1 tW FI 1lFll TIIL1 11bulli 1~lllbullll lllIIOT lIIJClION -11- OV TUI Mlbullrllll1Nft TOT1 I VOIIHIIbull 01- I Ill TOTmiddot I Imiddot I I 11-11~~ COi l NTlbulln
_LI__l_ ___l 4 bull P 7 bull P A R l n A 4 R 1 n 4 4 ___ 7 A ___ _RA____ 1 A 1A 8 1 SL__ 1 R 5 1 ~ 3 _1] ________ 111~____JJ__) 1 ~---~-~ 4 9 1 ----1___11
A0 1A4 JR 17 lAR A 41 4C 174 A ll~ JnR ll7 9 5 R12
----- --------- - ---- ---- -- middot-middot --- ---- -- ----middot-- ----middot ----~ 7 c ~ bull n 1 1 2 l 9 -s 1 -i s s ~ c c__ bull A____02 1 1 2 1 r n 1 c 0 r-- l ~ 1 o 1 0 1 t 1 o c- ----~--- -Pl Ol OP lA2 11A 61 c 0 ~ cq03 l 5 4 l 4 4 1 i r 7 -i r) c q 1 deg 1n
qA RO 107 ~_12_7 SA c 7 A04 _l _ s_ _________l 7 l RA____ 7 ______l _o 4 __ l nn_________ 0 _7________ R_l __ o ~ _____ J(__7 11__r ___ _t-__ o ____ ln ___ h_ ______ An____ c-_n __05 ]lA 140 170 7cA __2n1 qc- 7q 1n1
c 2c 7 lR 1 O 406 11 c A 0 4 n 3 4 7 l 4 R ~ 7 7 5
___ ___~ c Z bull A ~R_ ---2_h 2 bull bull 4 e R bull C07 l_ H l_ ri ------- R1 --- c ----------- AQ _middotbull 4_p____ r 7 ___1[_2_ _r--_______ _ A____________ )_3-____ _____ __t 9 ______ ~_____ _3 __ --~___()_______ __I__08 C7 lR9 9A 1n~ ll7 S AS 84 ~Q 33 11 11 1n R 3A 3309 7RJ 1R lA 14-1 17fI 77 4R 7(_
---~- L~ 1 7 -~ 1 7 4 1 l 1 9 S 3 9 C) 4 9 A ~ 9 q10 __ 7 c _ ___ _ 2 RA l A ____J__6_R 1 C 4 7 P C 1 rq I
-----~~a-R___-=4bull2_1 e __ middot-middot middot bulls __ - __ IL 9 1 ~_1_ ___3 7 40__19 1 ~~- gt q6_ (- l Q 144 43 _37 _ A f ---middot ~ bull r bull q ~ 9 IL ~ 1 _0 -~ bull 1 A 3420 l r -- I- I q4 l qo l c c 0
Valve-Reactor Heat Exchanger 200 009 2222 I I I 285 I 93 52
All measurement are for 100 CT streams
1--- w 0
air smicroace Saturation concentratiun at 20degc is amicroproxiriately 12 Measured
values were bY and 74 These values r~sulted from gas chromatographic
dn a I y s e s ot t rd n s t er re c1 sa111 p I e s fr() m t n e 1U U I i t e r Ted l a r b a y t i 11 e i nt e9r a t ed
sample ror tt1e tank car displdcement of 2008 ft 3 the total CT emitted based
on tile ctverctye is LUU8 x 007L = 144 ft 3bull
Tnen for a density of p =PMRT
P = (1)(1gt4) 64 gL = 04lbft 3
(082)(293)
Tl1erefore tne displacement weiyl1t of CT microer ()ff-load is equal to 144 ftJ x U4
l b ft J = ~ 7 bull 6 1 b bull F o r 2 5 U off - l o ad i n gs an n u c1 1I y we have 14 400 l b s bull Th i s
compares with UuPont I s rnectsure1nents of 9 3 vapor content arid 18800 lbyr
emissions Saturation vapor pressure concentration emission would yield
2400Ulo
l)reathing loss from the tank can be calculated from AP-42 as
Lt3(1bday) = 619 X 10-5 M( p ) 068 D 173 H 051 T 05F CKcmicro 147-P
5 0 68 173 = (619 X 10- )(154( 173 ) bull (155)
147-i73
(lj) 051 (26) 05 (1)(1)(1)
= S 63 lbday
3where a working range of zuu x 10 o 400 X 103 lbs were used to determine an
average l i quid nei grit Jn nu a 11 y e111 i s s ions would be 2057 lb Therefore
working losses dorrii nate tt1e totil elissions from tne microlant
Summary
The upper and lovJer bounos of caroori tetracriloride emission ~vere
determined as Ll467 lbyear (upper bound on fugitives DuPont measurement of
storage tank work i ny erni ss ions nur 1a I tanK oreattli ng) and lbgt 15 loyear
(lower bound on fuyitives s1 111eas--1re111ent of storage tank working emissions
no ri n a I tan k breatt1 i ny ) bull
13]
100
SOUfCE TESTS - STAUFFER CHEMICAL COMPANY
101 SITE OVtRVItw
Summary
Stauffer is located in the Carson area of Los Angeles County It is
a manufacturer of polyvinyl chloride from tl1e monomer (VCM) ll11hich is produced
on sit2 Stauffer is the only California producer of ethylene dichloride
(EUC) which is converted to VCM The VCM is regulated by several standards
including Cal OSHA EPA (emission standard) and CARB (ambient)
During Phase I the plant was inspected and possible emission
sources were identified Th~se cans~sted of EDC storage ta11ks fugitive
emissions from valves flanges and pumps process vat2r content gas
incinerator effluents and th~ loading of outbound tankers
Since the completion of Phase I several events have transpired which
a ffectebullj the test program to deterni ne foe i 1i ty emissions These wer2
~ the completion of a vent ~ldS incineration system tied to all significant EDC storage tdnk breathers
$ the completion of a comprehensive study by SAi staff to develop a nationwide material balance for EOC
bull inactivity in EOC importation for the plant and elimination of EDC exportation from the plant
Based 11pon this input and the plant inspection of February 3 1981
it was expected that a plant emission factor for EDC can be determined with
relatively little uncertainty It is further expected that si 1Jnificant
atmospheric release of EOC du~ to plant operation may not occur at the plant
itself but rather offsite These vmuld drise fro11 two sources - the process
water discharge from the plant and severcil off site EDC storaye tanks
Facilities Description
Ethylene dichloride is being produced at Stauffer by two processes
direct chlorination and oxychlorindtion In the former EDC is produced by
d i rec t ch I o r i nd t i on uf et 11y l t~ n e i n the pr 1~senc e of cHl Fe C 1 J crl t d lys t I n th e
latter process EUC is microroduced by the oxychl)rindtion of ethylene with
hydroyen chloride and oxyuen in tne prescrlCe of a cdtdlyst typically Cuc1 bull2
132
lhe microldrrt fluriti1~s tuC after tne microrimar y reactions by separation and digt ti l I dt i ur1 fgtr-ocJuLL 1 s ctured ctncl umiddot 2d ctS foed for VCM production All
sourclS agre~ tl1dt c1ission of EDC is (f concern in its productionstorage
process stages ctnJ O little importance in VCM production steps (J~B 1980)
At the timt of plant inspection some storage of EOC existed at three
ledsed storage tanks located at tne Port of Los lngeles 1-iaterial has not
been 1-Jithdravm from these tanks during the 1dst year and there had been
discussion to consolidate materictl into a single tank
All microrocess components handling EDC storage are now tied to a closed
ventilation incineration syste11 It is expected that virtually all
chlorinated hydrocarbons includiny EDC will be etfectively destroyed since the
system must demonstrdte VCM concentrations are reduced below 1 ppm Firebox
tempera tu re is aoodegF and residence ti me greater than 1 5 seconds under
heaviest flow conditions Note that by way of comparison combustion perforshy
mance datct for PC8s are 99995 destruction at 1832degF and 1 second residence
time and 9Y9Y9YY4 at 2 seconds dnd 2372degF (N Flynn SAi Personal
Communication)
Fugitive emissions from valves and flanges are monitored by the
plant according to the requirements of SCAQMD Rule 4661 Pumps and
compressor follow ~ule 466 Emission and control requirements for vinyl
chloride are specified by ~ules 1005 and 10051 EOC concentrations in
wastewater are monitored several times daily in tne primary EDC steam stripper
stream and once daily in the composite plant outflow stream Daily water
disct1arge limits are 2~ ppm wit11 typical montt1ly averages being in the 8 ppm
vicinity The discharge limit is ~mbodied in the discharge permit (number
5061) vJith tt1e Los Angeles County ~anitation District
Tnere are a number of points at which tt1e rnc can be released to the
dtmospnere and they are similar for both dir~ct chlorination and oxychlorinashy
tion These include the following along with their emissions factors for direct chlorination
A chlorinator vent 2xliJ- 5 IOdSS per unit mass EDC produced
I) I ight end column vent -52x lt)
C dist i I I at i un column vent lXlU-5
u storage tank breatning 7xlu- 5
133
E storaye tankworkiny loss 27x 0-u 5F fugitiv~ emission bxll -
-JLi ildSteJdter l gtX] U
Em~ssion factors are si~~lJ fer oxychlorination processes
Enlission factor vau-5 (~1rbulle Tdilt2~ frcn tne literature (JR13 198U) and
assume 9b iciency for incineration (A-E) It is also assumed above that
EIJC emissions to water are Lf~ uf ~rihsion to citir ard that in wastewater
treatment 100 of EOC discharged tc1 vater is rel easl~d to a i 1middot ~ These emission
factors 1bull1e1middote used to prioritize r-2ieaies iJut were cldrly uude approximations
to tre planL l)ecause of thlt ncine-xt~on system it was anticipated that
fdctors A-E would be ri2dtKcdft 1-middotactor - 1nii~t1t be reduced since a monitoring
program had been in force al20st o~e y2ar Factor G and the off site storage
tanks~ vni~cn are not tied into lt1n ircinerator system rniyht -iominate emissions
Emissmiddotion limits and concomitant regulations embodied in Rule l00S of
the SC-l)MU have necessitctted tlw incinera-cion system Ninety gas chronatograph
probes are located throughout t11e plant including the stack of the primary
incinerator Con cent rations of vc1v1 are reported es sent id I I y be I ow tt1e r~yu la -
tory limit of 10 ppm and in fact below the limit of detection somewhat less
tl1an 01 micropra
lUl MEASUK~MENT APPROACH
Tne objective of tt1e proyram is to determine a pl ant emissions for
EOC It is not accemicrotabl~ to dtmiddotvelop sucll informatmiddot1on basect upon published
indlllstry Jide esti11dtes of plant control efficiencivs Fortunately it was
possible to desiyn a monitoring and calculational progra1n to determine EUC
emissions for tne site and not ctbsorb a disproportionate share of program
resources
Hie re are tnre~ modes ot re I ease from tt12 micro Iant dnd d fourt11
offsite
bull Post-Process 1nci11er-dtion
Processes A-E of ecc ion lll 1 dre a 11 1ented into the pl ant
incinerdtio11 syste11 rt1e conceritrcttion of VCfmiddotl contir1uously iectsured in till=
stctck cts SJdS output remicroresents d rectsondble UjJper li11it to diJply tor EDC
con cent rat i ons s i nce i ts ef f i c i ency of i nc i nera t i on 1s d t 1 east c l1 u a I to tr1 at
134
ot VCibulll Tlerefore knowledge of the system dirflow rnd VCM concentrcttion are
tt1e necessary and sufficient conditions for determining the bounds of EDC
reledse fhe pldnt expn~sse I wi 11 inyne~)s to provide tnese data in order to
support the calculations
bull Fugitive Emisions - Valves Flanges Seals and Other Sources
Fugitive emission from the valves pumps and flanges can be
determined more precisely tnan was anticipated since Stauffer has completed a
comprehensive leak inventory of all such components in compliance with Rule
466 466 1 and 1005 The Inventory delineates all leaks uncovered by their
three man crew throughout tne year and indicates the screening level in ppmv
and component identity In consultation with Stauffer we will be able to
identify the total number of components associated with EDC handling systems
( 700) the distribution ot substance composition streams the distribution
and incidence of leaks by hardware component type and leak rate We will
utilize this information to provide historical data to compliment our Foxboro
OVA sampling at the site dased upon this monitoring we predict an EDC mass
emission rate for the fugitive releases The plant has agreed to provide the
necessary information We propose to meet witt plant personnel prior to the
start of monitoring and finlize the sdmpliny strategy to accomplish 100
coverage of the lines of hiuhest EDC composition The sampling approach and
calculation of mass emission are described in Section 72
bull ~~as tewater
From examination of the emission factors derived from published
literature (see Section 10~) it is clear that EDC release from wastewater to
air is potentially several 11rders of magnitude higher than any other plant
source l1ased upon microl ant ml~a sured concentrcJ ti ons of 8 ppm EDC in water a more
realistic emission factor wgtuld b 24xlo-4 1-iassunit mass EOC produced
Clearly this could still be the dominant source Furthermore the release
point would be expected to middotgte locctted between the plant and the sanitary
district treatment site whi1 his~ kmfrom the plant at 24501 S Figueroa We
believe it is necessary to ndemicroendently confirm the average 24 hour EDC
concentration in the dischar-qe valer by obt1ininy the refrigerated composite
sample We t1ave identified ttie I 1nes of int~rest and received agrt~ement to
sample and analyze for EDC in the stream ~e propose to draw a duplicate
sample from the compositor nd ani1lyze for its rnc content This will be
135
compared with Stauffers para11eI tnalysis It is not reliable to obtain
direct readings of EDC by survey instrument in the air above the water flow since concentrations at crny single point will be i 1 the near dmbient range
Emissions will be cc1lcJLtecl 1_sing the plants vulumetric daily flm
and ass~m~ng that complete degass40~ of the effluent will eventually occur
This assumption is based umicroon the relative volatility uf EDC and tne distances
involved However it should be noted that no direct experimental or
monitoring data is available with which to confirn this Th2 composition of
the discharge stream is unknomiddotwn since it merges into a larqe multisource flow
Conversation with the LA County Sanitation l)istrict (J f1ilne) reveals the
stream to be both exposed and covered Vents exist where me measurements
could De made to assess gross leaka~e at key points
We will obtain samp1es of several plant process discharge streams in
40 ml bottles witn no head space Transit time fr0m sample collection to
analysis will be minimized and wi l I not exceed 24 nours This time frame is
conservat i v e al though ED C i s a v o l ct ti l e mater i al ar1 ct vJi 11 undergo
concentration degradation and outgdssing lnalysi) will b(~ performed in tt1e
SAI Trace Environmental Chemistry Laborcttory utilizing pur 1Je and trap analysis
FID gas chromatography A trial a11alysmiddotis was conducted and the EOC
characteristic peak was distinctive down to the ppb level Therefore the
analysis should easily confirm concentrations in t1e 8 ppm range
Offsite Storage ranks
Three EOC leased storage tanks are located offsite at the Port
of Los Angeles and are owned by annther firm Annual emissions from the tanks
were very roughly estimated based un AP-42 JRB 1980) as 44 kkgyear
based on preliminary estimates of stored quantites
Considering tne magnitude of this source these storage tanks must
be investigated We will gather information about their configuration and
control i n order to ca I cu l ate their e111 i s s i on s bull
136
lU3 UEfERM I NAmiddot 10~~ OF MISS IONS
1031 Incinerator
Vinyl chl 11ride co11centration is monitored dt approximately 90
locations throughoul the pl trlt (Lanyner 19Bl) including the incinerator
output Conc~ntrations are reported by Stauffer dS less than 0l ppm at a
flow rate between 10000 ant 12000 SCFM Although no direct monitoring of
EDC is conducted it is posible to conservatively bound the concentration of
EDC at U 1 pp111 Tl1at conce 1ltrat ion is a suitable ci10i ce s i nee it represents a
conservative bound on tile VCM levels measured near the stack output
Furthermore the molar volume will be taken as 224 L rather than the higher
value it would nave because of the sl i~thtly elevated temperature at the
detector location
Using 12000 SCFM one has thE annual emission of EDC as 12 x 103
SCFM x 283 LSCF x 526x105 minyr x lmole x 97gmole x 10-7vv 224L
Therefore incinerator emissions of EOG are thought to be bounded by 773 kg or
170 1byear
10J2 Fugitive Emissions
Seven hundred (7UlJ) sources were surveyed comprising nearly 100 of
EDC service All accessibl~ plant areas with streams containing greater than
1 EUC were screened except for a small number of devices located in areas
11lere active maintenance was being condJCted It is believed unnecessary to
p2rf-gtrm any emission factor adjustment since it is estimated that gredter than
95 of the requisite components were screened
Three leaks above an arbitrary OVA reading threshold of 20 ppm were
parameters and mass emission numbers fhe upper bound leak rate was
determined to be close to twice the nominal leak rate which accounted for the
response fltlctor of amicroproximamiddotely 05 by Radirn for TLV detection of EDC
( 13 rown 1980) bull
Stauffer found and reported approximately 10 leaks in the EDC
service during 9 months previous to the pLrnt testing Assuming no111inal leak
137
) )
Table 103-1
STAUFFER CHEMICAL MASS EMISSIONS PREDICTION FROM OVA SCREENING VALUES
OVA SAI Actual Upper Device Response Response Cone Source Leak Rate Bound Location (ppm) Rae tor (ppm) a b Se I Re Function lbyr lbyr
Gate Valve Unit P1559A
Valve Heavy Ends Column Unit 14439
Compressor Seal EDC Storage Unit Tl062
I-- w co
300 11 273 -035 096 0 17 153 D 6
11 I500 114 439 II 241 C 1
r III II II o-is lUI bull2000 vl 1 1818 A 9~
All emissions 100 EDC (12-Dichloroethane)
11
values in the nei~hborhood of 2000 lbs~ total could be emitted annually
Therefore it appears ttiat major reductions in the mass emissions were achieved
by the company run inspection prugram and the fugitive emission source has now
become of seconqary importance as a fraction of total plant emissions
lUJJ Wastewater Discharge
Wastewater samples were collected in duplicate from four sites
within the plant for determination of ethylene dichloride (EDC) concentration
The samples were collected in EPP standard 40 ml VOA vials on July 1 1981
Analyses were performed using standard purge and trap techniques coupled with flame ionization detection gas chromatography The results are given in the
table below
Sample description EDC conc~ntration range (microJml)
EDC concentration average ( 1-19ml)
Approximate fl ow ( gpm)
PVC Interceptor Box 82 - 108 95 200
EDC Stripper number Chlorination area C1404
2
011 - 014 012 25
Final collection site for pH adjustmentP663 349 - 3B2 366 350
Final discharge site sanitary sewer 6 - 254 158 500
The water in the PVC lnterc1~ptor ~ox was warm and represents washings from the
PVC reactor which travels t 1J the Interceptor Box in a concrete drainage ditch
Water from the EDC ctripper was vPry t10t and because of its heat was difficult to collect The ~1eat ot trie wat11 r may in pdrt account for tne
relatively low concentratio11 of ElC here as the EDC would out-gas from the hot
~vater more readily The fir1ctl collection site tor pH adjustment is a large
concrete container middotvith mixers in it located just prior to the final discharge
site Water at the final dischdrg~ site is being constantly aerated due to
the speed that it flows thru~yn the concrete drainage trough This aeration
could account for tne relatively large range in the rnc conotent found here
The average EDC concentraticmiddot1 at tne final discharge site is middotbelow the daily
discharye requirement of 25 19ml EDC
139
Plant oersonnel indicdted monthly iverage readings are typically on
the order of 8 ppm but daily averages can re 1ch greater than 20 microgml In
addition LA County Sanitation Uistrict staff (J Milne 1981) indicated that
an on-site impoundment pond can become laden with EOC during certain abnormal
operating periods and discharge Vdriances arl requested by Stauffer This
mi 9t1t Jccur on the order of O1c0 aCii ecH c1 1d t1erefore is no-c expected to
signifcantly impact averag~ cnnJJI discharg ~ values lt s not expected at
this t1ime tEit evaporative eriss uns from this pond are significant except
infrequently during upset conditions and spi 11 control operations Program
staff ~~ 1er2 una111Jare of any periods ~h 1~n EDC cimtent in the ponds could be
appreciable and therefore no sampling was performed
Utilizing the average uf the two final discharge concentration
1readinJs (158 micrograrnsm~ 1 and 1G 1Ji gpm fl)W one has 34600 lbyear of EDC
released into the sanitary sEwer For the pJrposes of calculating population
exposures from plant releases it will be assJmed that the EDC is locally
emitted from the wastewater streams Tl1ere ire no monitoring data available
with which to develop a more accurate release profile
However emissions of me fron the plant wastewater discharge can be
calculated according to the method of Mcmiddotckay (197S) as modified by Dilling
(1977) It should be noted that this mLst be considered an estimate since
conditions of fl ow and the presence of other substances will influence
emission rates
Using Uilling (1977) for non~erated flow first recalculate
Henry 1 s law constant (dimensionless-my of chmiddot~mical per liter of air divided by
mg of chemical per liter of water) dS
1604 P M 1 Wl
TS l
)
wnere
H Henrys law constant dimensionlessl
Pi tile compounds pur1~ componen- vapor pressure in mm Hg at T
Mwi the 11olecular weiglt
T tile absol ~te temperature of he wastewater in K
Si tt1e compounds ~uluhility in myliter at T
Then for UJC
(l6u4)(7~)(9Y)H = = U0447 with 1
(2lJ4)(8690)
LIii bullul1Jl)i I 1 1y 1 middot UL i11 wt1Ler yiven -1L L0degC is 8u4JU UHI tile vapl)r pressure
1 4 psi ( 7 2mm) at 7 0 degF bull
Tl1e overall liquid mass-transfer coefficient Kil is given by Dilling (1975)
as
(2211)(06) in mhr
-f-0-4~ (M _) 12+ 100 Wl-H-
1
=U108111tH
Then from Mackay the percent desorption is given by
c 1 = exp (-Kil tL) where
c 0
c = the concentrd ti on at time t of EOC 1
co = tile initial concentration
L = the Ii quid depth (Ill)
t = tt1e retention time (in hr) of the liquid in the wastewater
system
Retention time is bc1s~d 11pon a flow velocity range of 3 ftsec as estimated by
the Los Angeles County Sdnitatii)middot1 Dibulltrict (1J Milne) over a middotdistance of
ctmicroprox i 1nate I y S km to the Joi n_t Wdter Po 11 ut ion Contra l Pl ant Sewer fl ow
depth throughout the entire route have been estimated by J Milne as typically
1 foot 030m) to the Davison Pump Plant and 3 feet (09m) to the Joint
Tredtment Plant distances assumed to bt~ 1 mile and 2 miles respectively
Trdnsit tiine l)eco111es between 05 tnd LU hours sequentially Then computing
the net emission reduction as the product of each leg
Therefore 26 of the EOC i erni t ted bdween the pl ant and the Joint Water
Pollution Control Plant Rtgtsfdence ti111e and conditions at the plant account
141
for additional releases and residudl content is forther emitted between the
plant cnd the ocean discharge poiiit ~-Herever the flow is r2tained aerated
or shallow emissions will be accentuated
Offsite Storage Tank Emissions
There are three storage tanks leased by Stauffer for EDC storage
1ocated at 22nd and Gaffey Sts ~ bull San Pedro These tan ks a ~e used for long
term storage rather than providing feed on a routine basis Therefore
breathing loss rather than v1orkir 1J lass is )f concern An tanks are white
two being 67 ft in diameter while t~2 third is 57 ft All are 40 ft 3 inches
high and have capacities of e~thi l0~10000 (2) or 840000 gallons The
tanks are cone roof type and are not tied into vapor recovery systems The
South Coast Air Quality r1dna~J~ment u~ str4 ct has based their estimate of
emissions on the specification of 12 foot vapor level Using the AP-42
formula for breathing loss with the vapor pressure of EDC at 20degc and an
average diurnal temperature variation of 26degF one hds for each of the two
larger tanks
l) 173 H bull51 T bull 5 LlI = (619 x 10-S) M( _P_ tH F CKcp 147-P
68 173 51 5 = (619 X 10-S) (9119~ 116 bull (67) ( I 5) (26) (1) (1)
14 7-1 16)
Thus for the two larger tanks LB= 472 lbday
For the third tank 0=57 and = 178 lbdayL8 The total emissions become 23724 lbyear ~ince he plant is in the process
of shutdown it is not clear what the liquid levels currently are Note that
if the material in the three tanks were combined into one totamiddot1 -~1nissions
would be reduced markedly Exposure to a population from this site was
determined separately since it is lbullgtcated over six miles fron tl1e plant
142
110
ASSESSMENT OF PUPULATlUN EXPUSUR~
111 OVEKVIEW
A major program goal wa~ to compare emissions from the various
sources and identify and rank any 11 tlot spots in California where the general
population was exposed to elevated concentrations of carcinogens A simple
Gaussian dispersion model was therefore used to obtain order-of-magnitude estimates of exposure of the genermiddotal population surrounding each source
Since this was essentially a screening study use of more sophisticated models
was not appropriate
It cannot be emphcsized too stronyly tnat most California residents
are exposed to emissions of hazardous substances from a variety of natural and
rnanmade sources Urban dwellers dre typically exposed to greater concentrashy
tions than rural residents however all are subjected to so called 11 background 11 levels from multiple sources In order to place stationary
source exposures in perspective the typical ambient levels of each substance
were identified from the literature and compdred with the concentrations due
to the emissions from each ~lant Exposures were thus expressed both as
absolute quantities and as increments above background
Comparison of plants presents a further difficulty in that various
substances are being consict~red No attempt was made to evaluate the relative
importance of exposure to two different substances such as chloroform versus carbon tetrachloride other than by ambient concentration
112 DATA SOURCES
1121 Meteorologicdl Llata
The dispersion model to be descritgted in Section 113 required input
of annual average wind speed and frequency of occurrence of wind from each
compass direction These data were obtained for most of the sites from the
South Codst Air wuality Management District (SCAQMD) In other cases (Kaiser
Jotms-Manvi l le l)ow and OuPlt)nt) only clai ly cJVerage wind speed and frequency
data were available In al I cases we used data from the meteorological
station nearest the modeled e11issrnn source Table 112-1 summarizes the
141
Table 112-1 METEOROLOGICAL DATA BASE
Observation Site Data Period of Pldnt (Distdnce Source Observation Remarks
from Plant)
A11 i ed Che11 i Ca 1 Lennox SCA~MD 1Y71-1974 Averaged hourly readings available ~ 1 Segundo (3 111iles)
S ff~r-- Uemical Long l3each SCA()MD 1962-1974 Averaged hourmiddoty leadir19s avai1able offsite Carson (3 miles) storage tanks appruxi1ntL~ly 5 miies from
Figure 114-3 Cumulative Population Exposure to Carcinogenic Polycyclic Aromatic Hydrocarbons From Coke Ovens at Kaiser Steel Corporation Steel Mill Fontana
CV)
0 -I X-V) V) a _J
s 0
C 0 micro co s micro C a u
l ~
14
1 --
J- t_bulli
_ C
uw m 0
1 -0 a bullbull micro
microdeg ()
0 micro tbullmiddot C QJ V)
0
X LLJ
C
middot4middot~middot C 0 micro rt1
-- - 0 ~ 0
0
n __ _______bull~ -______ --al_ - ~_
I bull
I I - bull
middot 1
-~1
l i l ~l
~ 0 I 1
Li bull I t_~I
i bull -1 _1-J
0fmiddotbullmiddot-middot+-middott-t-r+-t-+-+middotr+-t--+-1--t+ t f-t-J-t--r-+tmiddotmiddot~~bull+-t--1~~bull-+-1 Id middot t 4 E1 (
I ncrementa 1 Concentration (~gm3)
Figure 11 4-4 Cumulative Population Exposure to Asbestos From Johns-Manville Plant Stockton
I
)
--~ ~w t _ ---~- ----------~ --____~-=~ M
~
bull I bull I I I I ii i Q t I i E e- I Ii tl I I I 6 I a I I J I f t t bull I G fl I
a I e 1 1 a 1 1 1 1 iii bull bull 1 1 bull 1 1 bull 1 1 bull bull bull bull 1 1i bull bull 1 1 bull a bull 1 1 r ~ middot u ~ ~ I0
+H~ ~ ~~~~-~~~~-rmiddot-~~~~~~~~~-~--~~-~~-~t~~~~~-~J C
~1_ c bull 1 t~ bullbull bullabmiddot 11 l bullmiddot II bullbull C- I u~
0 rl
gtlt I I I- 20 () ()
CJ -1 ~_J
J Cbull s
I If I I I I0
C bull I Ii I I I I I I I I0 1E
bull-+l I t I I ro s Lower+l
I I I G I iii I bull
() 14 I I IC
u I0 I I II Iu
C
1-_~I fi
CJ ~ I-- (j) CJ I G I I p +l
ro +-l I I I1~3~V)
0 +) UDf)Pr u I I 11 I I
CJ 1iII
Vi I 0 I I It- I a X
0 ~
w I I I I bull
Lower C 0
middotr-
- 4+)
(1j
J Q_
0 bull1
ij a 3- ti I
D d I I I I
I I I bull
II B I t
i ] i amp I
1 ei ~ amp 3 ii
I I I I
e ft a a I bull G
I I I bull I
I I bull I I t If
UpPrf
I I
I I I-_L~ lI I I f
I I I I I
- I I
j I I II I -
lI ImiddotbullII
l I
l Li
Incremental Concentration gm3)
Figure 114-5 Cumulative Population Exposure to Perchloroethylene and Carbon Tetrachloride From Dow Pittsburg (Solid Lines) and to Carbon Tetrachloride From DuPont Antioch (Xs) Results For Upper Lower Bounds on Emission Estimates Are Shown
1i-1 - middotbullth~middot t =lttF--i-+-middotr+-r+middot~middoti-middot-~-4-4middoti t -~--is -~middot-rmiddotibull--+ u ~~ i--t4~ t-bull~~311 li-i1middoti- C - JJ _ CL
-- I 1 t middot1 c7 Ibull ) ~ I-middot bulli C bull 1-_ 11 11gt1l11 ~bull -a bull-bullbull t a- II bullbullbull bullI lt_I Iii _- I
Incremehtal Concentration (microgm 3)
Figure 114-7 Cumulative Population Exposure to Carbon Tetrachloride From Allied Chemical El Segundo (Curves Corresond to Upper and Lower Bounds on Emission Rate Estimate)
--- I
~__~-unt2wbull--o1WCUatJ41ia1
16 0amp (V)
0 ~
gtlt
14~11l 1l QJ _J
~
1 middot-0
c t1middotmiddotmiddot 0
micro ro
_ c micro ~ 1euro1pound1
0- QJ u-J c 0 u O =-middot ~-= QJ
c_1 micro ro +- V)
0 micro ~ t1 O QJ 1l 0 a 40gtlt w c 0
-
ro --
micro
20middot l ---a I ~ I a 0
0
pound1r-t k
Figure 114-8
8tWI-S~1111111N~-
a I __ I wi~ i awLi--~~ 2 I I
1--+ I ~--if bull =~~--bullJ-+--f-1-----middot+--+middot-middot--r9c- 1 c- J
bull _ bull -~
Incremental Concentration (microgm 3)
Cumulative Population Exposure to Carbon Tetrachloride and Chlorofonn From Allied Chemical El Segundo Assumingsix-months Operation With Each Feedstock (Curves Correspond to Upper and Lower Bounds on Emission Rate Estimate)
i u---J--bull-+ i i--+ --4--+-imf-bullbullQ~-+-+-+~bullbull-4 171 i R = Fi middot1 1bull1 r1__ bullltaa ~a
Incre0ental Concentration (microgm 3)
Figure 114-9 Cumualtive Population Exposure to Carbon Tetrachloride From Allied Chemical El Segundo During Offloading From Midnight to 6 am middot-middotmiddot------middot--middot
--bull ft ampLA 1111-~lliaIN_WPl1A 1JIU_Mt91 MJ1WWlaquoPI-W1111~ --------__----Et1
M gt~middotI
0 -I
-X
~ t~ bullM
V) middot-middot ~ V)
QJ _J
f 0
0 -
C -4 0middot~middotmicro re f micro C a u
10 Cdeg u 0 3 0middotmiddotmiddotbull~ -0 a
+-gt ro
+J V)
+-gt 0
2E1-0 QJ V)
0 a X
LLJ
C 0 1pound1--middot+-gt ro
-- Q_ 0 0
I I
~ -~middot bull I -bull-_ bull I
middot lbullgtbull___I I
frac14 llk I
I bull
-~
I
-1_ I
Ibull II__ JI
I~
(1 f---1-+ bull-r_--f bull middot~-bull +-- bull i -bull 1 -1-+--+bull-+-t-+bull--t1 ~ _ 4middot 6 E 1t1 1middot2
1 3 5 I middot~ _ 11
Incremental Concentration (ngm3)
Figure 11 4-10 Cumu1ative Population Exposure to Ethylene Dichloride From Stauffer Plant Carson
_~ ~ t M
C X E k1~
-
- c- 1-1 ~ C middot-middot -_J
I--- -J 4t1middotbullmiddotmiddota
D
-+-shy --middot liiit1J
r ~
ltL
t-1 _ V
0 -+-1
middotr -bull ~1
X
C
+- bull1 ~1 atr l -=
~~lll(k~IHmiddot~~-Wa~middot~nL1middot~~~ta KF u n P middot~dE-tveRS ~-tl~~~-u-~~~tmiddot_-~~~~liil-~-~~~ tc r
I bull
bull -r~l-yen--~1l~a l fl __ A4o bullbull bullbullbull ~ 111bull--middotbull
8 NJ industrial sites 28 samples in industrial areas with benzene consumption facilities Torrance CA Tuscaloosa AL National Forest 1000 sarip Ies-Torontt) Azusa CA El Monte CA Long Beach CA Upland CA Los Angeles CA Riverside CA
bull Assumes all year operation on either feedstock If plant operates 50 on each feed the population exposed to greater than 50 increment over background goes to 16025 - 19377 and 4044 - 16025 for 100
Ambient concentrations of benzene vdry over one order of magnitude in the literature and therefore make tnis calculation questionable For Kaiser therefore the carcinogenic PAH assessment was used to evaluate incremental population exposure above background
+ The partition of emissions from Oow are 78 carbon tetrachloride and 22 perchloroethylene by weight
173
1142 Comparison of Incremental and l3ackground Concentratiors
Those sites whicn elevate background concentrations greater than SO~~
City Kentucky 6600 The latter two are associated wi t11 EOC pl ants Data
taken in senFice s~ations ano lrctffic areas ir various cities range from -~
300-3640 ngm~ As previous~J mentioned the Stauffer plant is discontinuing
operations These two sources should be examined as part of any possible
start--up permitting activity
11422 Kaiser Steel Corporation
The Kaiser steel plant polycyclic aromatic hydrocarbon (PAH)
emissions cffmiddotise from the coke oven omicroerations Comfdrison of the cumulative
population exposure Figure 114-3 and Table 114-4 with the specified
background levels for urban areas illustrates the breadth of the exposure
distribution The five known PAH carcinogens that were isolated in the coke
oven emissions were quantified in tile ambient air of Los Angeles by Gordon
1Y76
Benzo[a~pyrene 046 ngm J
t3enzo~e]pyrene osio Benzaanthracene U lo
Chrysene U60
lndeno Cl23-cd~pyrene 134
Althouyn these concentrations ctr~ low co1npared with 1 number of other cities
cited in the literature and repres2nt a very I imited data base tt1e predicted
concentrations from the Kaiser plant ~enerally exceed these levels by d
174
significant margin ie greater t11an 30000 are predict2d to be exposed to
4redter than l()-lt t 1tbull 1111tlit~nt background of 3j nqm 3 It ic 1 iklly tlat
lHrllerte exmicroo~urt 11 Ilie drld dre also 11evatPd ovr r11nbient I1owever since
backyround concer1tr1tions of t)enzlmiddotne show large variation no population
calculation was specified
11 4 bull 2 bull 3 Dmv
Carl1on tetrdcnloride releases from Dow constitute approximately 78
of tl1e total CT plus perc emissions in Table 114-6 and were found to elevate
urban background concentrations greater than 50 in five census tracts Emissions are predominately from storage and check tank working and brething remiddot1 eases
1142~4 Allied Chemical
The Al lied plant was modeled several ways since the plant can
operrttt~ wiUl chlorotor111 dr cc1rbon tetrctlt hloridt fo(~d The c1bull-i pr~middot11ntlid in
ldlgtlt~o ll4- drld llil-B represent dflrlUil 01Hrdtion with eitt1er teed Partial
yedr operation with each feed can be scaled from the individual annual modes
and is presented for equal hltllf year operation in Table 114-9 As with the
Du Pont plant carbon tetrachloride emissions arise from feed tank working
loss fable 114-10 illustrates peak exposures predicted to arise during an
off loading cycle As expected concentrations in that six hour period far
~xceed annual average values Insufficient data were available to contrast
levels with backyround transient concentrations
175
l~O
AUHLAlHJ~ CONTROL n CHNil)UES
Alternative control approacnes for the most significant emission
sources dinmig ~he various pl1ants n ees fitie(J n the follo11 ric suJsections
EmpiiasVi nas Deen placed in dedlinq ~riU1 Uiose sources of 91reatest absolute
inaynHud2 (lbyr) and those const1 ut 0ing trie largest increm121middot1t middot] bulli~he
bdc kgrnund con cent ration of Hie em~ tted subs tcince i~o ctt tention Wi 11 be ~i ven
to the s~iondc1tY sources witt1middotn cK1 microant or to trlL case of J011ns-Manvi I le
since -UH imiddotajor source is less tran LOO 1byr and is not microrejicted to raise
bctckgro 1wd exposure levels to the ~~ntr21 population Furt12rmore a11
emission sources dt Kaiser Ste 1 1iC(1 v1 1~n directly dedlt -1ith in this
program r1re elltlteci ro t11e cc1lt1bull ovcmiddotr nperatmiddotions These faci 1ities are to be
closed du1rm and all primary reel 11i~ oH~rdtions discontinmiddot 2d Kaiser
forecasters have predicted further deterioration of the plant economics and
the phased closure has been accelerdted for primary steel making operations
Note that this closure is essentially irreversible cince differential
exmicroansion and contraction of the coke oven structur1 occurs in the cooliny
process and it would be improbable that ovens could be reheated without
extensive reuuildiny at major expense
12l STORAGE TANK EMISSIONS - fAUFFrn IHJW OUPUNf ANU ALLIED
At c1ll four sites emissions from storage tanks constitute either the
primary or near dominant (Stauffer) source of carcinogen release andor
genero1i iJOpulation exposure Curreritly the tanks of interest at each site are
permitted by the local Air Quality IJistricts hJwever they do not require
emission ontrol systems for vdrious reasons Tl1e estimated releases grounds
for exemption and other pertinent information are given in Table 121-1
In order to amicromicroreciate tth~ practicdl dlternatives for emission
controls the provisions of SCAWMU Rile 463 ar~ listed below which specify the
acceptable alternatives for tdnks r4uirin9 controls ie tanks 11aving
capacities greater than 39630 yal with suostances of true vapor pressure
176
Table 121-1
Site
Stduffer Uff-Site Tanks (J) (Sdn Pedro)
Allied (El Seyunrlo) - KdJ
1bulllctLer1 d I reld iturctjc
-J -J
Dow (Pittsbury)-Product Check Tanks (4)
Uow - microroduct sturag~
DuPont (Antioch) - Raw Materidl Feed Stordge
Approx Capacity~
6lUS x 1g (c) 84 X 10 (1)
lt3Y630
18000 each
70000L) 3UUOUO
30000 working 42636 liquid full
STORAGE TANK
Substance
Ethylene Uichloride
Carbon Tetrctci1 I Jr i J1
Perchloreshythylene and Carbon Tetrashychloride
CT perc
Carbon Tetracr1 l ori de
SUMMARIES
Emission Mode
Tank breattli ng
Tank v10rk i ng Tank breatttiny
PERC-~JOrk i ng PERC-b r2a t ~ i ng CT-working CT-breattling
Tank breathing Tank breatt1i ng
Tank workin9 Tank breathing
Vamicroor Pressure
014 at 70 F psi
1 63 psi at 68 F
16d psi at 68 F(CT) UJ9 psi at 68 F(perc)
168 i)Si dt 68 F
Grounds for Exemption
Exempt from control re4uirements to SCAQNJ Rule 463 since vapor microressur-2 dues nut iXC~ec
l~ io at sturdJ~ conditions
txe11pt TlJil fule -+b3 since tank capaL ~J ~ s u~der 39630 gal (lSO 11 )
Exempt from Regulation Rule v~~vu u~~~~~shytank capacity is under 39630 ga 1 andor substances are not classified as organic 1i4uicts
Exempt f ro1n Ru l ~ 85300 because subsLctnce is not classified as dn organic li4uid accordin to Regulation-Rule 8120
exceeding l~ microsi at storage conditions
~ floating roof tanks
~ fixed roof tanks with an internal floating type cover
bull a vapor recovery system with vapor collection and retllrn (or disposal) proce5sinJ lXctbullerlin~ 95 efficiency
ln a cllemicai plant a typical vapor recovery systein (KR Evans
SCAQMU) migm consist of a collection inanifold to d recovery system (suctl as a
vapor sphere) to a compression syst2m dnd subsequently to absorption and
recovery systems The absorpt~on st~n c~nsisting possibly of (rubber
stripper or activated carbon
In dealing with speciti( carcinogenic substances such as vinyl
chloride hiyhly efficient vapor control system perf0rmance has sometimes been
stipulated necessitating incineration systems
Fer th~ cases of corccrn realistic alternltltives are as follows
Stauffer Off-Site TanKs - Tnere are a number of options which can 1~arkedly 3
reduce ei1issicns from these tanks from th~ calculaL~d value of over 20 x 10
lbyr One alternative is to consolidat~ the material in tne three tanks
One of the large tdnks can contain the currently stured material and reduce 3emissions to apmicroroximately 13 0 x 10 lbyr Anotl1er alternative is to
transfer dll E0C to a single floating root tank Various types of such tanks
exist dlld are reviewed in the EPA report Lrganic Cht~mical Manufacturing Volume
J Storage Fugitive dnd Secondary Sources EPA-450J-80-025 eg internal and
external floating roofs and a variety of seal desiyn configurcttions Generalshy
ly such designs would be expected to reduce emissions to the order of
one-fourth to one-fifth the current level Cost of 1nstalliny ct contact
single seal internal floating roof wds estimated by BB Lumquist Pentrex
for EPA as $27770 in 1Y8U dollars for 7U ft diameter tdnk Cost to build an
external floating roof tank was estimated by G Stilt of Pittsburg Oes Moines
for EPA as ~14UUUO for D=67 H=4Uft Anottler com111on approach is to utilize
cdroon ctdsorption This works we11 witn nonpolar hydrocarbons dS VvC is
removed from the vapor phase A basic syte111 consists ot tvo carbon beds and
a regeneration system Regeneration is typically performed with steam or
vacuum Figure 121-1 illustrattS rhe systems (Basd2kis 1~110) Steam raises
tne VOC vctpor microressure The resulting sti~am-VOC mixture is condensed and
173
AC-TJA~D CA-e0-l
-----~---
LOW- ffiESiJ~C ~T~H----
PiltOCESS COOL-J~ At-JO oLOWER DRY t-J~ 2gtLOWE~
COgtJDENSER
J
COOLIN(a - --------
WATi=-A
RECOVeRE SOLVENT
WASTE WATER
Fi l 1 1 middot gure middot - Activated-c arbon Adsorption System
179
rout2d to c ~epordtor decanted and returned to storage In vacuum regenerashy
tion VOC vapor is desorbed by pu1 ling a vact1um on the carbon bed then condens-
ed a1d returned System efficiency ~ ~ est ii1a ted at 96 f(duc middot i m from fixed
roof levels (EPA 4503-80-025)
H~fri 9ctated vent concicr1sers dre une of the most cu1mon emission
reduction prJcesses -for conttol ~ ng hx2-d-- storage tai vUC Figure
12 1-- 2 i l 1ust rates a u n i t ( Er i scmiddot n 19HU ) eff i c i en cy of middot middot~ ~ c middotlt middotmiddot _y i s rate ct
bet1Jee~middoti 60-90 for the vapor pressure range of concern For such a large tank
the capital costs would be high ~igure 121-3 illustrat~s EPA estimates
(EPA l ~~SOb) for the condenser section Condenser sys tern a ea mu l d be in
excess of 1000 ftL
It should be noted t -at Jtc1 uf fer Uiemi cc1 l has arH1ounced the c 1 osu re
of its VCPVC plart PrevfoJsly tt12) had planned to purge tlle off-site tanks
of EUC Since any future po~sible micro1ant stdrt-umicro will necessitate a compreshy
hensive SCAQMD review this ltoument can assist in evaluating proposed control
measures
Allied and DuPont Feed Tanks - In bott1 of these cases the nore significant
quantity of emissions arise from tank workinltJ ie during the off-loading
activity rather than tank breathing Contnl ~easures taken for working
emissions are thus of primary concern Ther~fore no detailed discussion will
be provided on the alternatives for control gtf bredthiny emissions Additionshy
ally both tanks are in the range of 20U00 gallons which is a capacity where
floating roof tanks are almost nonexistent Out of 670 floating roof tanks
surveyd by EPA less then 15 were smaller than 30000 gallons in capacity
Therefore the utilization of any floating roof concept and seal combination
will not be considered
Commmerc i a11 y it is estimated by Ou Pont that carbon tetrachloride
feed costs approximately $017lb (E Taylor personal communication)
Ttierefore less than i2500 is lost to l)uPont annud l ly as a result of working
l eve l em i s s fo ns and 1es s fr om Al l i ed bull Thus i t is u n l i kel y that any
appreciable economic incentive exists to dev1~lop ci vapor recovery system
However a Cdndidate system could be pattern~d after the chloroform
feed-storage unit currently at Allied This is a dedicated vapor balance
system Chloroform is off-lndded from tank cars and stored in a closed
unvented tdnk As the 5tordye tank is filled the c1ir spdce d1spldced is fed
Figure 121-3 Installed Capital Cost vs Condenser Area for Various Materials of Cunstruction for a Complete Condenser Section
182
back to the tank car Also in this case the storage tank is not vented in its
normal breathing mode and is part of a closed feed system to the reactor
Vapor recovery system alternatives which are particularly effective
in loading and handling include r~frigeration adsorption andor absorption
Control efficienci~- 1re est1mateJ in the rlt1nge of 90-95 (EPA 1980b) and of
course depend on the speci tic su1)Stance and equipment used Carbon adsorpshy
tion systems were discussed above The smallest capacity carbon adsorptionmiddot
system priced by EPA (EPA 1980b) has 2 vertical beds of carbon (900 lbs - 4ft
diameter by 3 ft depth) witl1 an ilstalled cctpital cost of $135000 based on
December 1979 dollars
Dow Tanks - Dow has two pair of p~oduct check tanks which alternately are
filled dnd off-loaded Emhsions from th~sl tanks were calculated based upon
operating cycle (3 day fi 11) satubull-dtion vapor pressure and physicdl
chardcteristics Direct hec1d-spa1e testing was planned however it could not
be accomodated because of r 1stri c ed access due to unscheduled mai ntendnce on
the field test day
Dow has indicated that chey are studying the option of installing a
vapor controlrecovery systt~m in 1hese and their largerproduct storage tanks
The extent of their engi neri ng ind assessment work is unknown as are their
current plans It may be possibk to incorporate a vapor balance design into
the system whereby the disp1aced apor is trdnsferred to another process point
within tl1e system Alterna1ively a vapor r1middotcovery system such as carbon
adsmiddotorption is fedsible Hot-1ever since emissions for t11e check tanks are
primarily due to working loss tht~ use of a conservation vent or an adjustment
of its operational differential p~essure would be ineffective for tr1e
reduction of the bulk of such emissions Furthermore since check tank size
is re1at i ve 1y sma 11 no cons i derctt I on was given to conversion to a fl oat i ng
roof configuration for those tanks Conversion would be possible for
the large storage tanks
12~ WASTEWATE~ EMISSIONS - STAUFFER
Emissions of me throug11 ~-a~)te-1dter discharge are the largest single
source identified at the plctnt sites The discharge limit was set at 25 ppm
by the Los Angeles County Sc1nitation iistrict and was estdblished with the
occupational limit in mind of 50 ppm JVer an 8 hour work shift dt the District
18~
tredtrient center~ T11ere hctd been suine di~cuss1on between pdrti~s of possibly
raisiny tne discharge limit since it is felt by Stauffer that dilution of the
stream is sufficient to d 11 ow it Itmiddot shou Id he noted t10-1ever tna NIUSH has
recommended the permissible exposure limit be reduced to 5 ppm averaged over a
work period of 10 nours microer ddy 40 ours per Ieek witt1 a cei l iny level of 15
pmicroin dVeraged over a 15-mi nute microet~ v(J
It is not certain at 1~hrt rdte me 1rill be reledscd from the
wastewater stream At the microlant discharge points wastewater is both hot and
aerated thus favoring release~ No measurements have been ta~en downstream of
the plant after considerable dilut1on has occured The distance to the water
treatment pa11t is approximately 5 kin at 1i1ic~1 point anerobic diyestion is
conducted lt is presumed that a11 ELlC will be released before final ocean
di scnar~Ji
A possible emi ss i D~ cont ro process for nmiddotduct ion of the EDC in the
effluent is by ~recess adjust~Pnts or additio~s For example process
modifications to the EUC stripping stage coula dramatically reduce discharge
lev~ls A control alternative is the use of octivdted carnon or XA0-2 resin
to recover EOC in tne discharge stream Tests of Gulf South Researd1
Institute on XA0-2 and activated carbon (Coco 1980) show high recovery yields
for nonpolar organic carcinogens unaer a range of concentrations Viable
suggested alternatives oy Srnitn included regeneration of Ue trapping
materials dnu even consideration of burning tt1e concentrate carbon media (at
greater than 1000 pmicrom)
Control approaches to reduce emissions from so called secondary
sources in general and waste water emissions in particular fa 11 into four
categories - waste source control resour(e recovery alternative disposal and
add-on controls Alternative control pro(~esses which were considered but
appear to be inappropriate to this case i11clude chtmiddotmica means eg
neutralization precipitation coagulatior1 and chemical oxidation thermal
destruction of tne unconcentrated ~dste s1ream is ii1practical biological
treatment eg aerdtion and biomass-wasteater contcct ~ienerally relates to
the treatment of soluole degrddable organics in the concentration range
bet~~een U01 and l~~ terminal storagt eg lanctfilliny urface impoundment
and deep well injection dre either indppl 1cable or 1mmicroractical Therefore in
summary the poss i b I e contra l approac ties fur this ca e inc 1ude the improvement
r
184
of semicrodration efficiencies in steam stripmicroing tl1e internal recycle of ~aste
stredms ctnd the ddsorption b activdted carbon The design configuration
~fficiency and cost )bviously depend upon numerous microlarit specific factors and
their determination vJOuld rec-Jire detailed analyses
n1e ~~dsrewctter systbull-m of a model cl1emical production plant based
upon the average properties uf a composite of 30 chemicals was evaluated for
EPA by IT Enviroscience (EPA 1980b) Included prominantly among the 30 was
tUC with the hignest uncontrolled secondary emission wastewater release rate
ie ~ percent of tt1e production and 34 perc~nt of the emission Cost and
impact analyses were evaluated for alternative control systems to reduce
secondary voe emissions from wastewater Four systems were considered a
carbon adsorption system (eAS) for recovery of the voe from the wastewater a
cover to reduce secondary VOC emissions from the wastewater clarifier a cover
for the clarifier plus a carbon desorption system and a cover for the
clarifier plus ct eAS system usinu a fume incinerator The scale of the model
system was greatly in excess of the Stauffer plant thus further making
detailed comparison i mpract i cal bull However emission reduction factors ~-1ere 6diven as ~9 and cost effectiveness per 10 g reduction genera1middot1y ranged
between $450 to 1733 These factors would likely grossly underestimate the
system cost if scaled down to the range of the Stauffer plant ie the order
of 34600 lb annual discharge
The alternative approaches of improved steam stripping efficiency
and internal recycling of the stripper iischarge stream with a reduction in
makeup re4uirements could decreas~ net ~DC wastewater content release
123 CONTROLS FUR SECONDARY LEAD S~1tLTER STACK EMISSIONS
Given our findiny of low (16 kgyr) ~missions of arsenic from the
reverberatory furndce at RSR it middotJOuld ippear that the arsenic content of themiddot
lead feedstock is low andor that the microants system for reducing lead
emissions is dlso quite effective for arsenic ~SR it will be recalled from
Section 32 uses a quenchinu cnariber ctnd Dayiwuse filters to remove
pctrticuldte matter and a carbonate scruooer to remove sulfur dioxide
There are no federal new source microertormance standards for lead
eini ssions per se t10wever the Standdrds of Pbullrformance for Secondary Lead
SmeHers (40 CFR 6Ull2) limit total particuLte emission from a blast furnace
185
or reverberatory furnace to SU mgm 3 According to Augenstein et al (1978)
who reviewed the techno1ogy for control ing lead ~missions from ttwse sources
baghouse filters or wet scrubbers are yenerally used to contro1 particulate
emissions When fabric filters are used to control blast furnace emissions
they are nurma11y preceded by an afterburner which inciner~tes hydrocarbons
that bull~GU Id otnerwi se b 1 ind t ne f o1 ur-i c After-burners a re nor necessary for
reverb2middotmiddotatcry furnace emiss-ion smiddotnce the excess air and temperature are
usually sufficient to oxidize tn2 hydrocarbons
According to Augenstein et ali 11 sr1aker--type baghouse filters are
the 1nost effective means of controlling ead fume emissions from secondary
furnace opErations 11 Collection etficiencies can exceed 99 oercent One
advantage to this control approact is that l~ad oxide dust can be recovered
easily cmd recycled in the smeller Flue gases must be coated to below 300 degF
for dacro~ bags and to below ~00 degF for fiberglass bags (Hi~h et al 1977)
Although wet scrubbers are effecti1e under some circumstances in
controlling lead emissions it is more difficult to recover the lead oxide for
recycling In addition sulfur dioxide presi~nt in the flue gases becomes
oxidized to sulfuric acid and can cause corr1sion problems For this reason
sodium carbonate or other basic reagents ar~ added to the scrubber solution
Although it concerns a yold smelter an approach described by
Marchant dnd Meek (1980) provides an exampli~ of an arsenic control
alternative which might be apmicrolicatgtle to secmdary lead smelters At the
Campbe1 1 Red lake Gold Smelter in Balmerton Ontarion Canada Smelter gases
are first passed throuyh a hot electrostatic precipitator (ESP) The ESP is
heated so that the arsenic (which is principally in the form of As ) remains2o3 gaseous and is not yet collected This exclusion of arsenic allo~~s the ESP to
recover particulate gold rnore easily The ESP exhaust is then quenched with
ambient air to condense the arsenic trioxide Baghouse filters then remove
the arsenic along with other particulate matter
124 CUNT~OLS FUR STEEL MILL EMISSIONS
Given the imminent and irreversibli~ cessdtion of coking activities
at Ka 15 er Stee 1 Corpora t i on a rev I e~ o r t e c Imo 1 o gi es for ( on t r o 1 l i ng
~missions from the coke ovens rnd he Clt1ke byproduct recovimiddotry pldnt WdS not
deemed to be necessary Tnis is t1e only primary steel mill in California
1Pi6
KEFUENCE)
Al lf~n Jr Ll~ llJBU1 J 111odel to 1middotstimite hi1drdous (bullmissions from coke oven Juurs pr~pdrelt1 DY 1-kSecircn Tr1dnye7nrffut fur U~ Environmental Protection Agency Uffi ce of Air ~ua 1i ty and Standards fltesearch Tri ang I e Park Nortn Cdrolina KTl17362-UL
Jl len Jr CC 1~8Ub tstination of chdr in(i emissions for by-product coke oven microrepared by fltesedrch Tr1anglt Insc1tute for US nvironmental Protection 1-yency Uf f ice of Air l)ua l i ty and Standards Ke search Triangle Park North Cdrolina RTI17362-0
Allen Jr CC 198Uc 11 Environmeritdl c1ssess1ent of coke by-product recovery plants Proceedings First Symposium on Iron c1nd Steel Pollution Abatement Technology Ctncayo Illrno1s--Tin1ctober - 1 November 1979 EPA-600-9-80-012 pp 7~-dd
Augenstein ur1 T Cornin et al 1978 Con_trol techniques for lead air emissions prepared by PEDCU Envircnmental Inc for US Environmental Protection Agency Research Triangle Park North Carolina EPA-4502-77-012-B ( NT IS P88U-197551)
jarber WC 1977 Interim air pcllution a~sessment for eight t1igh-volume industrial organic chemicals 11 US Environin~ntal Protection Agency memorandum to Air and Hazardous Materials Uivision Kegions I III-X and to Uirector Environmental middotPrograms Uivision Region II
t3drrett KE WL Margard et al 1977 Sampling and analysis of coke-oven door emissions Prepared by ~dttelle-Columbus Laboratories for US Env 1 ronmenta I Protection Agen y Industri a I Environmental Research Laboratory Research Triangle Park North Carolina EPA-b002-77-213
t3 a r ret t flt bull r bull and P bull K bull ~~ebb 1Y78 bull 11 E f f e ct i venes s of a wet el e ct r o stat i c precipitator for controlling PUM emissions from coke oven door leakage Presented at 71st Meetiny of the Air Pollution Control Association Houston Texas 25-29 June 1978
l3asdekis HS 1980 Carbon Adsorption Control Uevice Evaluation IT Enviroscience Inc report in prepdration for US Environmental Protection Agency ESEU
8ee fltW et al 1974 Coke oven charging emission control test program Vol I c1nd II US Environmental Protection Jgency EPA-6502-74-062
~eimer RG 1978 Air pollution emission test Analysis of polynuclear aromatic hydrocarbons from coke oven effluents Prepared by TRW Uefense and Smicrodce Systems Groumicro for US tnvirorunental Protection Agency Office of Air wuality Planning and Standards Kes~arch Triangle Park North Carolina EMB Kemicroort 8-CKU-12
187
Braman RS 1976 Application of the arsine evolution methods to environmental analyses presented at the International Conference on Environmental Arsenic Ft Lauderdale Florida 5-8 October
Bratina J E 1979 11 Fabric filter applications on coke oven pushing operations J Air Poll Cont Assoc 29(9) lll6-920
tiuonicore AJ 1980 Environmental assessment of coke quench towers in proceedings First Symposium on Iron and Steel Pollution Abltlternent Technology Chicago~ Uirnois JO October - f N~vernber 1979 EPA-6009-~1b-62~ pp 112-142
t3usse A D and dR Zimmerrao1 1973 User 1 s guide for tdegh~ Climatological Uisper~ion Model US Environmental Protection Agency Resea~ch Triangle Park Nortl Carolina) EPA-R4--73~0024
California Air Resources 8oard 1976 Coke oven emissions miscellaneous emissions and their control at Kaiser Steel Corporations Fontana Steel making facihty Enforcement l3ranch) S2crcmento California L amp E-76-11
Calvert~ CA 1976 Envir __ Sci_ i( __ Technol 10256
Coco L~i 2t aL~ 1979~ 11 02c10pmert of trec1tment and control technology for 11refractory pet rochemi co was tt=1s r middotl 1d l Report Gu 1f South Research In st it ute
to U S tnv i ronmenta l Protection Agency EPA-ti002- 79-080
Colucci~ LM and CR Begemi 1971 Palynuclear aromatic hydrocarbons and other pollutants in Los Angeles air In Proceedings of the International Clean Jir Congress Vol 2 Academic Press pp 28-35
Cooper JA and JG Watson Jr 1980 Receptor oriented methods of air partkulate source apportionment 11 J Air Pollution Control Assoc 30(10) 1116-1125
Coutant R W JS McNulty and RD Grammar 1975 Final report on determi 11at ion of trace elements in a combustion sys tern Prepared by Batte11 e Columbus Laboratories for the Electric Power Research Institute EPRI 122-1
11 11Dilling W 1977 lnterphase transfer processes Envir Sci TechnoL 11 4 pp 405-409
Oow1irgl) MP JO Jeffry and AH Laube 1978 11 Reduction of quench tower emissions Presented at the 71st Air Pollution Conotrol Association Conference Houston Texas 25-30 June APCA No 78-92
EPA 1980a Review of Criteria for Vapor Phase Hydrocarbons 11 US Environmental Protection Agency ~eport 6008-80-045
EPA 1980b Organic Chemical Manufacturing Volume J Storage Fugitive and Secondary Source EPA report 450J-80-025
EPA 1981 Compilation of Air Pol lutdnt Emission Fdctors (Including Suppleshyments 1-7) Third Edition Suppl1111ent 12 Utfice of ir l)uality Planning dnd Stdndlt1rds AP-42-ED-3-Supp I - lt
188
Erikson Uli 1~80 11 Control Uevice Evaluation Condensdtion 11 IT Envioscience lnc report in prepdration for US Environmental Protection Agency ESEU
lrtel liL 197Y Quench tower particuldte emissions J Air Poll Cont Assoc 2Y(9) 913-916
Ev a n s K bull P bull 1981 Pe r son a l Commun i cat i on v i th R bull Zi s k i n d bull
Goodwin DR 1980 11 Air pollution emissions standards in Proceedings First Symposium on Iron and Steel Pollution Abatement Technology Chicago 11 lrno1s 30 October - 1 November 1979 EPA-6009-80-012 pp 37-45
Gordon GE 10 Receptor Models Environ Sci Tech 14(7)792-800
Gordon KJ 1976 11 lJistribution of airborne polycyclic aromatic hydrocarbons throuyhout Los Angeles Envir Sci Technol 370-373
Great Lakes Cdrbon Corporation 1977 Study of coke side coke-oven emissions~ Vol 1 Source testing of a stationary coke side enclosure EPA-340l-77-014A
lirimsrud EP and HA Rasmussen 1975 Atmos Envir Y 1010
Hartnan MW 1~80 Source test c1t US Steel Clairton Coke Ovens Clairton P~nnsylvania Prepared by rRW Environmental Engrneerrny D1v1s1on for US lnv1ronrnental Protection Agency Emissions Standards and Engineering l)ivision Research Trianyle Pdrk North Carolina EM~ Report 78-CKU-13
Hendriks Ilt V et a I 1Y7Y Urgani c air emissions from coke quench towers 11
Presented at 2nd Arrnual M(etiny ot the ir IJol lution Control Associdtiont Cir1cinndti Utlio June 1~7lJ l-dmicroer NU 9-391
Higt1 MU ME Lukey dnd TA Li Puma llJ77 Inspection manual for secondary lead smelters premicrolt1red b Enyineeriny Science lnc for US lnvironmentdl Protection Ayency Office 0f ~nforc~nent EPA-3401-77-001
Holzer G et al 197 11 Collection amp analyses of trace organic emissions from natura I sources 11 Journal of Cnromatogra flhY 142 pp 755-764
Science Applications lnc Final lemicroort to US Environmental Protection Agency tPA-560lJ-80-UU2
J bull Mi l n e l lJ 81 Person a I Cornm un i cat i on wi t t1 bull Zi ski nd bull
Kemner WF 1979 Cost tffectiven2ss model for pollution control at coking facilities Premicroarect oy PEDCo Environmental lnc for US Environmental Protection Agency Industrial Environmental Research Laboratory Research Tridnyle Park Nortll Carolind EPA-GU02-7l-1U5 (NTS PB 8U-118706)
Kessler T AG Sharkey Jr and kA Friedel 1973 Analysis of trace ele111ents in coal by smicroark-source mass spectrometry US 8ureau of Mines Report ot I n v es t 1 gd t 1on s No bull 77 14 bull
189
Kolak NP J Hyde and R Forrest~r 19 1 9 Particulate source contributions in tt1e Niltlgara Frontierffi Prepared t)_y Nr2bulll York State IJepartrnent of Env i ronm~ntc i Con serv d ti on for US En vi r inment ct l Protect ion Agency Region 11 EPA-9024-79-006
Mackay U and PJ Leinonen llt175 Hate of Evamicroordcion of low - solubility contaminants from vJater bodies to c1tmospmre 11 Envir Sci Tecnnol 9 13 pp 1178-libU
Magee tiv HJ Hctll and GM Vc~rlt_a Jr 1973 Potential Doimiddotiutants in fossi1 f1Jel Prepared by ESSO fbullt~scarcr middotnd Engineeriny Co tor US Enviromihm~ct Protection Agency Ufl-R2-7 --2l1~L (NTIS Pt3 as u~9)
Marcriant middotG0H ard RL Meek 1~gu Eva uation of technology for control of arsenic tmissfons at the CdlmpJel Hd Lakmiddot G~ld Smelter prepared for the US
1Env i ronmentd l Protection Agenc)-l-l(lU-sfrT l lnv i ronmenta l kesedrCiI Laboratory Cincirrn2ti Uh70 lYA-60U2-8U--141 (NTIS 8 oU-21Y36J)
Margler L$ HA Ziskind M8 lcgozen (Ind M Axelrod 1979 11 An inventory of care i nog)n i c substances re i ea sect into ne ambient air of California Vol une I - Scn~enin~ and id2nt~fication o-f r~1rci1ogens of greatest concern Science J- pp 1i ca middott~ L J s In c bull Fi nd l kcmicro~ rt ~ i I - U 6 C) - U- 5 0 4 to U1 e Ca I i ~- o r n i a Ai r kesources 130ard
Milne J 198L Persond1 C~mmunicdtmiddotion middot1itt1 R Ziskind
llurchio JC WrC Cooper and A ue Leo11 1970 Asbestos fibers in ambient air of California 11 Cal Horn i ct Air lesourc2s l3oa rd ARl3 4-0S4- l EHS Report 73-2 (March)
National Academy of Sciences 1Y72 Part1culate polycyclic organic matter Washington~ DC pmicro 4-ll
Nati ona I Kesedrcn Counc i I 197b Arsenic prepi1 red by Subcommittee on Arsenic Co111mittee on 1bull1edical ctnd t)ioloyicai Effects of Environmental Pollutants washin~ton uc
Oliver JF and JT Lane 1979 11 Control of visible emissions at CFampl 1 s coke plant-Pueblo Colorado J Air PolL Cont Assoc 29(9) 920-925
Pe I I i z z a r i E bull 0 bull anct J bull E bull l$ u n c h 1s 19 1 nbi en t ai r ca r c i n o gen i c v a po r s Improved samplin~ and analysis tectrnology EPA Contract 68-02-2764
Pellizzari E U 1977 11 Tle 11easure11ent of carcinogenic vamicroors in ambient middotatmospheres EPA-b007-77-J55
Phelps RG 1Y79 lSI-CPA-oatLlle c-ilte oven door sealing proyram J Air Pol 1 Cont Assoc 29(J) JlJ8--JlL
Porter RA 1976 Uismicroersion ~yuation s lucions Dy calculator I guide for air pollution engineers ctnd sci12nthts S cond edition rexas Air Control tioard ~MT1S P~ 262 lJ~)
Redburn T 1980 Kaiser battles aginst l~gacy of woes 11 Los Angeles Timesmiddot (7 Semicroternber irno) Part VI pp 111
Research Triangle lnstittJte 1977 11 Qudntification of benzene in 15D -1bullnoient air samples 11 for US Environmental 11rotection Agency Office of Air Quality Planning a11d Stitndards
Rinkus SJ and MS Legator 1979 Chemical characterization of 465 known or suspected carcinogens and their correlation with mutagnic activity in the Salmonella typhimurium system Cancer Research 393289-3318
Koberts R M 1980 11 An inventory of arc i nogen i c substances r~ leased nto the ambient air of California Tasks 11 and IV KVB Final Report KVB 26900-836 to Science Applications Inc
Rogozen MB LW Margler P Mankiemiddott1icz cind M Axelrod 1978 Water pollution control for coal slurry pipeiines Prepared by Science Appl1cat10ns Inc for US Department of Energy (NTIS SAI-068-79-516
Rogozen MB DF Hausknecht and RA Ziskind 1976 Methodology for ranking elements in fossil fuels according to their potential health impact Science Applications Inc Final Report 260-77-539 to the Electric Power ~esearch Institute
Russel 1 JW and LA Shadoff The sdmpl ing and determination of halocarbons in ambient air usin~ ~rncentration on porous polymers
Siebert PC CA Craig and Eb Coffey 1978 Prelimary assessment of the sources control and population to airoorne polycyclic organic matter as indicated by benzo(d)pyrene Prepare( by lnergy and Environmental Analyses Inc for US Environmental Protection Agency Office of Air Quality Planning and Standards Resedrch Triangle Park North Carolina
Simmonds PG et amiddot1 1974 Distribution of atmospheric halocarbons in the air over Los Angeles basin Atmos Envir Vol 8 pp 209-216
Singh HB LJ Salas and RE Stiles Distribution of Selected Gaseous Oryanic Mutagens and Suspected Carcinoyens in Ambient Air in Proceedings Annual Meeting Air Pollution Controrl Association 82-651 June 20-25 1982 New Url eans
Smith WM 1979 Evaluation of cok oven emissions in Yearbook of the American Iron and Steel Institute pp 163-179
SRI International 1980 Asessment oi human exposures to atmospheric benzene SJ Mara and SS Lee finai report to US Environmental Protection Agency EPA-4503-78-031
Taback HJ 1978 Control uf Hydrocarbon emissions from stationary sources in the California South Codst Air Basir Prepared by KVB Inc for the Cal1fornid Air Resources Bod-rdmiddotKVB No 5804-714
191
Trenholm J-f dnd LL )eek 1Y78 11 Assessment of t1azardous organic emissions fror1 slot type coke oven Datteries unpublished paper US Environmental l)rotection Ayency Ernssion Standdrds and Engineeriny Jivision
Turner U~ J970 Workbook of atmusphe~ic dispersion estiimiddot1Jt2s IJS Environmental Protection Agency k~sedrch Tri1ngle Park North Carolina
US Bureau of Mines lYl f-inercls yearbook 197U Vol 11 p 4~3
Van Os dE I l U bull iJ bull lL Ma rs l and et ct 1Y7~1 o En i r onmt~ nta I a s s es smen t of co k e Dy-product recovery plants prepareu by ReseMcn Trictngle lnstitute for US Environmental Protection AgerKy 11cJustrial Environ11ental R~search Laboratory Kesearcll Tri2n~e Idrk NorU1 Caroline EPA-6U02-7J-Ul6 (NTfS PB 293-278)
Westbrool( C W 1979a Level l ass~ssment of uncontrolled sinter plant emissions Prepared by fesearch Triangle Institute for US Environmenta1 Protection Agcncy Industria1 tnvirnnmental fltesearc-1 Laboratory kesearch Trianyle Park North Carolind EPA-bOU2-7Y-112
Westbroollt C~~ 197ltJb Level l css~ssment of uncontro1led l)-BOP emissions Prepared by Researcn Triangle institute for US Environmental Protection Agency Industrial Environmental Hesearch Laboratory Research Triangle Park ~Orth Carolina EPA-6002-79-190
Wetherold rLb LP Provost and CD S111ith 198U Assessment of atmospneric emissions from microetroleu111 refinin~ Volume 3 Appendix l3 Radian Corp Fina Report to US tnv i ron111enta I Protection Agency EPA-600 2-80-07 5c
Ziskind RA OF S1ilith JL Hdhn etnd G Smicroivey (1980) Determinants of Cancer and Carctiovasculdr Uisease Mortality in Asbestos Mining counties of California SAi Report 068-dl-514 l May
192
APPENDIX A
Rapid Screening and Identification of Airborne Carcinogens of Greatest Concern in California
Lawrence W Margler Michael B Rogozen Richard A Ziskind and Robert Reynolds Science Applications Inc Los Angeles California
This paper describes a method for establishing a priority list of airborne carcinogens within a
state jurisdiction In this case it was necessary to identify from among hundreds of potential
candidates the live to ten materials of greatest potential concern In California as airborne
carcinogens
Because no previous Inventory of carcinogens In California existed published lists rankings
and assessments of national scope were used to Identify candidates By systematic manipulation
and comparison of these data sources 47 materials of some notoriety were chosen for closer
scrutiny This selection was pared to 22 candidates largely by eliminating those which had
very little production and use In California (Substances primarily used as pesticides were
excluded from the scope of this study) The remc1ining candidates were then ranked by additive
and multiplicative algorithms and by a panel of experts The results of these rankings were
combined to produce a single selection of 11 priority candidates In alphabetical order they
ethylene dichloride N-nllrosoamlncs perchloroethylene and polycyclk aromatic hydroiarbons
In continuing studies a baseline emissions inventory Is being prepared and a source testing
program is being designed
In recent years concern has grown over su~pected arcinug-ens New Jersey seshythe possibility that certain materials lected ten volatile organic compounds released to the atmosphere through inshy and five htbullavy metals to be examined dustrial and commercial activity may be further and is currently measuring responsible for a significant portion of ambient atmospheric concentrations of the incidence of cancer in the general these substances in a variety of areas In population This concern is manifested the second year of the study the state at the federal level in the National h1s incre1sed the volatile organics Emissiltm Standards for Hazardous Air stt1died to 20 and begun measuring Pollutants (NESHAP) which limit heavier orianics associated with parshyemissions of the known carcinogens asshy ticulatcs2 In California a very different bestos beryllium and vinyl chloride 1 approach-was taken
Only a few states including New Jersey and California have begun efshy Overview of Californias Approach forts to identify airborne carcinogens of concern to the general public for the The California Air Resources Board purpose of setting state emission regushy (( ARB) is sponsoring a three-stage lations for these substances After reshy st 11d~bull of airbornC rmissions of carcinoshyviewing national use data for known and g1middot11s from anthropogenic atlivities The
November 1979 Volume 29 No 11
first stage which is the subject of this paper was to identify roughly five to ten materials which of the hundreds of known or suspected airborne carcinoshygens are most likely to be of greatest concern to Californias general populashytion Also of interest were those which in order to satisfy occupational health and safety regulations might be transshyferred from the workplace air to the outside atmosphere The second stage which is now underway includes pinshypointing of emission sources for each of the carcinogens of importance quantishyfiltation of emissions nnd design of source-testing methods A subsequent stage will consist of source testing and measuring public exposures to those substances for which data are unavailshyable The basis for regulatory action if appropriate will include the results of this program and other related reshysearch
Screening of Candidate Carcinogens
The National Institute for Occupashytional Safety and Health (NIOSH) lists 1905 chemicals which have reported neoplastigenic or carcinogenic effects and 510 which have otherwise received attention for their neoplastigenic poshytential1 The need to select five to ten materials from such a large number of potential candidates dictated that we devise a way to rapidly eliminate from further consideration the vast majority of the substances Given the paucity of published data on most of the candidate substances the screening method was designed to make best use of readilymiddot
C1111yrich1 19i9-Air Pullulion Control A~~ociatlon
1153
A- I
----- ---- ----- ---------------------available information The screening process was as follows (1) eight comshyplications of known and suspected carshycinogens were reviewed and those subshystances which were not used in Califorshynia were highly unstable in air or were very doubtfully carcinogenic were eliminated (2) after more detailed inshyformation was obtained for the reshymaining 25 substances candidates were rated by two different analytical methshyods (3) an expert panel was convened to review dossiers on the candidates and to independeitly nmk them (-i) from the eight to deven substances ranked highest by all three approaches eleven were selected for the emission identifishycation and source-testing design stages of the CAHB effort
lnitiI $cre~ng ot Potential Candidate Carcingtgens
65 compnunds middotwe ire selected from 642 industrial ur~nnic air pollutants comshypiled by MrTRE Corporation for the US Envircmnental Protection Agency (USEPA) in U1at study pollutants were scored by multiplying four explicshyitly defined rating factors annual US production fraction of production lost to the environment volaflity and toxshyicity To adapt this york to our pmpose we first selected the 114 substances listed as being carcinogeni 1~ or neoplasshytigenic Then the scores of each of these compounds under the criteria annual US production fraction of producshytion lost volatility and carcinogeshynicity were multiplied together Seshylected for further consideration were those substances which had a product score above 50 Another 15 substances listed as being carcinogenic but lacking information for one of the other rating factors were also selected This list of 65 was then compar~d with seven other lists of corcinogeris1
-middot 11 Materials comshy
mon to the reduced MITRE list and at least one of the other lists were chosen for further consideration Added as candidates were those suhstances which are regulated as occupational carcino-
gens by the Occupational Safety and Health Administration (OSHA) and certain inorganic carcinogens Finally substances were added which in our judgment should be investigated but had been eliminated at this point Exshyamples of these are bis(chloromethyl)shyether epichlorohydrin and hydrazine
Next the refined list which now contained 47 substances or chemical groups was pared further hy another rapid screening process Eliminated
were all candidates (1) whose production andor use in California was very low (under 10~ lbyr) and wns not thought likely to pose a risk to a localized popushylation (2) which are very unstable in 1ir or CH whilh should not on t lw hnsis of CUtrltbullnL cidlnre liebull considtbullnmiddotd r11rci-
1154
Tahllt I Suh tances n-iewed in letail
Candidate Substances
Arsenic Inorganic lead Ashestos Alkyl lead Benzene Maieic anhydride
nogenic The result of the initial the rating under that criterion and the screening was a list of 22 candidate mashy corresponding criterion weight The terials which is presented in Table I overall rating for pollutant i is then the
sum of the scores under all the crishyteriaRanking Candidates by AddiUve ancll
Mulllpllcatlve AigorUhms The additive approach has several virtues the main one being that it forces
Many screening or ranking systems the user to make all assumptions exshyfaH into one oft wo cat~middotgories additive plicit In the process of setting up such and multiplicative Some systems are a a ranking system new insights into the combination of the two while others problem under consideration may be combine nn ol1jective appronc 11 with g1ined Once the system is set up it is suhjlctive cvnluition of the remiddot ults 1l rdatively easy to use Where data for The L2 substances surviving th1middot initial scoring pollutants are unavailable arshyscreening were ranked independtmiddotntly by tificial scales can be constructed to the tvo approaches If the same subshy quantify subjective factors Finally the stance rated high~middot under both systems sensitivity of the results to the systems its importance to California was judged subjective aspects may be measured For to be more likely than if it had scored example one can determine the effect of highly in only one method changing criteria weights upon the final
Additive Approach In the additive pollutant ranking Similarly an appreshyapproach the user identifies one or more ciation may be gained of the significance criteria and rates each alternatiw subshy of the range of uncertainty for a particshystance against each criterion while sishy ular required data element by varying _ multaneously deciding the relative imshy rating factor values portance of the criteria Eq (1) shows its A fundamental problem with the apshymathematical formulation proach is that there is no logical basis for
adding the individual scores assignedRating for pollutant i = f W1 R11 under the criteria other than the asshy
rbullI sumption that this simulates or even
(1) improves upon the users thought proshyEach criterion or rating factor (1 ) is crss A major operational problem is assined a value for each pollutant and that of weighting the criteria A common each rating factor is weighted (b W1 ) practice is to give all critNia equal -icnrding to its importance reht veto wltgtight hut that is in itstbulllf a st1Ubullm1middotnl lhe otlwr rritnia llw sc11rtbull for 1111 aLnut the 1rcl11tive importance of Ow 1n1 1 undmiddotr nitn1 n j is the prod11middott of n1ttria
Journal of the Air Pollution Control Association
A-2
---------
Multiplicative Approach In the multiplicative approach the rating for each alternative is the product of the rat ihgs under each criterion
Hating for pollutant i --- 11Ill
UJ (~) Jbull I
A multiplicative approach can have ~omc altlvrntnges over additive onrs First in some cases the ratings can be physical parameters such as concenshytrations or volatilities there is then no need to weight the criteria and hence less controversy over subjective judgshyments Second multiplication generally provides a wider range of scores than does addition allowing clearer disshycrimination among alternatives Finally this approach provides results which are more intuitively acceptable As an exshyample of this last point suppose that exposure and harmfulness levels for candidate substance are each converted to values on aO to IO scale and that a certain substance is both extremely toxic and extremely rare An additive approach would give the compound a rating ofO + 10 = 10 which is equivalent to that of a moderately prevalent subshystance (rated say at 5) which is modshyerately harmful (rated also at =-) A multiplicative approach on the other hand would rate the first substance at 0 and the second at 25
Criteria The six criteria used in the additive and multiplicative ranking procedures are defined in Table 11 Asshysignment of values to the Rij was based upon data gathered from published litshyerature personal communications and panel discussion and has been fully documented 13
Because the purpose of this exercise was to determine the relative imporshytance of the suspected candidate carshycinogens R 1 was scaled to the most heavily used candidate substance benshyzene whose annual production and use in California is nearly 109 lb Materials with a use under 105 lbyr would be rated zero for R 1 and rejected Howevshyerbefore rejecting a substance by this criterion we considered whether its emissions could in particular circumshystances result in high exposures to a Joshycalized population
R2 takes into account the fact that the chemical industry is in continual change Substances of concern today may be phased out while the use of others may rise dramatically increasing their imshyporta nee asmiddot poll utan ts Information on developments which could likely result in a change in the growth rate was facshytored into the choice of a value for R2- As an example asbestos consumption in California has been stable in recent years However the pending phaseout of asbestos in motor vehicle friction materials will hasten the decline in asshybestos consumption hence we assigned a value of 1 for R2bull
November 1979 Volume 29 No 11
Ideally one cou1 1 use pouu1c11u
sion as a criterion However in this case emissions were to l e estimated in detail only for the five t1 ten carcinogens fishynally ~elected Thmiddotrtfor(bull a 11w1st1n of tt II iim pol cint ial vus 1~1middotd 1atbulld llpon
knowledge of the s11bstances manufacshyture and USP for R The higlwst rating went to substanc middots which are v idely used especially in consumer products A slightly lower rating went to subshystances which an~ routinely emitted from industrial processes during proshyduction and use Some materials are employed in such a way that emissions are quite low even though tight emission control may not be required by law Materials in this cntegory were assigned a value of two for H3bull Substances which under federal or st ate regulations may not be discharged to the exterior envishyronment but which could be dischJrged by accident received the lowest rating
Each candidate was evaluated (n the basis of its 1ropen-ity to decompbullgtSe in ambient air Materials with half-lives greater than eight hours were considered moderately to highly stable and rated five for R4 bull Low to moderate stability was assigned to substances with halfshylives between zero and eight hours Compounds known to exist in air for only a few minutes would be rated zero
Ta hie II Dcfinitinns of the criteria used
R1 Prbullbullsent use in California
100 of max (109 lbyr) 5 10 1)f max (108 lbyr) 4
1 of max (107 lbvr) 3 01 of max (106 lbyr) 2
001 of max (105 lbyr) l lt001 of max (lt105 lbyr) O
R2 Growth in California use
+ 20 5 +Oolti to +20 4 Positive growth to 0 3 Stahle or unknown 2 DPcline 1 B1middotin~ phased out 0
R Emision potential
Widespread use in ltonsumer products 5 R1middotlatively p1or con rot over emissions 4 Rdatively good control over emissions 2 Tihtly c1int rolled 1
R4 8tabilit~- in ambient Air4 tvlnderale t1 high st 1bility (l 1 middot2 gt 8 hr) 5 Lw to modmiddotrate s1 bility (t1 l -0-8 hr) 3 Unstahle(t2~febull minutes) 0
R Dispcision Potential
Emitted lar~ely as apor or iine 5 particulnte
Emit led largely as rnarse particulate
Rlt Evidence f Carcinu~enici ty
Known or suspected human carcinogen 5 Known mammalian carcinogtmiddotn 4 ~usptctcd 111ammalan carcino~en or 3
known m mmalin muta~tmiddotn Ames ltmiddotst p sitive 2 Prpoundgtcursor 01 co-car inogen
a t11 is the half-liftmiddot
A-3
tion state or anion associations may change in the atmosphere metals do not degrade and were considered stable Asbestos is likewise itahlc Many of the clcc-ompo~ilion rtbullnctions oi orinnic molecules are mediated by light Such substances if released at night would have several hours to disperse in the surrounding area
A rapid way of assessing the relative potential of different substances to spread from a release point is to note their physical state under normal amshybient conditions Accordingly we scored materials emitted as vapors or fine particulates the highest for R5 and coarse particulates the lowest Intershymediate values are possible for varying amounts of fine and coarse particulate emissions from the same source or from different sources
There is as yet no widely agreed upon measure of the relative potenciemiddots of carcinogens although some ranking systems have been proposed14 Extrapshyolating data from in vitro techniques such as the Ames bacterial mutagenicity test and from laboratory animal studies to humans is problematic Therefore a less quantitative measure of the carcishynogenic potential of each candidate substance was used The candidates receiving the highest scores for Rs would be those for which there is strong evishydence of carcinogenesis in humans Exshyamples are asbestos which is implicated in mesothelioma vinyl chloride which has been identified as the agent of liver cancer in exposed workers and bisshy(chloroinethyl)ether shown by epideshymiological studies to cause lung cancer in resin workers The next highest rated substances are those for which human carcinogenicity is unknown but which have produced cancer in one or more mammalian species in laboratory tests Next are those which have not been shown to be carcinogens but which have proven to be mutagenic in test animals Substances for which the only knowlshyedge of carcinogenic potential is a posishytive Ames test (producing mutations in histidine-requiring strains of Salmoshynella) are rated 2 Finally substances which are implicated only as precursors or co-carcinogens would be rated lowest
Substances unequivocally associated with carcinogenesis were considered as carcinogens in this study Conditions of emission and exposure including the presence of co-carcinogens were facshytored into the evaluation of each candishydate where possible Carcinogenic subshystances derived from the metabolism of a precursor were considered as carcinoshygens However ubiquitous substances which have been hypothesized to be precursors (eg secondary amines nishytrous acid and nitric oxide which combine under certain circumstances to
1155
form N-nitrosoamines) were not conshysidered because of uncertainties in the importance of their link to the carcinoshygenic compound and the practical conshysiderations demanded by the scope of the study
It was beyond the scope of this study to jud~e the validity and interpietation of the experimental ad epidemiological evidence upon which the carcinogenicity of candidate substances has been esshytablished We accepted the conclusions about carcinogenicity drawn by the Inshyternational Agency for Research on Cancer or the National Cancer Institute and did not consider dosage or route of administratioll of tested substances However considerations of test validity did enter into the subjective evaluation by the panel of experts
Weight lr1 the sdditive ranking scheme each rating criterion is weighted according to its limportance relative to the other criteria Little precedent exists for assigning these weights In our judgment and as generally agreed by the panel of experts (see below) R 1 RJ and R6 are more important than the other criteria Evidence of carcinogeshynicity was considered to be the most important criterion of aB so W 15 was assigned a valtze of 3 W 1 aid W 3 were set at 2 and W W-i and W5 were set at 1 In order to discern the potential senshysitivity of the rankings to the weight assignments the candidates were also ranked using equaily weighted crishyteria
Ranking Candidates by Panel o~ Experts
Anine-member panel of experienced governmental industrial and academic scientists whose disciplines include~ -organic and physical chemistry indusshytrial hygiene toxicology epidemiology and iegulatory control of toxic sub~ stances was convened to provide addishytional data for our ranking algorithms
to discuss our candidate --ubstanc -sand rejections to suggest pos-ibie ne suushystances for consiceration and Lt rank the candidates indepen11mtly ( f our own ranking Tvo v-ee ilts he fore the meeting pa-d uember-s were jven one-to three-page dossiers on eac I canshydidaie substance
At th~ start of the meeting- before any group discussion the pand was asked to rate each candidate suhtance with a score from Oto 5 Next cmiddotach candidate was discussed at length e provided ~n ovrview and summarized critical issues identifi2d up to that Pbullgtnnt Through materials brought to thtmiddot meetin and their persoril experience panel memshyber Veimiddote able to provide much useful information on the candidates and adshyditional insight in to our ~a ting criteria
At the encl of the two-d1y sc~sion the pa-iel ugain -ated the cai 1didates
Flnal SelediDn
Tabe III show- the highest-scoring substance a~ det(rraine1 by the 1ddishytive and mlltiplictive approaches and by th~ panel in ~e additive approach 21 single r1nking as obuined by avershyaging the two ran kings resulting from using equal and uneguai weights The ran kings ltif most candidi tes were unafshyfected hut carbo11 tetralt hloride chloshyroform chromium and inorganic lead changed more tban thmiddotee positions_ Because uncertainties in the data base predude imputing signifiance to small differences in the Lnal ordering the lists in Table III are prtmiddotsented in alphabetishycal order However it is of interest to point out that benzene consistently ranked hihest
Becausimiddot some cindidates had equal rating scores we ltould nut choose exshyactly ten candidat ~s from the additive and multiplicative rankings Instead the
top nine and eleven were selected from the two exercises respectively We also considered the ten substances scored highest hy the panel at the end of the session The final consensus selection cornsistt-d of the 11 candidate substances appearing on at least two of the three lisL- For the substances included in the cun~ensus ranking a baseline emissions inventory is being conducted and a source testing program is being deshysigned
1ejected Substances
Some comments about certain subshySiF1ces not appearing on the final list are in order inasmuch as they include known carcinogens and compounds which hae received considerable atshyllttion in recent years as occupational c11c111ogens
rouisionally Rejected Substances Appended to the consensus ranking (T1ble III) were vinyl chloride gasoline and engine exhausts tobacco smoke and pesticides No further action by the CRB is recommended at this time for vinyl chloride because it is already subject tu the USEPA emissions stanshydard a CARB ambient air quality strndard and an OSHA standard
Gasoline and tobacco smoke were appcndlmiddotd to this list because each ocshycurs very widely and contains several of the candidate substances reviewed in this study some of which are in the final listing For example gasoline contains benzene ethylene dibromide ethylene dilmiddothloride and alkyl lead compounds the last three being in leaded grades only Both gasoline and diesel combusshytion products include PAHs Tobacco smoke contains among other neoplasshytigenic substances nitrosamines PAHs nickel arsenic cadmium and other heavv metals Manv individuals are inshyvolu~tarily and i~ many situations
Table HI Hi~hest ranked candidates from each rnnkin~ methodbull
Highly ranked but no inventory
l I ighesl consensus recommended Additive Multiplicativ(~ Pan I of eqwrts rnnkmg at this time
1156 Journal of the Air Pollution Control Association
A-4
lirtualy unavoidably exposed to t11-k1ltco smoke Because the sources of gasoline its combustion products and tobacco smoke emissions are well known no specific action was recomshymended for these materials during the tbullmissions innnlory and -ource testing design stages of the present study It was considered importint ho-vemiddoter to draw attrntion to the general publics exposhysure tot hrse substances PrsticidltS are listed for the same reason thou~h ad(bullmiddot tailed examination of pesticides w1s beyond the scope of this study Many pesticides are widely used and some of them are known to be carcinogenic
Other Rejected Substances Acryshylonitrile and vinylidene chloride were placed in a provisionally rejected group because of the panels suspicion that imports of these compounds to Cnlifornia from Japan may be appreshyciable yet are hard to substantiate Should such imports be verified in the future these two compounds would take on greater importance Formaldehyde was also provisionally rejected because the preponderance of evidence indicates that it is not carcinogenic and that bisshy(chloromethyl)ether is not formed from formaldehyde in appreciable quintities in industrial environments 15 The ocshycupationally controlled carcinogens ethyleneimine and beta-propiolactone were rejected in part because of their reactivity in air At the time of this study DBCP a pesticide was no longer heing middotproduced in California and was therefore rejected from further considshyeration
Occupational Regulations and Community Exposure
A question of interest to the CARB was whether the regulation of acknowlshyedged occupational carcinogens adshyverselv affects the ambient air outside the v~rkplace Our general findings can be illustrated by the example of asshybestos
Asbestos is a very widely used mateshyrial for which no ambient air standard exists Concentrations in the workplace are limited to an eight-hour timtmiddotshyweighted average concentration of two fiberscm=1 of air and a ceiling concenshytration of ten fihcrscm 1bull In meeting this standard exhausting air containin~ asshybestos to the ambient air is not reshystricted except by the USEPAs reshyquirement that there shall he no visible emissions containing asbestos piut ides from s11eh foci lit i1bulls (bullxd11din~ hrak(bull shops 1 Considning that undc-r certain conditions 101 asbestos fiherscm1 could ht Pmitted without being visihl( 11 this standard mamiddot allow considerable asshybestos emissicns It is unlikely however that emissions would actually approach these levels First since the OSHA standard cannot generally be achieved
November 1979 Volume 29 No 11
by ventilation n ine the generation or asbestos partid in the workplace must be greatly reltlu ed Second the air is usually filtered 11gt prevent recirculation of asbestos to 1 workplace Asbestos waste must bed 1sposed of in sealed imshypermeable bag~ or containers Thus under current onupational regulations the amhient air 1enerally appCars to be afford(bulld greatn protection than it would without s11ch regulations 1
Conclusion
The screenin and ranilting methodshyology presented in this paper proved to be a feasible approach to establishing a priority list of Jirborne carcinc)gens in California We ftel that it is an efficient means of focu~ing further efforts on emissi()ns inventories source testing and ambient measurement for it not only identifie all the carcinogens of potential concern but it also permits the state regulatory agency to direct its reshysearch resourCtmiddots toward those subshystances of partilmiddotular interest within its jurisdiction
References l National Elllission Standards for Hazshy
ardous Air P llutants A Compilation as of April 1 19 3 PEDCO Environmenshytal Inc for US Environmental Proshytection Ag1middotncy EPA3401-78008 (NTIS No I B 288 205) 19i6
S Gray Ne Jersey Department of Enshyvironmental l 1rotection Trenton NJ private com111unication 1979
1 H E Chrislllsen E J Fairchild and R J Lewis Sr middotmiddotSuspected Carcinogens A Suhfile of th1middot NIOSH Regii-try of Toxic Effects of Chemical Substances 2nd Ed National Institute for Occupational Safety and HPalth NIOSH i7-149
4 B 8 Fuller J Hushon 1 Kornreich R Ouellette L Thomas and P Valker Scoring of rganic air pollut1nts for US Environrnental Protection Agency Mitre Corpt ration T1middotchnical Report MTR-6248 I 9i6
S F R Lozan11 Toxicants Selected for Priority En- 1ronmental Assessment corresponde11ce with EPA Region VI Administrat r 19i7
1gt Survey of inyl Chloride Levels in the Vicinity of Keysor-Century Saugus California US Environmental Proshytection AJet middoty National Enforcement lnvestiiatim Center Denn-rand San Francisco E A-n02-77-017a 19i8
0 Bardin l middotstimon~middot in Hearings on Helationshi1 Bet wfen Cancer and the Environmei 1 US llltuse of Hepreshysentatives bull th Congress 2nd Session Committee (In Interstate and Foreicn Commerce ~crial No 94-141 1976 pp 7~i0
middot T H Ma11middoth Carcinogens in the workplace lwrE to start d1middotaning up Sc-imiddotrmmiddot l17 1-1110) lfi8 (197)
I CH ar1111 Classif1r11t1on of the Exshyposure Haza d Prtbullstgtnled Ii~middot Thirt~middot-Six Clwmirals 011 I llixllmmiddots lternorandum from Clllm1middot I Offi(middote of thtgt Assmmiddotiate lgt1rector for 1 middotarcinoltnesis Division of Cancer Cau~ l and Prevention lo SushyJwrmiddoti-or~middot ( 0 Auditor Gtmiddotneral Acshycounting Oft lmiddote19ii1
I0 Chemicals ith Sufficient Evidence of ( arcinogenit ty in Experimental Ani-
A-5
ma1s Jnlernauona1 Agency 1or 1c-curcn on Cancer IAHC Working Group Reshyport Lyon France 1978
11 Threshold Limit Values for Chemical Substances and Physical Agents in the Workroom Environment with Intended Changes for 1977 American Conference of Governmental Industrial Hygienists 1977
12 M H Rogozen D 1 Hausknecht and R A Zisk ind 1976 A Methodology for Ranking Trace Elements in Fossil Fuels According to Their Potential Health Impact by Science Applications Inc for Electric Power Research Institute 1976
1t L Margler R Ziskind M Rogozen and M Axelrod An Inventory of Carcinoshygenic Substances Released into the Amshybient Air of California Vol I Final Reshyport Screening and Identification of Carcinogens of Greatest Concern by Science Applications Inc for California Air Resources Board 1979
14 T H Maugh Estimating potency of carcinogens is an inexact science Science 202 (4363) 38(1978)
15 C C Yao and G C Miller Research Study on ais(Chloromethyl)Ether Forshymation and Detection in Selected Work Environments by the Bendix Corposhyration for National Institute for Occushypational Safety and Health Cincinnati Ohio Contract No 210-75-0056 1979
16 Occupational Cancer Control Unit Calshyifornia Department of Industrial Relashytions Los Angeles California private communications with staff members 19i9
This paper was developed while Dr Margler was a Staff SriPntist in the Energy-Environment Systems Division of Science Applications Inc (SAIi 1801 Avenue of the Stars Suite 1205 Los Angeles CA 90067 and Mr Reynolds was a Project Manager with the California Air Reshysources Board Sacramento CA 95812 Dr ltar~ler is currently emshyployed as DirEgtrtnr of Bnvironmental Scitbulllltcs for WinzlN nnd Kelly Conshysultini EnintNi t1 Third Stretbullt EurekH CA 9gtgt01 and Mr Heynolds is Air Pollution Control Diretbulltor for the Lake County Air Pollution Conshytrol District 55 N Forhes Street Lakeport CA 95-13 Dr Rogozpn is a Staff Scientist and Dr Ziskind is Deputy Manager of the Energy-Enshyvironment Systems Division of SAi
1157
Appendix f~
_l)ispobullition of Miscellaneous Sites
Gould Inc Ver_n)_~ St~condary Lead S~_n_elter
The emissions of arsenic from the four large secondary lead smelt~rs
in Cali forni) J1~1 ~ bull~tiilnted in the program This estimate was based upon a
uniform fugitive emission f1cJr nt WPll supported by measurement inforriation
Therefore source monitoring was r2cq11)nded 1nd Gould Inc Vernon being
~he largest was singled out It was however proposed to monitor both Gouli
and RSR Corp (Quemetco) in the City of Industry since analysis of the plants
revealed significant differences in plant equipment and engineering The
latter being more typical of 1 nodern facility
As a r2sult Jf pr-~-middott -iiscussions and plant inspections we became
aware that Gould was actively in the process 1)f constructing a new facility
which would completely replace the existing one We have monitored progress
on the new site and concluded it would he inappropriate to utilize program
resources to conduct field tests at Gould ~rr1issi 1rns from the new Gould
faci 1ity wi 11 be bas~rl 11p11) test results from RSR
PG ampE - Pittsburg PG ampE - Salinas a~d So Cal Edison - Long Beach Oil Fuel Power Plants
It was app ro p ri ate tJ cons i dP r the (~mission of arsenic from power
plants during the initial study stage inc~ imiddotrace quantities in the fuel oil
are known to be ernitted Because of the population distribution in tile
vicinity of three plants and some 11realistically conservative estimat2s of
erlissiJn c-1ctors tht~ facilities appearc~d among the top seventeen stationary
sources of potential c1Jn-r-n HovJ~ver as a result of a reexamination of
emission data it was concluded that an error ~dj ~een made in the material
balance of arsenic - resulting in tn emission factor equivalent of a 300
release ~io lit~rature was found to justify 1reater than a 30 transfer
function Thus we estimated at the outside th~ emission factor should be
reduced from 013 lb per 1000 lb to approximately 0013 lb and resulting
ars9nic 2missions for the entire state w~r~ cJ11servatively estimated to be
1760 lbs (from 17600) divided amongst 311 the states power plants Clearly
then the four secondary lPad smelters estimat~ of S9410 lbs of arsenic(
B-1
emissions per year makes consideration of poer pl int ernissio1s if ffsenic a
very low priority
In a literdture review by SAI - Methodokgy __f_or Ranking Trace
Elements in Fossil Fuels Accordi~9__t_~_their Potential Health Iripact (Roqozen
1976) - emissions estimates f0r 15 trac~ substances were researched for c11l
and fuel oil pomiddot1er plant c 1J1v---r-sion Source content combustion pr-)~~
transfer functions and contrcl methodology were co1~idered to develop e~ission
factors The output to input ~atio computed and utilized n -~i-i~ study for
arsenic was 002 to 03 (ie tra1sfer function betw~en 2 11d JO~) reflecti~q
the wide variety of fuels processses and controls nationwiJe The upper end
reflecti1g both high arsenic coals and poor e11iss10n contro 1 devices Clearly
neither of those conditions accurtely apply to th2 thr~~ si~~~ r1nrl thus
substantiate the rlecision not to perform emissions neasur-i-~n~s
Calaveras Asbestos Company - Copperopo l middotis Asbestos Mini n(i and Mi 11 i nq
In the past (late sixties and early seventies) ~h2 ~3laveras
Asbestos Company came under heavy criti-is1 1fter imiddot1sp~ction rieasurements
revealed serious problems However significant reduction of e1nissi0ns have
occured prompted by NESHAP regulation and occupcttional standards SAI
inspected the site in f)~c~1b-2r 1979 under an EPA contract An missions
inventory was published under that work (Ztskind 1980) and the bottom line
conclusion is that currently no significant ernissioi1 are gteing released as a
r~sult of hl-isting which would reach the publ le i 1le 0pen pit is some 900
feet deep with blasting at the bottom 0ver 80 percent of emissions in the
CARB Emission Inventory System were attributej to pit blasting In fact at
its current depth blasted 1aterial does not reach the mine surface with tile
explosive implacement used Additionally the site is more remote than might
be realized The ~ituation is vastly different today than in the past and
currently attention should be paid to the issues of occupational exposure and
breakdown of ventilation syste111 controls in t1e inilling operationsmiddot He
recorrrnended that no furtw~ corsideration be Jiven to this site for the
purposes of tl1is w)JJil ie identification )f silt~middot1ificant releases vhich
might he responsible for causi1J middot1 fmiddot middotgtbullgt middot 1= 1(11 1bull11 tinn Pxposure
8-2
Vctri0J5 Refineries
It was est=tblished early in the anrilysis prmiddotJr111 V1at gtenzene
emissions from refinery operations coulrl in th~ 1guregate constitute a
significrnt source Since there ilre a large rrnrnber 0f refineri~s in
California (46 in Los Angeles County itself -it t(~ i 1~ middot)f r-1ltariination) with a
wide variety of types sizes ages etc their eva l uat 11) 1 i =r~sent -3~ i)7
monumental task It was noted immediately t 13t ~~it hi n the total scope of the
study the design and conduct )f c r~f i r1~ y testing program which would develop
a complete benzene emissions in1~ntory for 01e site was impractical let alone
to d1aracteri ze emissions fro111 three ie those listed among the 17 rnost
s i g n if i ca n t potent i a l st a t i on a r y sou r c e em it 1-_ l rs bull
The three refineries were singled out for special attention in the
previol1s phase ~ecause they uriiquely had components which process materials
containing 10 or more per~nt hy middot1-bulli1 11t )-~112~ne (46 FR 1165 Jan 5 1981 pg
1491 ) Est i mates by E P A ( F e d bull r l Rr-~ g i s t ~ r 1q81 ) i n d i cate that 90 p e r cent o r
more of the total benzene fu 1Jitive emissions arise from such components
Therefore attention is appropriately focused on the three henzene production
consumption refineries Chevron (Richmond) Arco (Crson) and Chevron (El
Segundo) SAI staff considered the possibliJ )f -1 testing program at one of
these refineries and cone l url 1middotd it would not 1)e cost-effect i v2 for o number of
reasons
California is a minor producer and consumer of benzene with
approximately 15 of the 114 hill ion pounds produced and
consumed nation1lly in 1977 This can 1)e riY1pared with the fact
that California has approximately 10 of th~ ~ations population
Benzene exposur~ to the generil nJmiddot1lation has been partitioned
among the various sources Aproximately 90 of exposure is
estimated (SRI 1978) to be caused hy gasoline distribution
activities and 1ehiclEmiddot emissions in 1 ir~an areas with nearly all
of the balance )Y benene hrndl ing operations (refineries and
chemical plants) In the case middotlf California this percent1g~ Jill
be even JT1ore di middotpruJor-tionate b~cause of its greater s1are of the
population and ~esser share of the henzene ha~dling Furthermore
since the three refin 1bullry cites ire heavily urbanized no new rural
population cegmmiddotnt is beinq exposed
B-3
Californias most heavily urbanized Air Quality Management
Di st r i cts have a l ready adopted t en zen e f ugit i v e em i s s i on
staridards comparable and vith S( 1ine features VJr stringent than
the proposed national (bullmission lttandarrl (4f-i FR 1165 Jan 5
1981) Thus the concmiddotlusion rleri1ed above (Le refinery emissions
are secondary cause of exposure to the urb~n microopu~ation) is
further reinforced since H is irojected that re1eases from
components in benzene service (gt10 benzene) will be reduced by
73 by the proposed F~deral Standards T~~ ~istrict rules (eg
South Coast Air Quality Management District 456 and 4fi6l) are
not restricted to berz~ne per senor to only components in
benzene service The rjLs 1lso i-lclude flanges in addition to
the components ca i led uut in the propose nat i 01a l standard
The EPA estimated that if the proposed emission standard were adopted the
1~aximum annual benzene concertr0Vioi1 -~01middot 1 plant would be 36 ppb at a
distance of 0 1 kilomeL~~ avay Cor1paring this ground level concentration
with the general urban background i1 California of 19 ppb shows the latt2r t1
dominate In recognition of the secondary im0ortance of these thr~~ i~~s i~
was decided to utilize program resources to d~velop field data at other sites
1111here little emissions informati1)i1 1as cll-1aila1)le
8-4
APPENlJIX C
US tn v 1 ronrnen ta I Protection Agency
middotJc k
METHOD 108 - UETERMINATION OF PA~TICULATE ANlJ GASEOUS ARSENIC EMISSION
FRUM NONFE~RUUS SMELTERS
1 Applicaoil ity and Principle
11 Apmicroli1ability This rnetnod amicropliltS to tt1e determination of
i nor~an i c d rsen i c ( s) e1n is sions trom non f errou Sm(~ I ters dnd other sources as
smicroecified in the reuulations
12 Principle Particulate and gaseous As emissions are withdrawn
isokinetically from the source and collected on a glass mat filter and in
water The collected As is ther1 analyzed by means of atomic aosorption
spectrophotometry
2 Apparatus
21 Sa111pliny Train A schematic of tt1e sampling train is shown in
fiyure 421 it is similar to tt1e Method 5 train of 40 CRF 60 Appendix A
Tne sampliny trdin consists of tt1e followiny components
Note H1i s and a11 subse~uent nferences to ottler methods refer to the Methods in 40 CFflt 60 Appendix I
3 Annual Book of AST11 Stdndarctbull Part 31 Water Atmospheric
Analysis American Society for T~sting and 1-1aterials Philadelphia PA
1974 micromicro 40-4~
C-15
APPENDIX D
~f-~__d_y_ e s f o r Pr o c e s s i ng a n d An a l y s i s of PA H i n
Co~e Oven Emissions from Kaiser Steel
Global Geochemistry Corporation
10 Samples expected
l 1 Connective tubes between section welded to oven port cover and remainder of sampling system - These are to be stainless steel tubing disconnected and capped off for each sample They will be about 10-15 long
1 2 Impingers in series first two water-filled third dry - The center tubes will be removed and wrapped in clean foil and the impingers stoppered as is for shipping One set for each sample Protect from light during shipping
20 Extraction -procedure
Sample fractions are to be covered with foil or in opaque containers with inert gas flush during storage in freezer Solvents used in extraction will be BampJ glass distilled pesticide grade low residue) Working bench area will be screened with amber plastic and light bulbs
2 l Connective tubes will be rinsed with dichloromethane by partial filling capping tilting several times and draining into a beaker This will be repeated until the washings are colorless but at least three times
22 Impinger contents will be transferred to a separatory funnel The impingers and center tubes will be rinsed inlo the funnel using a measured amount of dichloroshymethane in portions The funnel is shaken well and the extract drawn The rinsing and shaking is repeated with two more portions of solvent The extracts and washshyings are combined over Na 2 S0 4 bull
A separate portion of the water used for impingers is extracted in the same way
For each liter of water 100 ml dichloromethane will be used per extraction ie 300 ml total
Internal standard is added to the combined exshytracts at this p0int
D-1
_rmiddot
30
40
Concentration of extracts
Bulk solvent is removed on a rotary evaporator in a water bath at not over 40deg stopp~ng short of dryness The res~dual volume should be just less than the f i n a l v o l ti 11 e o wl1 i ch the s o l uti on i s to be d i l u t e d (accurately) The object is a final solution with roughly five mgml solids concentration in dichloromethane After the vol w1 e ~ s made up to the mar k i n a vol umet r i c flask an aliquot is removed and evaporated on a tared disk in a nitrogen stream and weighed to determine yields The remaining solution is stored in a ial or ampoule sealed with a Teflon-lined septum A split for SAI QA may be made at this point (see Section 43 also)
Gravity column c1eanup chromatogram
4 1 Into a 10 min 1 d column with co2rse sinter and Teflon stopcoc~ 6 g 120 mesh silica gel (activated at 120deg) 1s sluiry packed in pentane cooling the column by evaporation from a tissue wrapped around it and wet with acetone) This minimizes vapor gaps in the column A cap of 3g sodium sulshyfate is added to the top The column must not dry out before the chromatogram is complete
42 A sample of the extract concentrate is taken This is to be based on experience of expected PAH conshytents but initially would be 5-10 mg This is exchanged with pentane by adding 10 ml portions of pentane and evaporating over about 100 mg silica to small volume Three to five portions of pentane are used The final residue is rinsed onto the column with pentane
4 3 The chromatogram is developed with four solvent steps
l 25 ml pentane 2 l 0 ml 2 0 ~~ dichloromethane in pentane
II II ii II3 l 0 ml 50 u II II4 l 0 ml 5 0 ~~ methanol
The column volume is approximately 8 cc This is allowed for while collecting fractions During the first chromatogram the graduated cylinder collector is observed for visible solvent changes and a more precise estimate of the column volume
T h e f o u r f r a c t i o n ~ a r e e v a p o r a t e d i n a n 1 t r o g e n stream at not oveimiddot 40 to l 0 ml then stored under nitrogen in the freezer The principal constituents
0-2
of the fractions will be
l Aliphatics 2 Lower molecular weight PAH
11 113 Higher 11
4 Polar materials
Fractions 2 and 3 are to be analyzed by HPLC the others retained for future interest or in case of an incomplete or defective separation Splits for SAI QA may also be made from these fractions
50 HPLC Analysis
5 l The column is a reverse phase partition column C18 bonded to microparticle silica 4 x 250 mm The solvent system is a gradient from 6040 acetonitrile water to 100 acetonitrile initial slope setting l 5min The flow rate is one mlmin the injecshytion sample volume one to ten microl depending on conshycentration Detectors are UV absorbance and fluorescence in series The absorbance is set routinely at 296 nm but may be varied in certain instances to aid in identification of peaks The fluorescence filtars are narrow pass excitation at 360 nm and cut-off emission above ~10 nm Both d e t e c to rs a re rec o rd e d s i 1i ul ta n e o u s l y
52 Reference calibration is based initially on indishyvidual chromatograms of single compounds and roushytinely interspersed chromatograms of multicomponent reference mixtures The proposed calibration stanshydards are the following at minimum
Phenathrene Pyrene Chrysene Perylene Benzo(a)pyrene Benzo (ghi)pcrylene I n d e n o ( l 2 ~ c d ) p y r e n e Coronene
The internal standards (added in Section 22) are 910-Dimethylanthracene
9-Phenylanthracene 9 10-Diphenylanthracene
60 Peak Identification
Significant peaks not recognizable as present in the reference standard will be investigated by alternate methods for identification
D-3
6 bull I Stopped-flow examination of UV absorbance for maxima
by manual wavelength varmiddotiation allows checking against known reference spectra (published or measured) This is usually sufficient confirm a component already suspected to be present
62 Collection of the fraction contairino the unknown peak allows later determination o~ t~e full absorbance andor the fluorescence spectrum for comparison with reference compounds
63 If not already run on GCMS the sample can be sent to SAI for that purpose possibly allowing the comshyponent to be identified Since the HPLC elution order is not the same as the GC retention sequence it may be necessary in affibiguous cases to collect the HPLC fraction and run it separately by GCMS
The~e may be numerous minor constituents not identified The decision on how many of these to track down and identify will be made in consultation with CARB adv~sors a~d with SAI since at some point the returns ute not worth further expense
_64 IdeAtified peaks are quantified by comparing their areas to the areas of the known concentrations of reference compounds A recovery factor based on the added internal standard nearest in elution order to the sample component is then applied the amount of internal standard originally added to the sample is divided by the amount found in the HPLC analysis This ratio is the multiplier used to correct for losses in concentration and chromatography steps
70 Quality Assurance
71 Transmittal of sample ~plits to SAI provides for external QA This should be done with 15 (or one in seven) of all samples
72 Blank water extracts are carried through the same analytical procedure (10 of total)
73 The cleanup and HPLC analysis are duplicated on 15 (or one in seven) of all extracts
74 The multicomponent reference is run first every day then elution volumes and peak area for each comshyponent are plotted on a control chdrt to pinpoint developing problems At least evetmiddoty tenth chromato-gram will be this standard It wi 11 also be rerun after any service on the instrument~ such as a column change
D-4
APPENDIX E
Gas chromatograph 1mass spcmiddotctromlt~try protocols used by SAI Trace
Organic Chemistry Laborator1
Aliquots of sampl=s received from Global Geochemistry Corporation
were analyzed by combined c11s chromatography mass spectrometry (GCMS)
Liquid extracts were spiked with an internal standard compound (deuteriumshy
labeled phenanthrene) priot to analysis by gas chromatography mass spectroshy
metrymiddot A Finnigan model 4( 21 Quadrupole GCMS system including an Incas data
system was used for all analysis Liquid samples (1 microliter) were injected
into a 30-miter x 025-mm ID SE-54 fused silica open tubular 9as chromatoshy
graphy column (JampW Scientific) A programmed GC oven rate of a 4 minute hold
at 30degc followed by a 40degCmin temerature increase to 160degc and a s0 cmin
temperature increase to 270degc was used for sample analysis The detector
system was a quadrupole mass spectrometer operated in the electron ionization
mode at 70eV The quadrupole mass filter was scanned from 35 to 475 atomic
mass units in 095 seconds A hold time of 005 seconds was used to stabilize
the electronics prior tote next scan Data were continuously acquired using
a Finnigan Incas Data system which writes time intensity mass spectral data
to a magnetic disc The mass spectrometer was operated under computer control
during acquisition Figures 33-7 through 33-9 show the total ionization
chromatograms for each of the three samples that were analyzed
Following analyses the data files were searched for priority
pollutant polynuclear aromatic hydrocarbons (PNAs) In addition to the
quantitative data reduction used for priority pollutant PNAs a survey analysis
was performed on all significant GC peaks (ie signal to noise of greater
than 101) The survey analysis was based upon a computerized library search of
individual background subtracted peak maxima scans for qualifying GC peaks
The library used for qualitative identification AJas a combined NIHEPA
National Bureau of Standards library 1hich contains approximately 26000
entries
Tables 33-11 through 3-13 present the results of this survey analysis As expected a series of al~yl Slbstituted PNAs were observed in addition to heterocyc l i c compounds con ta i ni n~ nitrogen iind sulfur moieties Further
investigation of the mass spectre of the nitrogen substituted identifications
confirmed the presence of benzonitrile isoquinoline andor quinoline acridine
carbazole and alkyl substibted ~yridines Ion specific searches for N-nitroso
substituted amines were neg1tive
E-1
APPENDIX F
Detailed Fiber and Mass Concentration Data - Johns Manville
=-- -i l L - [- - =-- -= f- - =-- -l J middot - -=-
i-- L =-- - ~ ~ L ~ ~ =--shy r---- --
- 7 l -=----7- -- - - - -- =-- - - ~ =-- - 1- -~- l
g~~~~ r_-middot~ ~-i [-- L ~ L Ldeg
middot--~ l - --------=-- r- L r- r-
~~~ =-- -- r- = - -r---
a- - - - - _ - -I~=-- L -- l 7 -~--
gtZlt _
- _(_ -lt lt =- = ZJZ~
-
__ _
~ -- __ middot- -~ ~~~ = L ~f~ Z ~ C
middot -
__ ~ lt lt =- _~~ijE
gt = z lt gt
rbullshy__
F-2
SI I HMllLfc J H 1 I OBr I lll-H ANI) NS~ coHcrrrllliTJ ON CUfllllTI VImiddot SIJMMIIY
TIITIL JnI~ SfiNtlEO TOTtl Ill- or FI ITETI TOTI Illlt ~111-11 11 ltllOT FIIACT I Ori 1111 OF ~middotWI MUllllNJbull lOTI 01IIMImiddot 01- JII TOTI FI lilmiddott~ UgtIJH nD
= t nwo 00 so MI CllONH = t I 1JOO SO (M = II 900 SO CM t bull (0000 l1T~
I I JO SO CM 70000 CUil JC Ml-TEHS
= ~lt O V l lWfl~
Manville D-2 Baghouse
723 347 -633
CIIHYSOTI LE JfllII I nou J11BICUOUS NO PTTFllN NON-SBI-STOS tLL FIIIEIIS
FI 111-11 COllflT ltFIIWll-
ITllClmiddotrir r lllrf
2gt()
) ( I f S bull1 bullbull
I 0
l ~-Irbullmiddot
00
fl JI)(~
00
0 OUU~~
00
0 OflO
~60
1110 000~~
FI IWII ( ()[IClmiddotNTHT I ON ltI- I mn ll11 ~41 GI OF FI LTEH) (FIBlmiddotn~ ll-11 CIJB f-11bullTl-H OF IH)
I (2fi71bull+0G (aG(l-+05
( gt027 F+04 OfmiddotHH
OOOOOE+OO OOOOOEtOO
OOOOOF+00 o uuoul~middotH)O
0 O(OOE+OO o 0000I+oo
J bull f 1)07 E+O( fi r71HE+OG
I
w
TcfLI ~~lllllmiddotCE 1lllmiddot lt~O Ctl lFII ~~O UI OF FIiTEi) t q UI 11-11 Cll II fllmiddotTlbullll OF J I H)
I 590)1+07 6 t 90llbull-HH
( HMOE+O 2 40411bull-1 o
0 OIHHlF+OO o 00001+oo
0 OUO(I 1 00 OOOOOJi+00
() 10110~ Ji)
OOOOOE+OO 1 lHHllJ i 00
0 00001middotgt1 00
M ~~~ middotor-ic Irrrn TI or
( PGRAMS 1bull111 ~c UI OF F J LTEH) (PGRAMS 1111 UJI illI Lil ()j Ill)
IUitTII Ml-Ml ltMI tJIOfl~~) ~Tlgt lgtEV
Ml- H Lot c1middoton nr COlmiddotF I II
I) I MWTEll Ml-N ltMI c11or1~~) STI) tHV
MEtl IOC cImiddotor1 r1N CCWF V11
AIt lbullI rt 1bulli I~ (Sil MIC) STIgt 1gt1-V
Ml- N 101 c1-ori r1u COlbulltmiddot V1ll
I 0 I ~if1+07 bull tJf I (11bull-HH
r 1gta 1 1) bull 2 (10()
0 lt700 l 1Hi4~ 2G44
() ~( 1 027B
- l bull 6 I 1bull1middot 0 J l)J 09J~fgt
1a 1071 1~ 09rn
0 2(jlt)fi l ~HVil 4040fi
~ HltF+O~ U (H~61bull+o I
o rooo 0()()00
-0GJOB 0 (000 0 (000
OOfiOO 00000
- )()57
OOGOO 00000
0 0 1)02 00000
- I J 0 0 0 1W 00000
00000 0 ()000 00000 00000 00000
00000 00000 00000 0 (WHO 00000
0()000 0()()()()
0 ()()()) 00000 00000
00000 00000 0000() 00000 00000
00000 00000 00000 00000 00000
00000 00000 00000 00000 00000
00000 00000 00000 00000 00000
00000 00000 00000 00000 00000
00000 00000 0 (WOO 00000 ()0000
( ~1701 l B bullJl~H 0 lt~( I BIJI ~ ~i Ill
o~-~ 0 ~j
- I ult-fc 0 I Hbull)I o bull)abullgt~
12 (1 41 17bullgt7
0 1 l I H) 4 ltmiddotI 1
VOUJME (ClJB MIC)
MFiN ~Tl) 111-V ru-1N 10C Ct-OM MN COE- Vtll
2lHH7 gt HUB
-2 7m 0 060 1)
( IJl)()C)
0 001 00000
-ltj 74a 1gt 0001~ 00000
00000 0()000 00000 00000 00000
00000 00000 00000 00000 00000
ooono 0 1)000 00000 00000 0 ltWOO
2 7 0 1J lJ I J J
-2J+H () ()~(
B 77gtB
lt ~
lt
C
J c
~ lf
_
ii
(]j (fJ
J I 0
a1 c r--- l)-- Clltr (Y) r- ro C)(V) gt deg CN C1 (0 I ~ C M
N --- r--
w J
s rJ gt-
c
~
LJ
t-
~ - -lt-[-bull
0- ~ l L L l~ -18~~
L a- - ~ I-
-I - ---
-
-- -- -- - - ~l
-1_ - - --
--~--- [-middot-~--- t-- -- [- -_ 3J ~I r-
z g ---
l
t--[-
l l c +-shyc
~ Ci - - I
-1--r-middot~- ~
L l- L ~
L
II V
C
c
t--
-= -- - - ---- - _
middot= - middot-= l J l~ - ~ ~ L - ~ L ~ [--
II II II II II II II
_
---
--
z
- I
==== lt _ --middot --~ 7 _
- ~~- _
z ~ ~ =lt ~~~ti3
f
gt =2 lt_
2 z _ lt=lt i~iS
-- z ~ lt
~~~t3
~lt~ =--~~~sect~ _ -- -middotImiddot_
L -_ -
F-4
(lJ II) I
~ Or--M
ltlJCsr-M ~ O) bullbullbullbull ~ lCMIO bull- ci gt (= N lC I M ~ 0 N
_
J J
---r--co - - ~l r-j = ~i
~~ 2-
88 rJ--_-ltI -
lt ~lt
z
Cl J ~ _ e-l ~~
- ~ middot~ e-- l ~ l~ [- - ~ -~=----~
_ L
L
--o---
-o-~shyc L~ ~ - 0 - Ll-C d o~-
11middot II 11 11 II II II
-_ 0- _
--
-- lt== ~~ ltZ z-== -~~=E~8C~l~-=rI - lt ~ fJ
ltZ lt---=lt===~~~- _ ltltlt=gt ~~=~~~~~ -
~ ~ lt -=z~ z _ -shy middot--z =-
-middotmiddotmiddotmiddot middot-middot -----
r bull bull _
- _ - ~-middot __ _
~ _
middot- - --- middotmiddot
_
- l
~=~r -~= ==
lt =~~~ti
F-5
FIHJ-ll Nil MAS~ CONCFNTllJTlOi SJ I SMllLE I l) Al OJ CUMJIATI JIbull t-llHMHY
TOTL iHf- SCNNFD lOTiL Illmiddot OF FILTFTI TOJI Ill- JSlllmiddot~ll I I OIIOT FlltCT [ ON 1111middot OImiddot ~1-M iIHlllllJNE TO 1 1 I ()lIJMImiddot ()I t I H TOTI V 1111-11~ COUNTED
= 1080000 SQ MI~RlozNS = 11 900 SQ CM = 1 l bull )00 SO CM = 1 00000 liHlS
11 IH) so CM 2G70000 CUBIC METfiHS
1 ) fi FI JllIIS
Man vi l1 e D-2 Baghouse
723 0 347 -633
Cl Ill Y-OJ l LE MIPII I BOLF Jf1BI(UOUS NO PATTEnN NON-ARHFSTOS ALL FI BFllf~
F 11l-ll f()llrlr ( I 1IFH-- l 17 G o 00 00 00 19fi
Tl fL P1 Sl~JNNFD Tl lI ltl- OF 1bullrn~n H 11I HF ~IJlbullI) lll 1 ll(IT 1-ILCTON IL CW ~U-1 HErlllllNE
THl1 (11llrJI~ OF 1IH Tt1T-d 11111middotH COIJrfflI)
= 697~oo so MICRONS l l 1)00 so CM t I JOO t-o CM
= 1 00(00 lHT~ = I l 1)0 ~o GI
ant()Oo cun1r NETEns = 7 0 I l lWllS
Manville (upwind)
CIIIIYsOllY AMPII I nou JMfl[CUOUS NO PA1vrrHN NON-A~~HFSTOS ALL FIBFHS
v1mI1 Cttllffl l 1-- I II I fl~ 1 () () oo l 0 00 40 70
1middot1111unmiddot ToT1L f J BFllS o ooo~ ooow- t OTT~ ( 01rn 57 I 1~m too ooo~
1middot 1111--11 1or111rnnT I ON I Imiddot I 111 II~~ 1111 ~o CM OF f I LTFTl) 1 11111 1middot111 CUil middot11bullnn Illmiddot J l ll)
O OOOlJEmiddotH)O UUUUOL+OO
0 00001-1-00 0 00011 lbull-Imiddot00
G 1 I nn+OG 1 HJ llmiddot~+OG
00001+oo 0001101-1-00
r n~+H+uG l 7llHmiddotlE+Ofl
l bull l 1)1-n+O(i I J ~ 1)(11-middotl()i
T11T I ~Ull FiCImiddot 1 ll f ( ~ l lrl ll11 tn Ii OF F 1ITFll)
f ill I I I i 11 I U l ( i I Ill J
0 OOl)Ol ll() () U()0 I Imiddot()()
0 UOOOlmiddotgt1-00 O OOlgtUlbullgt100
l 0 1) ) I 1 lJ J ~ tl( 1H-HH
0 UUUOlmiddotH)I) o 00001-+oo
~ lt1Hlt1(E+OG 7(H(Ml-middotHM
0 OOOOJbullH)O 0 0000lbullgtHlO
77 I
N ~
rl gt- 10NCJfffll1T I ON
r orR AMS ITll ~q lfI (II- FIT n~ll) - LIi LIIll 111middotTU OI 1lll)(PGRAMS II i1 Ill MlmiddotN
bull I t bull I 1 f I ~ middotr1 Ill lmiddotl 1- [~ IOC ( Hl1I f r~ Ull-F ll
0 0000 Imiddot bull-110 0 0000 Ibull HO
0 ltHgtOO 000110 t) (HJ()()
0()()()()
0()000
llOOOOF+Oo 0 00001bull-100
0 ()000 OOtlOO 00000 0 (WOO 00000
0 (((7
0 ~0-i -0 middotlbullribull)bulll-
0 ltl l lt 0 oHi
00000 00000 00000 0(0HO 0000()
077[0 0 -1117
-0 ~bull)I)()
0 7middot1middot1 (i () L~bull)7
0 7~B( 0 I I
-o alt17n 0 (1)~1 0 17lt1 I
11 I 11ITlt I j i 1l 1 111bull I
f 11middotgt N gt I ii ii I V 111 ii IOC tHlfl flf1 Clll-tmiddotmiddot VAIi
00000 0()000 0000() 00000 O 0000
041000 ()0000 () 0000 o 0(1()0
0000()
0 llfl(i7 OltHJ
-~ lt-11 I 0 07 I 1 () I)lt)
0 (H)ll()
00000 00000 0 ()()()() 0 0(100
0 0 Omiddotlmiddot l
-c07 1)
0 I 1 I( 0 1101
0 l l j--
OOGli -~bullI) ( (J
0 1lt11 I 077G7
I 1 (71 rJI C)
tll N -r11 111-V rmiddotllfl trn CI llf I 111bull1 COIT rll
() ())(l(J
00000 0 ()(1()()
0000() () (1()(10
00000 00000 00000 0 (I()(()
00000
0 ~ I Ifdeggt 0 I Imiddot~ I
- I H 1)7
O I GOO I I i ) l)
0 0001) 00000 00000 00000 00000
0 1 1)17 o~ViH
-1 0 1)gtG 01lbullH 0 7 I till
0 l l lt G 0 ~r1
- J bull J1 1) I 0 17 I bull at-lt
lllllrmiddot lt Clll ii IC)
~I Imiddotmiddot N ~middot1 D 1)1bullI f-1 Imiddotgt f IOC ClltHI rm uw-- i ll
O(lOOO 00000 00000 () ()(J()(J
(l(t(()()
0()()00 00000 00000 (1(1(1(()
00000
000fi7 OOOGO
-G Jll7l 0011G 1 H20
00000 00000 00000 000()() 00000
0 0 I bulllCi 001w
-4 iHd 00101 I ~700
OOIOH 0 0 I Ilt
-(1 j l)~G
000(i 1tHH
--
--
L c~
U c t c
(
J
J -
C
C
c g 0 0
+ + e--c~ =~ - r~ C- l~
--
+ +---
- -C 0 o- -c--1
I
lt t w
lt
C
ta J ~ ltCi
lt ~
a---~ -- c gt C 0 rtl 5 -
VJ C f-rJJ c
[J_
ltI ~ C z
z _ ~
~ lte 0 ~
C
tJ
e
C
~ c C
L N
~~ C 0
C
l~ 1 0
t CI ~ ----~ -
i-
cc cc + + t=~= g~ --
-~ C + + t~ ---[w (~ lC
-
ccc ~t c5 ~ ee 00000
co--oc I
- - - - -_____
------ _ - _ _--__ ---__ ---- _ - _
ooocc
rr ~ c-~ E lt
C
-
~ C g L-
gt o + + ~~ ( L C- lN o -l
t l - -I
~ - L~ [ti e- [
l _ 1
- - ~_ C
gg -- -- ~ -0000-tr--- ____
----C __
- __ - _ --ggggg
- -
C
+ + -- --- -~_shy-----_ _ -----___ ooo~
-----__ ___ _ - - - _ - _ _- - - - ___ 0~000
--__ ---_ __ -- --c5 ooco
C cc
1 II II 11 11 II II
z
-
r ~ c
z
~ t lt i C z __ -- -~tt ~ middot~ ~
-middot middotmiddotmiddot _ - -----=___
_ =~ - _ -=shy ~
-- -~== 2~= ~ () V)
_middot ~ c( c(
0 er ~- ~ ~ lt 0 0___
-
- amp lt~ ~~~t8
F-21
) ) ) )
TOT1L AllE SCiNrmn TIIT1 L ll lbull1 tW FI 1lFll TIIL1 11bulli 1~lllbullll lllIIOT lIIJClION -11- OV TUI Mlbullrllll1Nft TOT1 I VOIIHIIbull 01- I Ill TOTmiddot I Imiddot I I 11-11~~ COi l NTlbulln
_LI__l_ ___l 4 bull P 7 bull P A R l n A 4 R 1 n 4 4 ___ 7 A ___ _RA____ 1 A 1A 8 1 SL__ 1 R 5 1 ~ 3 _1] ________ 111~____JJ__) 1 ~---~-~ 4 9 1 ----1___11
A0 1A4 JR 17 lAR A 41 4C 174 A ll~ JnR ll7 9 5 R12
----- --------- - ---- ---- -- middot-middot --- ---- -- ----middot-- ----middot ----~ 7 c ~ bull n 1 1 2 l 9 -s 1 -i s s ~ c c__ bull A____02 1 1 2 1 r n 1 c 0 r-- l ~ 1 o 1 0 1 t 1 o c- ----~--- -Pl Ol OP lA2 11A 61 c 0 ~ cq03 l 5 4 l 4 4 1 i r 7 -i r) c q 1 deg 1n
qA RO 107 ~_12_7 SA c 7 A04 _l _ s_ _________l 7 l RA____ 7 ______l _o 4 __ l nn_________ 0 _7________ R_l __ o ~ _____ J(__7 11__r ___ _t-__ o ____ ln ___ h_ ______ An____ c-_n __05 ]lA 140 170 7cA __2n1 qc- 7q 1n1
c 2c 7 lR 1 O 406 11 c A 0 4 n 3 4 7 l 4 R ~ 7 7 5
___ ___~ c Z bull A ~R_ ---2_h 2 bull bull 4 e R bull C07 l_ H l_ ri ------- R1 --- c ----------- AQ _middotbull 4_p____ r 7 ___1[_2_ _r--_______ _ A____________ )_3-____ _____ __t 9 ______ ~_____ _3 __ --~___()_______ __I__08 C7 lR9 9A 1n~ ll7 S AS 84 ~Q 33 11 11 1n R 3A 3309 7RJ 1R lA 14-1 17fI 77 4R 7(_
---~- L~ 1 7 -~ 1 7 4 1 l 1 9 S 3 9 C) 4 9 A ~ 9 q10 __ 7 c _ ___ _ 2 RA l A ____J__6_R 1 C 4 7 P C 1 rq I
-----~~a-R___-=4bull2_1 e __ middot-middot middot bulls __ - __ IL 9 1 ~_1_ ___3 7 40__19 1 ~~- gt q6_ (- l Q 144 43 _37 _ A f ---middot ~ bull r bull q ~ 9 IL ~ 1 _0 -~ bull 1 A 3420 l r -- I- I q4 l qo l c c 0
sectJ~ BLVDII EJ SEi~D~ 1 nu tmiddot-middotmiddot-middot J _
tl~ r-middotn ----i J middot--1_ ___ ~1-1
A f gtiff_ ffmiddotc YYt)II ~ r bull Ja------ ijlttQ~QW~
___ ~Q -~imiddotbull1 ~H fl(--middot
-lt) bullbull
1 middot
middot--middot middot-- _
middotmiddot_ u - -_-~~middot_1r
~-- ~ _
~ -middotmiddot bullY
j ~ ~ L ~ ll c_l
~ pf-
111-r ~ (i middot) ~
bI middot_FT )-I Imiddot ~middotmiddot () --
I filfltf) J ~3 ilj-t ~ C _G~ llli_A T1 AN B~ACI
I~
603 middot Alondro Pork
l U (
Goll Coorll
~ htyen 1~J-(_
d n t U Wik c rr 1 i t_y of Indus try
(
H-7
Plant Site - Dow ch em i cal US A Pittsburg
I bullbull middot1 Sfl Wi - bull AUIIUlllJLII+
BELMOr r l -- -- 7iSHOfgt ITHACA lN I
N[VAOA lN TGERS llltI
Q LN
GfORC~
72copy~~~~-0AYTON l
I I+
__ middot
1- - -~ middot
I I
- ----------
Plrint lite - ilu rnnt Antioch
) j
(
H-9
Plant Site - Johns rlanville ctockton
a a
LEGEND
CENSUS T1ACT1
51 )I 51 07
1980 CENSUS TRACTS SAN JOAQUIN COUNTY
37
air smicroace Saturation concentratiun at 20degc is amicroproxiriately 12 Measured
values were bY and 74 These values r~sulted from gas chromatographic
dn a I y s e s ot t rd n s t er re c1 sa111 p I e s fr() m t n e 1U U I i t e r Ted l a r b a y t i 11 e i nt e9r a t ed
sample ror tt1e tank car displdcement of 2008 ft 3 the total CT emitted based
on tile ctverctye is LUU8 x 007L = 144 ft 3bull
Tnen for a density of p =PMRT
P = (1)(1gt4) 64 gL = 04lbft 3
(082)(293)
Tl1erefore tne displacement weiyl1t of CT microer ()ff-load is equal to 144 ftJ x U4
l b ft J = ~ 7 bull 6 1 b bull F o r 2 5 U off - l o ad i n gs an n u c1 1I y we have 14 400 l b s bull Th i s
compares with UuPont I s rnectsure1nents of 9 3 vapor content arid 18800 lbyr
emissions Saturation vapor pressure concentration emission would yield
2400Ulo
l)reathing loss from the tank can be calculated from AP-42 as
Lt3(1bday) = 619 X 10-5 M( p ) 068 D 173 H 051 T 05F CKcmicro 147-P
5 0 68 173 = (619 X 10- )(154( 173 ) bull (155)
147-i73
(lj) 051 (26) 05 (1)(1)(1)
= S 63 lbday
3where a working range of zuu x 10 o 400 X 103 lbs were used to determine an
average l i quid nei grit Jn nu a 11 y e111 i s s ions would be 2057 lb Therefore
working losses dorrii nate tt1e totil elissions from tne microlant
Summary
The upper and lovJer bounos of caroori tetracriloride emission ~vere
determined as Ll467 lbyear (upper bound on fugitives DuPont measurement of
storage tank work i ny erni ss ions nur 1a I tanK oreattli ng) and lbgt 15 loyear
(lower bound on fuyitives s1 111eas--1re111ent of storage tank working emissions
no ri n a I tan k breatt1 i ny ) bull
13]
100
SOUfCE TESTS - STAUFFER CHEMICAL COMPANY
101 SITE OVtRVItw
Summary
Stauffer is located in the Carson area of Los Angeles County It is
a manufacturer of polyvinyl chloride from tl1e monomer (VCM) ll11hich is produced
on sit2 Stauffer is the only California producer of ethylene dichloride
(EUC) which is converted to VCM The VCM is regulated by several standards
including Cal OSHA EPA (emission standard) and CARB (ambient)
During Phase I the plant was inspected and possible emission
sources were identified Th~se cans~sted of EDC storage ta11ks fugitive
emissions from valves flanges and pumps process vat2r content gas
incinerator effluents and th~ loading of outbound tankers
Since the completion of Phase I several events have transpired which
a ffectebullj the test program to deterni ne foe i 1i ty emissions These wer2
~ the completion of a vent ~ldS incineration system tied to all significant EDC storage tdnk breathers
$ the completion of a comprehensive study by SAi staff to develop a nationwide material balance for EOC
bull inactivity in EOC importation for the plant and elimination of EDC exportation from the plant
Based 11pon this input and the plant inspection of February 3 1981
it was expected that a plant emission factor for EDC can be determined with
relatively little uncertainty It is further expected that si 1Jnificant
atmospheric release of EOC du~ to plant operation may not occur at the plant
itself but rather offsite These vmuld drise fro11 two sources - the process
water discharge from the plant and severcil off site EDC storaye tanks
Facilities Description
Ethylene dichloride is being produced at Stauffer by two processes
direct chlorination and oxychlorindtion In the former EDC is produced by
d i rec t ch I o r i nd t i on uf et 11y l t~ n e i n the pr 1~senc e of cHl Fe C 1 J crl t d lys t I n th e
latter process EUC is microroduced by the oxychl)rindtion of ethylene with
hydroyen chloride and oxyuen in tne prescrlCe of a cdtdlyst typically Cuc1 bull2
132
lhe microldrrt fluriti1~s tuC after tne microrimar y reactions by separation and digt ti l I dt i ur1 fgtr-ocJuLL 1 s ctured ctncl umiddot 2d ctS foed for VCM production All
sourclS agre~ tl1dt c1ission of EDC is (f concern in its productionstorage
process stages ctnJ O little importance in VCM production steps (J~B 1980)
At the timt of plant inspection some storage of EOC existed at three
ledsed storage tanks located at tne Port of Los lngeles 1-iaterial has not
been 1-Jithdravm from these tanks during the 1dst year and there had been
discussion to consolidate materictl into a single tank
All microrocess components handling EDC storage are now tied to a closed
ventilation incineration syste11 It is expected that virtually all
chlorinated hydrocarbons includiny EDC will be etfectively destroyed since the
system must demonstrdte VCM concentrations are reduced below 1 ppm Firebox
tempera tu re is aoodegF and residence ti me greater than 1 5 seconds under
heaviest flow conditions Note that by way of comparison combustion perforshy
mance datct for PC8s are 99995 destruction at 1832degF and 1 second residence
time and 9Y9Y9YY4 at 2 seconds dnd 2372degF (N Flynn SAi Personal
Communication)
Fugitive emissions from valves and flanges are monitored by the
plant according to the requirements of SCAQMD Rule 4661 Pumps and
compressor follow ~ule 466 Emission and control requirements for vinyl
chloride are specified by ~ules 1005 and 10051 EOC concentrations in
wastewater are monitored several times daily in tne primary EDC steam stripper
stream and once daily in the composite plant outflow stream Daily water
disct1arge limits are 2~ ppm wit11 typical montt1ly averages being in the 8 ppm
vicinity The discharge limit is ~mbodied in the discharge permit (number
5061) vJith tt1e Los Angeles County ~anitation District
Tnere are a number of points at which tt1e rnc can be released to the
dtmospnere and they are similar for both dir~ct chlorination and oxychlorinashy
tion These include the following along with their emissions factors for direct chlorination
A chlorinator vent 2xliJ- 5 IOdSS per unit mass EDC produced
I) I ight end column vent -52x lt)
C dist i I I at i un column vent lXlU-5
u storage tank breatning 7xlu- 5
133
E storaye tankworkiny loss 27x 0-u 5F fugitiv~ emission bxll -
-JLi ildSteJdter l gtX] U
Em~ssion factors are si~~lJ fer oxychlorination processes
Enlission factor vau-5 (~1rbulle Tdilt2~ frcn tne literature (JR13 198U) and
assume 9b iciency for incineration (A-E) It is also assumed above that
EIJC emissions to water are Lf~ uf ~rihsion to citir ard that in wastewater
treatment 100 of EOC discharged tc1 vater is rel easl~d to a i 1middot ~ These emission
factors 1bull1e1middote used to prioritize r-2ieaies iJut were cldrly uude approximations
to tre planL l)ecause of thlt ncine-xt~on system it was anticipated that
fdctors A-E would be ri2dtKcdft 1-middotactor - 1nii~t1t be reduced since a monitoring
program had been in force al20st o~e y2ar Factor G and the off site storage
tanks~ vni~cn are not tied into lt1n ircinerator system rniyht -iominate emissions
Emissmiddotion limits and concomitant regulations embodied in Rule l00S of
the SC-l)MU have necessitctted tlw incinera-cion system Ninety gas chronatograph
probes are located throughout t11e plant including the stack of the primary
incinerator Con cent rations of vc1v1 are reported es sent id I I y be I ow tt1e r~yu la -
tory limit of 10 ppm and in fact below the limit of detection somewhat less
tl1an 01 micropra
lUl MEASUK~MENT APPROACH
Tne objective of tt1e proyram is to determine a pl ant emissions for
EOC It is not accemicrotabl~ to dtmiddotvelop sucll informatmiddot1on basect upon published
indlllstry Jide esti11dtes of plant control efficiencivs Fortunately it was
possible to desiyn a monitoring and calculational progra1n to determine EUC
emissions for tne site and not ctbsorb a disproportionate share of program
resources
Hie re are tnre~ modes ot re I ease from tt12 micro Iant dnd d fourt11
offsite
bull Post-Process 1nci11er-dtion
Processes A-E of ecc ion lll 1 dre a 11 1ented into the pl ant
incinerdtio11 syste11 rt1e conceritrcttion of VCfmiddotl contir1uously iectsured in till=
stctck cts SJdS output remicroresents d rectsondble UjJper li11it to diJply tor EDC
con cent rat i ons s i nce i ts ef f i c i ency of i nc i nera t i on 1s d t 1 east c l1 u a I to tr1 at
134
ot VCibulll Tlerefore knowledge of the system dirflow rnd VCM concentrcttion are
tt1e necessary and sufficient conditions for determining the bounds of EDC
reledse fhe pldnt expn~sse I wi 11 inyne~)s to provide tnese data in order to
support the calculations
bull Fugitive Emisions - Valves Flanges Seals and Other Sources
Fugitive emission from the valves pumps and flanges can be
determined more precisely tnan was anticipated since Stauffer has completed a
comprehensive leak inventory of all such components in compliance with Rule
466 466 1 and 1005 The Inventory delineates all leaks uncovered by their
three man crew throughout tne year and indicates the screening level in ppmv
and component identity In consultation with Stauffer we will be able to
identify the total number of components associated with EDC handling systems
( 700) the distribution ot substance composition streams the distribution
and incidence of leaks by hardware component type and leak rate We will
utilize this information to provide historical data to compliment our Foxboro
OVA sampling at the site dased upon this monitoring we predict an EDC mass
emission rate for the fugitive releases The plant has agreed to provide the
necessary information We propose to meet witt plant personnel prior to the
start of monitoring and finlize the sdmpliny strategy to accomplish 100
coverage of the lines of hiuhest EDC composition The sampling approach and
calculation of mass emission are described in Section 72
bull ~~as tewater
From examination of the emission factors derived from published
literature (see Section 10~) it is clear that EDC release from wastewater to
air is potentially several 11rders of magnitude higher than any other plant
source l1ased upon microl ant ml~a sured concentrcJ ti ons of 8 ppm EDC in water a more
realistic emission factor wgtuld b 24xlo-4 1-iassunit mass EOC produced
Clearly this could still be the dominant source Furthermore the release
point would be expected to middotgte locctted between the plant and the sanitary
district treatment site whi1 his~ kmfrom the plant at 24501 S Figueroa We
believe it is necessary to ndemicroendently confirm the average 24 hour EDC
concentration in the dischar-qe valer by obt1ininy the refrigerated composite
sample We t1ave identified ttie I 1nes of int~rest and received agrt~ement to
sample and analyze for EDC in the stream ~e propose to draw a duplicate
sample from the compositor nd ani1lyze for its rnc content This will be
135
compared with Stauffers para11eI tnalysis It is not reliable to obtain
direct readings of EDC by survey instrument in the air above the water flow since concentrations at crny single point will be i 1 the near dmbient range
Emissions will be cc1lcJLtecl 1_sing the plants vulumetric daily flm
and ass~m~ng that complete degass40~ of the effluent will eventually occur
This assumption is based umicroon the relative volatility uf EDC and tne distances
involved However it should be noted that no direct experimental or
monitoring data is available with which to confirn this Th2 composition of
the discharge stream is unknomiddotwn since it merges into a larqe multisource flow
Conversation with the LA County Sanitation l)istrict (J f1ilne) reveals the
stream to be both exposed and covered Vents exist where me measurements
could De made to assess gross leaka~e at key points
We will obtain samp1es of several plant process discharge streams in
40 ml bottles witn no head space Transit time fr0m sample collection to
analysis will be minimized and wi l I not exceed 24 nours This time frame is
conservat i v e al though ED C i s a v o l ct ti l e mater i al ar1 ct vJi 11 undergo
concentration degradation and outgdssing lnalysi) will b(~ performed in tt1e
SAI Trace Environmental Chemistry Laborcttory utilizing pur 1Je and trap analysis
FID gas chromatography A trial a11alysmiddotis was conducted and the EOC
characteristic peak was distinctive down to the ppb level Therefore the
analysis should easily confirm concentrations in t1e 8 ppm range
Offsite Storage ranks
Three EOC leased storage tanks are located offsite at the Port
of Los Angeles and are owned by annther firm Annual emissions from the tanks
were very roughly estimated based un AP-42 JRB 1980) as 44 kkgyear
based on preliminary estimates of stored quantites
Considering tne magnitude of this source these storage tanks must
be investigated We will gather information about their configuration and
control i n order to ca I cu l ate their e111 i s s i on s bull
136
lU3 UEfERM I NAmiddot 10~~ OF MISS IONS
1031 Incinerator
Vinyl chl 11ride co11centration is monitored dt approximately 90
locations throughoul the pl trlt (Lanyner 19Bl) including the incinerator
output Conc~ntrations are reported by Stauffer dS less than 0l ppm at a
flow rate between 10000 ant 12000 SCFM Although no direct monitoring of
EDC is conducted it is posible to conservatively bound the concentration of
EDC at U 1 pp111 Tl1at conce 1ltrat ion is a suitable ci10i ce s i nee it represents a
conservative bound on tile VCM levels measured near the stack output
Furthermore the molar volume will be taken as 224 L rather than the higher
value it would nave because of the sl i~thtly elevated temperature at the
detector location
Using 12000 SCFM one has thE annual emission of EDC as 12 x 103
SCFM x 283 LSCF x 526x105 minyr x lmole x 97gmole x 10-7vv 224L
Therefore incinerator emissions of EOG are thought to be bounded by 773 kg or
170 1byear
10J2 Fugitive Emissions
Seven hundred (7UlJ) sources were surveyed comprising nearly 100 of
EDC service All accessibl~ plant areas with streams containing greater than
1 EUC were screened except for a small number of devices located in areas
11lere active maintenance was being condJCted It is believed unnecessary to
p2rf-gtrm any emission factor adjustment since it is estimated that gredter than
95 of the requisite components were screened
Three leaks above an arbitrary OVA reading threshold of 20 ppm were
parameters and mass emission numbers fhe upper bound leak rate was
determined to be close to twice the nominal leak rate which accounted for the
response fltlctor of amicroproximamiddotely 05 by Radirn for TLV detection of EDC
( 13 rown 1980) bull
Stauffer found and reported approximately 10 leaks in the EDC
service during 9 months previous to the pLrnt testing Assuming no111inal leak
137
) )
Table 103-1
STAUFFER CHEMICAL MASS EMISSIONS PREDICTION FROM OVA SCREENING VALUES
OVA SAI Actual Upper Device Response Response Cone Source Leak Rate Bound Location (ppm) Rae tor (ppm) a b Se I Re Function lbyr lbyr
Gate Valve Unit P1559A
Valve Heavy Ends Column Unit 14439
Compressor Seal EDC Storage Unit Tl062
I-- w co
300 11 273 -035 096 0 17 153 D 6
11 I500 114 439 II 241 C 1
r III II II o-is lUI bull2000 vl 1 1818 A 9~
All emissions 100 EDC (12-Dichloroethane)
11
values in the nei~hborhood of 2000 lbs~ total could be emitted annually
Therefore it appears ttiat major reductions in the mass emissions were achieved
by the company run inspection prugram and the fugitive emission source has now
become of seconqary importance as a fraction of total plant emissions
lUJJ Wastewater Discharge
Wastewater samples were collected in duplicate from four sites
within the plant for determination of ethylene dichloride (EDC) concentration
The samples were collected in EPP standard 40 ml VOA vials on July 1 1981
Analyses were performed using standard purge and trap techniques coupled with flame ionization detection gas chromatography The results are given in the
table below
Sample description EDC conc~ntration range (microJml)
EDC concentration average ( 1-19ml)
Approximate fl ow ( gpm)
PVC Interceptor Box 82 - 108 95 200
EDC Stripper number Chlorination area C1404
2
011 - 014 012 25
Final collection site for pH adjustmentP663 349 - 3B2 366 350
Final discharge site sanitary sewer 6 - 254 158 500
The water in the PVC lnterc1~ptor ~ox was warm and represents washings from the
PVC reactor which travels t 1J the Interceptor Box in a concrete drainage ditch
Water from the EDC ctripper was vPry t10t and because of its heat was difficult to collect The ~1eat ot trie wat11 r may in pdrt account for tne
relatively low concentratio11 of ElC here as the EDC would out-gas from the hot
~vater more readily The fir1ctl collection site tor pH adjustment is a large
concrete container middotvith mixers in it located just prior to the final discharge
site Water at the final dischdrg~ site is being constantly aerated due to
the speed that it flows thru~yn the concrete drainage trough This aeration
could account for tne relatively large range in the rnc conotent found here
The average EDC concentraticmiddot1 at tne final discharge site is middotbelow the daily
discharye requirement of 25 19ml EDC
139
Plant oersonnel indicdted monthly iverage readings are typically on
the order of 8 ppm but daily averages can re 1ch greater than 20 microgml In
addition LA County Sanitation Uistrict staff (J Milne 1981) indicated that
an on-site impoundment pond can become laden with EOC during certain abnormal
operating periods and discharge Vdriances arl requested by Stauffer This
mi 9t1t Jccur on the order of O1c0 aCii ecH c1 1d t1erefore is no-c expected to
signifcantly impact averag~ cnnJJI discharg ~ values lt s not expected at
this t1ime tEit evaporative eriss uns from this pond are significant except
infrequently during upset conditions and spi 11 control operations Program
staff ~~ 1er2 una111Jare of any periods ~h 1~n EDC cimtent in the ponds could be
appreciable and therefore no sampling was performed
Utilizing the average uf the two final discharge concentration
1readinJs (158 micrograrnsm~ 1 and 1G 1Ji gpm fl)W one has 34600 lbyear of EDC
released into the sanitary sEwer For the pJrposes of calculating population
exposures from plant releases it will be assJmed that the EDC is locally
emitted from the wastewater streams Tl1ere ire no monitoring data available
with which to develop a more accurate release profile
However emissions of me fron the plant wastewater discharge can be
calculated according to the method of Mcmiddotckay (197S) as modified by Dilling
(1977) It should be noted that this mLst be considered an estimate since
conditions of fl ow and the presence of other substances will influence
emission rates
Using Uilling (1977) for non~erated flow first recalculate
Henry 1 s law constant (dimensionless-my of chmiddot~mical per liter of air divided by
mg of chemical per liter of water) dS
1604 P M 1 Wl
TS l
)
wnere
H Henrys law constant dimensionlessl
Pi tile compounds pur1~ componen- vapor pressure in mm Hg at T
Mwi the 11olecular weiglt
T tile absol ~te temperature of he wastewater in K
Si tt1e compounds ~uluhility in myliter at T
Then for UJC
(l6u4)(7~)(9Y)H = = U0447 with 1
(2lJ4)(8690)
LIii bullul1Jl)i I 1 1y 1 middot UL i11 wt1Ler yiven -1L L0degC is 8u4JU UHI tile vapl)r pressure
1 4 psi ( 7 2mm) at 7 0 degF bull
Tl1e overall liquid mass-transfer coefficient Kil is given by Dilling (1975)
as
(2211)(06) in mhr
-f-0-4~ (M _) 12+ 100 Wl-H-
1
=U108111tH
Then from Mackay the percent desorption is given by
c 1 = exp (-Kil tL) where
c 0
c = the concentrd ti on at time t of EOC 1
co = tile initial concentration
L = the Ii quid depth (Ill)
t = tt1e retention time (in hr) of the liquid in the wastewater
system
Retention time is bc1s~d 11pon a flow velocity range of 3 ftsec as estimated by
the Los Angeles County Sdnitatii)middot1 Dibulltrict (1J Milne) over a middotdistance of
ctmicroprox i 1nate I y S km to the Joi n_t Wdter Po 11 ut ion Contra l Pl ant Sewer fl ow
depth throughout the entire route have been estimated by J Milne as typically
1 foot 030m) to the Davison Pump Plant and 3 feet (09m) to the Joint
Tredtment Plant distances assumed to bt~ 1 mile and 2 miles respectively
Trdnsit tiine l)eco111es between 05 tnd LU hours sequentially Then computing
the net emission reduction as the product of each leg
Therefore 26 of the EOC i erni t ted bdween the pl ant and the Joint Water
Pollution Control Plant Rtgtsfdence ti111e and conditions at the plant account
141
for additional releases and residudl content is forther emitted between the
plant cnd the ocean discharge poiiit ~-Herever the flow is r2tained aerated
or shallow emissions will be accentuated
Offsite Storage Tank Emissions
There are three storage tanks leased by Stauffer for EDC storage
1ocated at 22nd and Gaffey Sts ~ bull San Pedro These tan ks a ~e used for long
term storage rather than providing feed on a routine basis Therefore
breathing loss rather than v1orkir 1J lass is )f concern An tanks are white
two being 67 ft in diameter while t~2 third is 57 ft All are 40 ft 3 inches
high and have capacities of e~thi l0~10000 (2) or 840000 gallons The
tanks are cone roof type and are not tied into vapor recovery systems The
South Coast Air Quality r1dna~J~ment u~ str4 ct has based their estimate of
emissions on the specification of 12 foot vapor level Using the AP-42
formula for breathing loss with the vapor pressure of EDC at 20degc and an
average diurnal temperature variation of 26degF one hds for each of the two
larger tanks
l) 173 H bull51 T bull 5 LlI = (619 x 10-S) M( _P_ tH F CKcp 147-P
68 173 51 5 = (619 X 10-S) (9119~ 116 bull (67) ( I 5) (26) (1) (1)
14 7-1 16)
Thus for the two larger tanks LB= 472 lbday
For the third tank 0=57 and = 178 lbdayL8 The total emissions become 23724 lbyear ~ince he plant is in the process
of shutdown it is not clear what the liquid levels currently are Note that
if the material in the three tanks were combined into one totamiddot1 -~1nissions
would be reduced markedly Exposure to a population from this site was
determined separately since it is lbullgtcated over six miles fron tl1e plant
142
110
ASSESSMENT OF PUPULATlUN EXPUSUR~
111 OVEKVIEW
A major program goal wa~ to compare emissions from the various
sources and identify and rank any 11 tlot spots in California where the general
population was exposed to elevated concentrations of carcinogens A simple
Gaussian dispersion model was therefore used to obtain order-of-magnitude estimates of exposure of the genermiddotal population surrounding each source
Since this was essentially a screening study use of more sophisticated models
was not appropriate
It cannot be emphcsized too stronyly tnat most California residents
are exposed to emissions of hazardous substances from a variety of natural and
rnanmade sources Urban dwellers dre typically exposed to greater concentrashy
tions than rural residents however all are subjected to so called 11 background 11 levels from multiple sources In order to place stationary
source exposures in perspective the typical ambient levels of each substance
were identified from the literature and compdred with the concentrations due
to the emissions from each ~lant Exposures were thus expressed both as
absolute quantities and as increments above background
Comparison of plants presents a further difficulty in that various
substances are being consict~red No attempt was made to evaluate the relative
importance of exposure to two different substances such as chloroform versus carbon tetrachloride other than by ambient concentration
112 DATA SOURCES
1121 Meteorologicdl Llata
The dispersion model to be descritgted in Section 113 required input
of annual average wind speed and frequency of occurrence of wind from each
compass direction These data were obtained for most of the sites from the
South Codst Air wuality Management District (SCAQMD) In other cases (Kaiser
Jotms-Manvi l le l)ow and OuPlt)nt) only clai ly cJVerage wind speed and frequency
data were available In al I cases we used data from the meteorological
station nearest the modeled e11issrnn source Table 112-1 summarizes the
141
Table 112-1 METEOROLOGICAL DATA BASE
Observation Site Data Period of Pldnt (Distdnce Source Observation Remarks
from Plant)
A11 i ed Che11 i Ca 1 Lennox SCA~MD 1Y71-1974 Averaged hourly readings available ~ 1 Segundo (3 111iles)
S ff~r-- Uemical Long l3each SCA()MD 1962-1974 Averaged hourmiddoty leadir19s avai1able offsite Carson (3 miles) storage tanks appruxi1ntL~ly 5 miies from
Figure 114-3 Cumulative Population Exposure to Carcinogenic Polycyclic Aromatic Hydrocarbons From Coke Ovens at Kaiser Steel Corporation Steel Mill Fontana
CV)
0 -I X-V) V) a _J
s 0
C 0 micro co s micro C a u
l ~
14
1 --
J- t_bulli
_ C
uw m 0
1 -0 a bullbull micro
microdeg ()
0 micro tbullmiddot C QJ V)
0
X LLJ
C
middot4middot~middot C 0 micro rt1
-- - 0 ~ 0
0
n __ _______bull~ -______ --al_ - ~_
I bull
I I - bull
middot 1
-~1
l i l ~l
~ 0 I 1
Li bull I t_~I
i bull -1 _1-J
0fmiddotbullmiddot-middot+-middott-t-r+-t-+-+middotr+-t--+-1--t+ t f-t-J-t--r-+tmiddotmiddot~~bull+-t--1~~bull-+-1 Id middot t 4 E1 (
I ncrementa 1 Concentration (~gm3)
Figure 11 4-4 Cumulative Population Exposure to Asbestos From Johns-Manville Plant Stockton
I
)
--~ ~w t _ ---~- ----------~ --____~-=~ M
~
bull I bull I I I I ii i Q t I i E e- I Ii tl I I I 6 I a I I J I f t t bull I G fl I
a I e 1 1 a 1 1 1 1 iii bull bull 1 1 bull 1 1 bull 1 1 bull bull bull bull 1 1i bull bull 1 1 bull a bull 1 1 r ~ middot u ~ ~ I0
+H~ ~ ~~~~-~~~~-rmiddot-~~~~~~~~~-~--~~-~~-~t~~~~~-~J C
~1_ c bull 1 t~ bullbull bullabmiddot 11 l bullmiddot II bullbull C- I u~
0 rl
gtlt I I I- 20 () ()
CJ -1 ~_J
J Cbull s
I If I I I I0
C bull I Ii I I I I I I I I0 1E
bull-+l I t I I ro s Lower+l
I I I G I iii I bull
() 14 I I IC
u I0 I I II Iu
C
1-_~I fi
CJ ~ I-- (j) CJ I G I I p +l
ro +-l I I I1~3~V)
0 +) UDf)Pr u I I 11 I I
CJ 1iII
Vi I 0 I I It- I a X
0 ~
w I I I I bull
Lower C 0
middotr-
- 4+)
(1j
J Q_
0 bull1
ij a 3- ti I
D d I I I I
I I I bull
II B I t
i ] i amp I
1 ei ~ amp 3 ii
I I I I
e ft a a I bull G
I I I bull I
I I bull I I t If
UpPrf
I I
I I I-_L~ lI I I f
I I I I I
- I I
j I I II I -
lI ImiddotbullII
l I
l Li
Incremental Concentration gm3)
Figure 114-5 Cumulative Population Exposure to Perchloroethylene and Carbon Tetrachloride From Dow Pittsburg (Solid Lines) and to Carbon Tetrachloride From DuPont Antioch (Xs) Results For Upper Lower Bounds on Emission Estimates Are Shown
1i-1 - middotbullth~middot t =lttF--i-+-middotr+-r+middot~middoti-middot-~-4-4middoti t -~--is -~middot-rmiddotibull--+ u ~~ i--t4~ t-bull~~311 li-i1middoti- C - JJ _ CL
-- I 1 t middot1 c7 Ibull ) ~ I-middot bulli C bull 1-_ 11 11gt1l11 ~bull -a bull-bullbull t a- II bullbullbull bullI lt_I Iii _- I
Incremehtal Concentration (microgm 3)
Figure 114-7 Cumulative Population Exposure to Carbon Tetrachloride From Allied Chemical El Segundo (Curves Corresond to Upper and Lower Bounds on Emission Rate Estimate)
--- I
~__~-unt2wbull--o1WCUatJ41ia1
16 0amp (V)
0 ~
gtlt
14~11l 1l QJ _J
~
1 middot-0
c t1middotmiddotmiddot 0
micro ro
_ c micro ~ 1euro1pound1
0- QJ u-J c 0 u O =-middot ~-= QJ
c_1 micro ro +- V)
0 micro ~ t1 O QJ 1l 0 a 40gtlt w c 0
-
ro --
micro
20middot l ---a I ~ I a 0
0
pound1r-t k
Figure 114-8
8tWI-S~1111111N~-
a I __ I wi~ i awLi--~~ 2 I I
1--+ I ~--if bull =~~--bullJ-+--f-1-----middot+--+middot-middot--r9c- 1 c- J
bull _ bull -~
Incremental Concentration (microgm 3)
Cumulative Population Exposure to Carbon Tetrachloride and Chlorofonn From Allied Chemical El Segundo Assumingsix-months Operation With Each Feedstock (Curves Correspond to Upper and Lower Bounds on Emission Rate Estimate)
i u---J--bull-+ i i--+ --4--+-imf-bullbullQ~-+-+-+~bullbull-4 171 i R = Fi middot1 1bull1 r1__ bullltaa ~a
Incre0ental Concentration (microgm 3)
Figure 114-9 Cumualtive Population Exposure to Carbon Tetrachloride From Allied Chemical El Segundo During Offloading From Midnight to 6 am middot-middotmiddot------middot--middot
--bull ft ampLA 1111-~lliaIN_WPl1A 1JIU_Mt91 MJ1WWlaquoPI-W1111~ --------__----Et1
M gt~middotI
0 -I
-X
~ t~ bullM
V) middot-middot ~ V)
QJ _J
f 0
0 -
C -4 0middot~middotmicro re f micro C a u
10 Cdeg u 0 3 0middotmiddotmiddotbull~ -0 a
+-gt ro
+J V)
+-gt 0
2E1-0 QJ V)
0 a X
LLJ
C 0 1pound1--middot+-gt ro
-- Q_ 0 0
I I
~ -~middot bull I -bull-_ bull I
middot lbullgtbull___I I
frac14 llk I
I bull
-~
I
-1_ I
Ibull II__ JI
I~
(1 f---1-+ bull-r_--f bull middot~-bull +-- bull i -bull 1 -1-+--+bull-+-t-+bull--t1 ~ _ 4middot 6 E 1t1 1middot2
1 3 5 I middot~ _ 11
Incremental Concentration (ngm3)
Figure 11 4-10 Cumu1ative Population Exposure to Ethylene Dichloride From Stauffer Plant Carson
_~ ~ t M
C X E k1~
-
- c- 1-1 ~ C middot-middot -_J
I--- -J 4t1middotbullmiddotmiddota
D
-+-shy --middot liiit1J
r ~
ltL
t-1 _ V
0 -+-1
middotr -bull ~1
X
C
+- bull1 ~1 atr l -=
~~lll(k~IHmiddot~~-Wa~middot~nL1middot~~~ta KF u n P middot~dE-tveRS ~-tl~~~-u-~~~tmiddot_-~~~~liil-~-~~~ tc r
I bull
bull -r~l-yen--~1l~a l fl __ A4o bullbull bullbullbull ~ 111bull--middotbull
8 NJ industrial sites 28 samples in industrial areas with benzene consumption facilities Torrance CA Tuscaloosa AL National Forest 1000 sarip Ies-Torontt) Azusa CA El Monte CA Long Beach CA Upland CA Los Angeles CA Riverside CA
bull Assumes all year operation on either feedstock If plant operates 50 on each feed the population exposed to greater than 50 increment over background goes to 16025 - 19377 and 4044 - 16025 for 100
Ambient concentrations of benzene vdry over one order of magnitude in the literature and therefore make tnis calculation questionable For Kaiser therefore the carcinogenic PAH assessment was used to evaluate incremental population exposure above background
+ The partition of emissions from Oow are 78 carbon tetrachloride and 22 perchloroethylene by weight
173
1142 Comparison of Incremental and l3ackground Concentratiors
Those sites whicn elevate background concentrations greater than SO~~
City Kentucky 6600 The latter two are associated wi t11 EOC pl ants Data
taken in senFice s~ations ano lrctffic areas ir various cities range from -~
300-3640 ngm~ As previous~J mentioned the Stauffer plant is discontinuing
operations These two sources should be examined as part of any possible
start--up permitting activity
11422 Kaiser Steel Corporation
The Kaiser steel plant polycyclic aromatic hydrocarbon (PAH)
emissions cffmiddotise from the coke oven omicroerations Comfdrison of the cumulative
population exposure Figure 114-3 and Table 114-4 with the specified
background levels for urban areas illustrates the breadth of the exposure
distribution The five known PAH carcinogens that were isolated in the coke
oven emissions were quantified in tile ambient air of Los Angeles by Gordon
1Y76
Benzo[a~pyrene 046 ngm J
t3enzo~e]pyrene osio Benzaanthracene U lo
Chrysene U60
lndeno Cl23-cd~pyrene 134
Althouyn these concentrations ctr~ low co1npared with 1 number of other cities
cited in the literature and repres2nt a very I imited data base tt1e predicted
concentrations from the Kaiser plant ~enerally exceed these levels by d
174
significant margin ie greater t11an 30000 are predict2d to be exposed to
4redter than l()-lt t 1tbull 1111tlit~nt background of 3j nqm 3 It ic 1 iklly tlat
lHrllerte exmicroo~urt 11 Ilie drld dre also 11evatPd ovr r11nbient I1owever since
backyround concer1tr1tions of t)enzlmiddotne show large variation no population
calculation was specified
11 4 bull 2 bull 3 Dmv
Carl1on tetrdcnloride releases from Dow constitute approximately 78
of tl1e total CT plus perc emissions in Table 114-6 and were found to elevate
urban background concentrations greater than 50 in five census tracts Emissions are predominately from storage and check tank working and brething remiddot1 eases
1142~4 Allied Chemical
The Al lied plant was modeled several ways since the plant can
operrttt~ wiUl chlorotor111 dr cc1rbon tetrctlt hloridt fo(~d The c1bull-i pr~middot11ntlid in
ldlgtlt~o ll4- drld llil-B represent dflrlUil 01Hrdtion with eitt1er teed Partial
yedr operation with each feed can be scaled from the individual annual modes
and is presented for equal hltllf year operation in Table 114-9 As with the
Du Pont plant carbon tetrachloride emissions arise from feed tank working
loss fable 114-10 illustrates peak exposures predicted to arise during an
off loading cycle As expected concentrations in that six hour period far
~xceed annual average values Insufficient data were available to contrast
levels with backyround transient concentrations
175
l~O
AUHLAlHJ~ CONTROL n CHNil)UES
Alternative control approacnes for the most significant emission
sources dinmig ~he various pl1ants n ees fitie(J n the follo11 ric suJsections
EmpiiasVi nas Deen placed in dedlinq ~riU1 Uiose sources of 91reatest absolute
inaynHud2 (lbyr) and those const1 ut 0ing trie largest increm121middot1t middot] bulli~he
bdc kgrnund con cent ration of Hie em~ tted subs tcince i~o ctt tention Wi 11 be ~i ven
to the s~iondc1tY sources witt1middotn cK1 microant or to trlL case of J011ns-Manvi I le
since -UH imiddotajor source is less tran LOO 1byr and is not microrejicted to raise
bctckgro 1wd exposure levels to the ~~ntr21 population Furt12rmore a11
emission sources dt Kaiser Ste 1 1iC(1 v1 1~n directly dedlt -1ith in this
program r1re elltlteci ro t11e cc1lt1bull ovcmiddotr nperatmiddotions These faci 1ities are to be
closed du1rm and all primary reel 11i~ oH~rdtions discontinmiddot 2d Kaiser
forecasters have predicted further deterioration of the plant economics and
the phased closure has been accelerdted for primary steel making operations
Note that this closure is essentially irreversible cince differential
exmicroansion and contraction of the coke oven structur1 occurs in the cooliny
process and it would be improbable that ovens could be reheated without
extensive reuuildiny at major expense
12l STORAGE TANK EMISSIONS - fAUFFrn IHJW OUPUNf ANU ALLIED
At c1ll four sites emissions from storage tanks constitute either the
primary or near dominant (Stauffer) source of carcinogen release andor
genero1i iJOpulation exposure Curreritly the tanks of interest at each site are
permitted by the local Air Quality IJistricts hJwever they do not require
emission ontrol systems for vdrious reasons Tl1e estimated releases grounds
for exemption and other pertinent information are given in Table 121-1
In order to amicromicroreciate tth~ practicdl dlternatives for emission
controls the provisions of SCAWMU Rile 463 ar~ listed below which specify the
acceptable alternatives for tdnks r4uirin9 controls ie tanks 11aving
capacities greater than 39630 yal with suostances of true vapor pressure
176
Table 121-1
Site
Stduffer Uff-Site Tanks (J) (Sdn Pedro)
Allied (El Seyunrlo) - KdJ
1bulllctLer1 d I reld iturctjc
-J -J
Dow (Pittsbury)-Product Check Tanks (4)
Uow - microroduct sturag~
DuPont (Antioch) - Raw Materidl Feed Stordge
Approx Capacity~
6lUS x 1g (c) 84 X 10 (1)
lt3Y630
18000 each
70000L) 3UUOUO
30000 working 42636 liquid full
STORAGE TANK
Substance
Ethylene Uichloride
Carbon Tetrctci1 I Jr i J1
Perchloreshythylene and Carbon Tetrashychloride
CT perc
Carbon Tetracr1 l ori de
SUMMARIES
Emission Mode
Tank breattli ng
Tank v10rk i ng Tank breatttiny
PERC-~JOrk i ng PERC-b r2a t ~ i ng CT-working CT-breattling
Tank breathing Tank breatt1i ng
Tank workin9 Tank breathing
Vamicroor Pressure
014 at 70 F psi
1 63 psi at 68 F
16d psi at 68 F(CT) UJ9 psi at 68 F(perc)
168 i)Si dt 68 F
Grounds for Exemption
Exempt from control re4uirements to SCAQNJ Rule 463 since vapor microressur-2 dues nut iXC~ec
l~ io at sturdJ~ conditions
txe11pt TlJil fule -+b3 since tank capaL ~J ~ s u~der 39630 gal (lSO 11 )
Exempt from Regulation Rule v~~vu u~~~~~shytank capacity is under 39630 ga 1 andor substances are not classified as organic 1i4uicts
Exempt f ro1n Ru l ~ 85300 because subsLctnce is not classified as dn organic li4uid accordin to Regulation-Rule 8120
exceeding l~ microsi at storage conditions
~ floating roof tanks
~ fixed roof tanks with an internal floating type cover
bull a vapor recovery system with vapor collection and retllrn (or disposal) proce5sinJ lXctbullerlin~ 95 efficiency
ln a cllemicai plant a typical vapor recovery systein (KR Evans
SCAQMU) migm consist of a collection inanifold to d recovery system (suctl as a
vapor sphere) to a compression syst2m dnd subsequently to absorption and
recovery systems The absorpt~on st~n c~nsisting possibly of (rubber
stripper or activated carbon
In dealing with speciti( carcinogenic substances such as vinyl
chloride hiyhly efficient vapor control system perf0rmance has sometimes been
stipulated necessitating incineration systems
Fer th~ cases of corccrn realistic alternltltives are as follows
Stauffer Off-Site TanKs - Tnere are a number of options which can 1~arkedly 3
reduce ei1issicns from these tanks from th~ calculaL~d value of over 20 x 10
lbyr One alternative is to consolidat~ the material in tne three tanks
One of the large tdnks can contain the currently stured material and reduce 3emissions to apmicroroximately 13 0 x 10 lbyr Anotl1er alternative is to
transfer dll E0C to a single floating root tank Various types of such tanks
exist dlld are reviewed in the EPA report Lrganic Cht~mical Manufacturing Volume
J Storage Fugitive dnd Secondary Sources EPA-450J-80-025 eg internal and
external floating roofs and a variety of seal desiyn configurcttions Generalshy
ly such designs would be expected to reduce emissions to the order of
one-fourth to one-fifth the current level Cost of 1nstalliny ct contact
single seal internal floating roof wds estimated by BB Lumquist Pentrex
for EPA as $27770 in 1Y8U dollars for 7U ft diameter tdnk Cost to build an
external floating roof tank was estimated by G Stilt of Pittsburg Oes Moines
for EPA as ~14UUUO for D=67 H=4Uft Anottler com111on approach is to utilize
cdroon ctdsorption This works we11 witn nonpolar hydrocarbons dS VvC is
removed from the vapor phase A basic syte111 consists ot tvo carbon beds and
a regeneration system Regeneration is typically performed with steam or
vacuum Figure 121-1 illustrattS rhe systems (Basd2kis 1~110) Steam raises
tne VOC vctpor microressure The resulting sti~am-VOC mixture is condensed and
173
AC-TJA~D CA-e0-l
-----~---
LOW- ffiESiJ~C ~T~H----
PiltOCESS COOL-J~ At-JO oLOWER DRY t-J~ 2gtLOWE~
COgtJDENSER
J
COOLIN(a - --------
WATi=-A
RECOVeRE SOLVENT
WASTE WATER
Fi l 1 1 middot gure middot - Activated-c arbon Adsorption System
179
rout2d to c ~epordtor decanted and returned to storage In vacuum regenerashy
tion VOC vapor is desorbed by pu1 ling a vact1um on the carbon bed then condens-
ed a1d returned System efficiency ~ ~ est ii1a ted at 96 f(duc middot i m from fixed
roof levels (EPA 4503-80-025)
H~fri 9ctated vent concicr1sers dre une of the most cu1mon emission
reduction prJcesses -for conttol ~ ng hx2-d-- storage tai vUC Figure
12 1-- 2 i l 1ust rates a u n i t ( Er i scmiddot n 19HU ) eff i c i en cy of middot middot~ ~ c middotlt middotmiddot _y i s rate ct
bet1Jee~middoti 60-90 for the vapor pressure range of concern For such a large tank
the capital costs would be high ~igure 121-3 illustrat~s EPA estimates
(EPA l ~~SOb) for the condenser section Condenser sys tern a ea mu l d be in
excess of 1000 ftL
It should be noted t -at Jtc1 uf fer Uiemi cc1 l has arH1ounced the c 1 osu re
of its VCPVC plart PrevfoJsly tt12) had planned to purge tlle off-site tanks
of EUC Since any future po~sible micro1ant stdrt-umicro will necessitate a compreshy
hensive SCAQMD review this ltoument can assist in evaluating proposed control
measures
Allied and DuPont Feed Tanks - In bott1 of these cases the nore significant
quantity of emissions arise from tank workinltJ ie during the off-loading
activity rather than tank breathing Contnl ~easures taken for working
emissions are thus of primary concern Ther~fore no detailed discussion will
be provided on the alternatives for control gtf bredthiny emissions Additionshy
ally both tanks are in the range of 20U00 gallons which is a capacity where
floating roof tanks are almost nonexistent Out of 670 floating roof tanks
surveyd by EPA less then 15 were smaller than 30000 gallons in capacity
Therefore the utilization of any floating roof concept and seal combination
will not be considered
Commmerc i a11 y it is estimated by Ou Pont that carbon tetrachloride
feed costs approximately $017lb (E Taylor personal communication)
Ttierefore less than i2500 is lost to l)uPont annud l ly as a result of working
l eve l em i s s fo ns and 1es s fr om Al l i ed bull Thus i t is u n l i kel y that any
appreciable economic incentive exists to dev1~lop ci vapor recovery system
However a Cdndidate system could be pattern~d after the chloroform
feed-storage unit currently at Allied This is a dedicated vapor balance
system Chloroform is off-lndded from tank cars and stored in a closed
unvented tdnk As the 5tordye tank is filled the c1ir spdce d1spldced is fed
Figure 121-3 Installed Capital Cost vs Condenser Area for Various Materials of Cunstruction for a Complete Condenser Section
182
back to the tank car Also in this case the storage tank is not vented in its
normal breathing mode and is part of a closed feed system to the reactor
Vapor recovery system alternatives which are particularly effective
in loading and handling include r~frigeration adsorption andor absorption
Control efficienci~- 1re est1mateJ in the rlt1nge of 90-95 (EPA 1980b) and of
course depend on the speci tic su1)Stance and equipment used Carbon adsorpshy
tion systems were discussed above The smallest capacity carbon adsorptionmiddot
system priced by EPA (EPA 1980b) has 2 vertical beds of carbon (900 lbs - 4ft
diameter by 3 ft depth) witl1 an ilstalled cctpital cost of $135000 based on
December 1979 dollars
Dow Tanks - Dow has two pair of p~oduct check tanks which alternately are
filled dnd off-loaded Emhsions from th~sl tanks were calculated based upon
operating cycle (3 day fi 11) satubull-dtion vapor pressure and physicdl
chardcteristics Direct hec1d-spa1e testing was planned however it could not
be accomodated because of r 1stri c ed access due to unscheduled mai ntendnce on
the field test day
Dow has indicated that chey are studying the option of installing a
vapor controlrecovery systt~m in 1hese and their largerproduct storage tanks
The extent of their engi neri ng ind assessment work is unknown as are their
current plans It may be possibk to incorporate a vapor balance design into
the system whereby the disp1aced apor is trdnsferred to another process point
within tl1e system Alterna1ively a vapor r1middotcovery system such as carbon
adsmiddotorption is fedsible Hot-1ever since emissions for t11e check tanks are
primarily due to working loss tht~ use of a conservation vent or an adjustment
of its operational differential p~essure would be ineffective for tr1e
reduction of the bulk of such emissions Furthermore since check tank size
is re1at i ve 1y sma 11 no cons i derctt I on was given to conversion to a fl oat i ng
roof configuration for those tanks Conversion would be possible for
the large storage tanks
12~ WASTEWATE~ EMISSIONS - STAUFFER
Emissions of me throug11 ~-a~)te-1dter discharge are the largest single
source identified at the plctnt sites The discharge limit was set at 25 ppm
by the Los Angeles County Sc1nitation iistrict and was estdblished with the
occupational limit in mind of 50 ppm JVer an 8 hour work shift dt the District
18~
tredtrient center~ T11ere hctd been suine di~cuss1on between pdrti~s of possibly
raisiny tne discharge limit since it is felt by Stauffer that dilution of the
stream is sufficient to d 11 ow it Itmiddot shou Id he noted t10-1ever tna NIUSH has
recommended the permissible exposure limit be reduced to 5 ppm averaged over a
work period of 10 nours microer ddy 40 ours per Ieek witt1 a cei l iny level of 15
pmicroin dVeraged over a 15-mi nute microet~ v(J
It is not certain at 1~hrt rdte me 1rill be reledscd from the
wastewater stream At the microlant discharge points wastewater is both hot and
aerated thus favoring release~ No measurements have been ta~en downstream of
the plant after considerable dilut1on has occured The distance to the water
treatment pa11t is approximately 5 kin at 1i1ic~1 point anerobic diyestion is
conducted lt is presumed that a11 ELlC will be released before final ocean
di scnar~Ji
A possible emi ss i D~ cont ro process for nmiddotduct ion of the EDC in the
effluent is by ~recess adjust~Pnts or additio~s For example process
modifications to the EUC stripping stage coula dramatically reduce discharge
lev~ls A control alternative is the use of octivdted carnon or XA0-2 resin
to recover EOC in tne discharge stream Tests of Gulf South Researd1
Institute on XA0-2 and activated carbon (Coco 1980) show high recovery yields
for nonpolar organic carcinogens unaer a range of concentrations Viable
suggested alternatives oy Srnitn included regeneration of Ue trapping
materials dnu even consideration of burning tt1e concentrate carbon media (at
greater than 1000 pmicrom)
Control approaches to reduce emissions from so called secondary
sources in general and waste water emissions in particular fa 11 into four
categories - waste source control resour(e recovery alternative disposal and
add-on controls Alternative control pro(~esses which were considered but
appear to be inappropriate to this case i11clude chtmiddotmica means eg
neutralization precipitation coagulatior1 and chemical oxidation thermal
destruction of tne unconcentrated ~dste s1ream is ii1practical biological
treatment eg aerdtion and biomass-wasteater contcct ~ienerally relates to
the treatment of soluole degrddable organics in the concentration range
bet~~een U01 and l~~ terminal storagt eg lanctfilliny urface impoundment
and deep well injection dre either indppl 1cable or 1mmicroractical Therefore in
summary the poss i b I e contra l approac ties fur this ca e inc 1ude the improvement
r
184
of semicrodration efficiencies in steam stripmicroing tl1e internal recycle of ~aste
stredms ctnd the ddsorption b activdted carbon The design configuration
~fficiency and cost )bviously depend upon numerous microlarit specific factors and
their determination vJOuld rec-Jire detailed analyses
n1e ~~dsrewctter systbull-m of a model cl1emical production plant based
upon the average properties uf a composite of 30 chemicals was evaluated for
EPA by IT Enviroscience (EPA 1980b) Included prominantly among the 30 was
tUC with the hignest uncontrolled secondary emission wastewater release rate
ie ~ percent of tt1e production and 34 perc~nt of the emission Cost and
impact analyses were evaluated for alternative control systems to reduce
secondary voe emissions from wastewater Four systems were considered a
carbon adsorption system (eAS) for recovery of the voe from the wastewater a
cover to reduce secondary VOC emissions from the wastewater clarifier a cover
for the clarifier plus a carbon desorption system and a cover for the
clarifier plus ct eAS system usinu a fume incinerator The scale of the model
system was greatly in excess of the Stauffer plant thus further making
detailed comparison i mpract i cal bull However emission reduction factors ~-1ere 6diven as ~9 and cost effectiveness per 10 g reduction genera1middot1y ranged
between $450 to 1733 These factors would likely grossly underestimate the
system cost if scaled down to the range of the Stauffer plant ie the order
of 34600 lb annual discharge
The alternative approaches of improved steam stripping efficiency
and internal recycling of the stripper iischarge stream with a reduction in
makeup re4uirements could decreas~ net ~DC wastewater content release
123 CONTROLS FUR SECONDARY LEAD S~1tLTER STACK EMISSIONS
Given our findiny of low (16 kgyr) ~missions of arsenic from the
reverberatory furndce at RSR it middotJOuld ippear that the arsenic content of themiddot
lead feedstock is low andor that the microants system for reducing lead
emissions is dlso quite effective for arsenic ~SR it will be recalled from
Section 32 uses a quenchinu cnariber ctnd Dayiwuse filters to remove
pctrticuldte matter and a carbonate scruooer to remove sulfur dioxide
There are no federal new source microertormance standards for lead
eini ssions per se t10wever the Standdrds of Pbullrformance for Secondary Lead
SmeHers (40 CFR 6Ull2) limit total particuLte emission from a blast furnace
185
or reverberatory furnace to SU mgm 3 According to Augenstein et al (1978)
who reviewed the techno1ogy for control ing lead ~missions from ttwse sources
baghouse filters or wet scrubbers are yenerally used to contro1 particulate
emissions When fabric filters are used to control blast furnace emissions
they are nurma11y preceded by an afterburner which inciner~tes hydrocarbons
that bull~GU Id otnerwi se b 1 ind t ne f o1 ur-i c After-burners a re nor necessary for
reverb2middotmiddotatcry furnace emiss-ion smiddotnce the excess air and temperature are
usually sufficient to oxidize tn2 hydrocarbons
According to Augenstein et ali 11 sr1aker--type baghouse filters are
the 1nost effective means of controlling ead fume emissions from secondary
furnace opErations 11 Collection etficiencies can exceed 99 oercent One
advantage to this control approact is that l~ad oxide dust can be recovered
easily cmd recycled in the smeller Flue gases must be coated to below 300 degF
for dacro~ bags and to below ~00 degF for fiberglass bags (Hi~h et al 1977)
Although wet scrubbers are effecti1e under some circumstances in
controlling lead emissions it is more difficult to recover the lead oxide for
recycling In addition sulfur dioxide presi~nt in the flue gases becomes
oxidized to sulfuric acid and can cause corr1sion problems For this reason
sodium carbonate or other basic reagents ar~ added to the scrubber solution
Although it concerns a yold smelter an approach described by
Marchant dnd Meek (1980) provides an exampli~ of an arsenic control
alternative which might be apmicrolicatgtle to secmdary lead smelters At the
Campbe1 1 Red lake Gold Smelter in Balmerton Ontarion Canada Smelter gases
are first passed throuyh a hot electrostatic precipitator (ESP) The ESP is
heated so that the arsenic (which is principally in the form of As ) remains2o3 gaseous and is not yet collected This exclusion of arsenic allo~~s the ESP to
recover particulate gold rnore easily The ESP exhaust is then quenched with
ambient air to condense the arsenic trioxide Baghouse filters then remove
the arsenic along with other particulate matter
124 CUNT~OLS FUR STEEL MILL EMISSIONS
Given the imminent and irreversibli~ cessdtion of coking activities
at Ka 15 er Stee 1 Corpora t i on a rev I e~ o r t e c Imo 1 o gi es for ( on t r o 1 l i ng
~missions from the coke ovens rnd he Clt1ke byproduct recovimiddotry pldnt WdS not
deemed to be necessary Tnis is t1e only primary steel mill in California
1Pi6
KEFUENCE)
Al lf~n Jr Ll~ llJBU1 J 111odel to 1middotstimite hi1drdous (bullmissions from coke oven Juurs pr~pdrelt1 DY 1-kSecircn Tr1dnye7nrffut fur U~ Environmental Protection Agency Uffi ce of Air ~ua 1i ty and Standards fltesearch Tri ang I e Park Nortn Cdrolina KTl17362-UL
Jl len Jr CC 1~8Ub tstination of chdr in(i emissions for by-product coke oven microrepared by fltesedrch Tr1anglt Insc1tute for US nvironmental Protection 1-yency Uf f ice of Air l)ua l i ty and Standards Ke search Triangle Park North Cdrolina RTI17362-0
Allen Jr CC 198Uc 11 Environmeritdl c1ssess1ent of coke by-product recovery plants Proceedings First Symposium on Iron c1nd Steel Pollution Abatement Technology Ctncayo Illrno1s--Tin1ctober - 1 November 1979 EPA-600-9-80-012 pp 7~-dd
Augenstein ur1 T Cornin et al 1978 Con_trol techniques for lead air emissions prepared by PEDCU Envircnmental Inc for US Environmental Protection Agency Research Triangle Park North Carolina EPA-4502-77-012-B ( NT IS P88U-197551)
jarber WC 1977 Interim air pcllution a~sessment for eight t1igh-volume industrial organic chemicals 11 US Environin~ntal Protection Agency memorandum to Air and Hazardous Materials Uivision Kegions I III-X and to Uirector Environmental middotPrograms Uivision Region II
t3drrett KE WL Margard et al 1977 Sampling and analysis of coke-oven door emissions Prepared by ~dttelle-Columbus Laboratories for US Env 1 ronmenta I Protection Agen y Industri a I Environmental Research Laboratory Research Triangle Park North Carolina EPA-b002-77-213
t3 a r ret t flt bull r bull and P bull K bull ~~ebb 1Y78 bull 11 E f f e ct i venes s of a wet el e ct r o stat i c precipitator for controlling PUM emissions from coke oven door leakage Presented at 71st Meetiny of the Air Pollution Control Association Houston Texas 25-29 June 1978
l3asdekis HS 1980 Carbon Adsorption Control Uevice Evaluation IT Enviroscience Inc report in prepdration for US Environmental Protection Agency ESEU
8ee fltW et al 1974 Coke oven charging emission control test program Vol I c1nd II US Environmental Protection Jgency EPA-6502-74-062
~eimer RG 1978 Air pollution emission test Analysis of polynuclear aromatic hydrocarbons from coke oven effluents Prepared by TRW Uefense and Smicrodce Systems Groumicro for US tnvirorunental Protection Agency Office of Air wuality Planning and Standards Kes~arch Triangle Park North Carolina EMB Kemicroort 8-CKU-12
187
Braman RS 1976 Application of the arsine evolution methods to environmental analyses presented at the International Conference on Environmental Arsenic Ft Lauderdale Florida 5-8 October
Bratina J E 1979 11 Fabric filter applications on coke oven pushing operations J Air Poll Cont Assoc 29(9) lll6-920
tiuonicore AJ 1980 Environmental assessment of coke quench towers in proceedings First Symposium on Iron and Steel Pollution Abltlternent Technology Chicago~ Uirnois JO October - f N~vernber 1979 EPA-6009-~1b-62~ pp 112-142
t3usse A D and dR Zimmerrao1 1973 User 1 s guide for tdegh~ Climatological Uisper~ion Model US Environmental Protection Agency Resea~ch Triangle Park Nortl Carolina) EPA-R4--73~0024
California Air Resources 8oard 1976 Coke oven emissions miscellaneous emissions and their control at Kaiser Steel Corporations Fontana Steel making facihty Enforcement l3ranch) S2crcmento California L amp E-76-11
Calvert~ CA 1976 Envir __ Sci_ i( __ Technol 10256
Coco L~i 2t aL~ 1979~ 11 02c10pmert of trec1tment and control technology for 11refractory pet rochemi co was tt=1s r middotl 1d l Report Gu 1f South Research In st it ute
to U S tnv i ronmenta l Protection Agency EPA-ti002- 79-080
Colucci~ LM and CR Begemi 1971 Palynuclear aromatic hydrocarbons and other pollutants in Los Angeles air In Proceedings of the International Clean Jir Congress Vol 2 Academic Press pp 28-35
Cooper JA and JG Watson Jr 1980 Receptor oriented methods of air partkulate source apportionment 11 J Air Pollution Control Assoc 30(10) 1116-1125
Coutant R W JS McNulty and RD Grammar 1975 Final report on determi 11at ion of trace elements in a combustion sys tern Prepared by Batte11 e Columbus Laboratories for the Electric Power Research Institute EPRI 122-1
11 11Dilling W 1977 lnterphase transfer processes Envir Sci TechnoL 11 4 pp 405-409
Oow1irgl) MP JO Jeffry and AH Laube 1978 11 Reduction of quench tower emissions Presented at the 71st Air Pollution Conotrol Association Conference Houston Texas 25-30 June APCA No 78-92
EPA 1980a Review of Criteria for Vapor Phase Hydrocarbons 11 US Environmental Protection Agency ~eport 6008-80-045
EPA 1980b Organic Chemical Manufacturing Volume J Storage Fugitive and Secondary Source EPA report 450J-80-025
EPA 1981 Compilation of Air Pol lutdnt Emission Fdctors (Including Suppleshyments 1-7) Third Edition Suppl1111ent 12 Utfice of ir l)uality Planning dnd Stdndlt1rds AP-42-ED-3-Supp I - lt
188
Erikson Uli 1~80 11 Control Uevice Evaluation Condensdtion 11 IT Envioscience lnc report in prepdration for US Environmental Protection Agency ESEU
lrtel liL 197Y Quench tower particuldte emissions J Air Poll Cont Assoc 2Y(9) 913-916
Ev a n s K bull P bull 1981 Pe r son a l Commun i cat i on v i th R bull Zi s k i n d bull
Goodwin DR 1980 11 Air pollution emissions standards in Proceedings First Symposium on Iron and Steel Pollution Abatement Technology Chicago 11 lrno1s 30 October - 1 November 1979 EPA-6009-80-012 pp 37-45
Gordon GE 10 Receptor Models Environ Sci Tech 14(7)792-800
Gordon KJ 1976 11 lJistribution of airborne polycyclic aromatic hydrocarbons throuyhout Los Angeles Envir Sci Technol 370-373
Great Lakes Cdrbon Corporation 1977 Study of coke side coke-oven emissions~ Vol 1 Source testing of a stationary coke side enclosure EPA-340l-77-014A
lirimsrud EP and HA Rasmussen 1975 Atmos Envir Y 1010
Hartnan MW 1~80 Source test c1t US Steel Clairton Coke Ovens Clairton P~nnsylvania Prepared by rRW Environmental Engrneerrny D1v1s1on for US lnv1ronrnental Protection Agency Emissions Standards and Engineering l)ivision Research Trianyle Pdrk North Carolina EM~ Report 78-CKU-13
Hendriks Ilt V et a I 1Y7Y Urgani c air emissions from coke quench towers 11
Presented at 2nd Arrnual M(etiny ot the ir IJol lution Control Associdtiont Cir1cinndti Utlio June 1~7lJ l-dmicroer NU 9-391
Higt1 MU ME Lukey dnd TA Li Puma llJ77 Inspection manual for secondary lead smelters premicrolt1red b Enyineeriny Science lnc for US lnvironmentdl Protection Ayency Office 0f ~nforc~nent EPA-3401-77-001
Holzer G et al 197 11 Collection amp analyses of trace organic emissions from natura I sources 11 Journal of Cnromatogra flhY 142 pp 755-764
Science Applications lnc Final lemicroort to US Environmental Protection Agency tPA-560lJ-80-UU2
J bull Mi l n e l lJ 81 Person a I Cornm un i cat i on wi t t1 bull Zi ski nd bull
Kemner WF 1979 Cost tffectiven2ss model for pollution control at coking facilities Premicroarect oy PEDCo Environmental lnc for US Environmental Protection Agency Industrial Environmental Research Laboratory Research Tridnyle Park Nortll Carolind EPA-GU02-7l-1U5 (NTS PB 8U-118706)
Kessler T AG Sharkey Jr and kA Friedel 1973 Analysis of trace ele111ents in coal by smicroark-source mass spectrometry US 8ureau of Mines Report ot I n v es t 1 gd t 1on s No bull 77 14 bull
189
Kolak NP J Hyde and R Forrest~r 19 1 9 Particulate source contributions in tt1e Niltlgara Frontierffi Prepared t)_y Nr2bulll York State IJepartrnent of Env i ronm~ntc i Con serv d ti on for US En vi r inment ct l Protect ion Agency Region 11 EPA-9024-79-006
Mackay U and PJ Leinonen llt175 Hate of Evamicroordcion of low - solubility contaminants from vJater bodies to c1tmospmre 11 Envir Sci Tecnnol 9 13 pp 1178-libU
Magee tiv HJ Hctll and GM Vc~rlt_a Jr 1973 Potential Doimiddotiutants in fossi1 f1Jel Prepared by ESSO fbullt~scarcr middotnd Engineeriny Co tor US Enviromihm~ct Protection Agency Ufl-R2-7 --2l1~L (NTIS Pt3 as u~9)
Marcriant middotG0H ard RL Meek 1~gu Eva uation of technology for control of arsenic tmissfons at the CdlmpJel Hd Lakmiddot G~ld Smelter prepared for the US
1Env i ronmentd l Protection Agenc)-l-l(lU-sfrT l lnv i ronmenta l kesedrCiI Laboratory Cincirrn2ti Uh70 lYA-60U2-8U--141 (NTIS 8 oU-21Y36J)
Margler L$ HA Ziskind M8 lcgozen (Ind M Axelrod 1979 11 An inventory of care i nog)n i c substances re i ea sect into ne ambient air of California Vol une I - Scn~enin~ and id2nt~fication o-f r~1rci1ogens of greatest concern Science J- pp 1i ca middott~ L J s In c bull Fi nd l kcmicro~ rt ~ i I - U 6 C) - U- 5 0 4 to U1 e Ca I i ~- o r n i a Ai r kesources 130ard
Milne J 198L Persond1 C~mmunicdtmiddotion middot1itt1 R Ziskind
llurchio JC WrC Cooper and A ue Leo11 1970 Asbestos fibers in ambient air of California 11 Cal Horn i ct Air lesourc2s l3oa rd ARl3 4-0S4- l EHS Report 73-2 (March)
National Academy of Sciences 1Y72 Part1culate polycyclic organic matter Washington~ DC pmicro 4-ll
Nati ona I Kesedrcn Counc i I 197b Arsenic prepi1 red by Subcommittee on Arsenic Co111mittee on 1bull1edical ctnd t)ioloyicai Effects of Environmental Pollutants washin~ton uc
Oliver JF and JT Lane 1979 11 Control of visible emissions at CFampl 1 s coke plant-Pueblo Colorado J Air PolL Cont Assoc 29(9) 920-925
Pe I I i z z a r i E bull 0 bull anct J bull E bull l$ u n c h 1s 19 1 nbi en t ai r ca r c i n o gen i c v a po r s Improved samplin~ and analysis tectrnology EPA Contract 68-02-2764
Pellizzari E U 1977 11 Tle 11easure11ent of carcinogenic vamicroors in ambient middotatmospheres EPA-b007-77-J55
Phelps RG 1Y79 lSI-CPA-oatLlle c-ilte oven door sealing proyram J Air Pol 1 Cont Assoc 29(J) JlJ8--JlL
Porter RA 1976 Uismicroersion ~yuation s lucions Dy calculator I guide for air pollution engineers ctnd sci12nthts S cond edition rexas Air Control tioard ~MT1S P~ 262 lJ~)
Redburn T 1980 Kaiser battles aginst l~gacy of woes 11 Los Angeles Timesmiddot (7 Semicroternber irno) Part VI pp 111
Research Triangle lnstittJte 1977 11 Qudntification of benzene in 15D -1bullnoient air samples 11 for US Environmental 11rotection Agency Office of Air Quality Planning a11d Stitndards
Rinkus SJ and MS Legator 1979 Chemical characterization of 465 known or suspected carcinogens and their correlation with mutagnic activity in the Salmonella typhimurium system Cancer Research 393289-3318
Koberts R M 1980 11 An inventory of arc i nogen i c substances r~ leased nto the ambient air of California Tasks 11 and IV KVB Final Report KVB 26900-836 to Science Applications Inc
Rogozen MB LW Margler P Mankiemiddott1icz cind M Axelrod 1978 Water pollution control for coal slurry pipeiines Prepared by Science Appl1cat10ns Inc for US Department of Energy (NTIS SAI-068-79-516
Rogozen MB DF Hausknecht and RA Ziskind 1976 Methodology for ranking elements in fossil fuels according to their potential health impact Science Applications Inc Final Report 260-77-539 to the Electric Power ~esearch Institute
Russel 1 JW and LA Shadoff The sdmpl ing and determination of halocarbons in ambient air usin~ ~rncentration on porous polymers
Siebert PC CA Craig and Eb Coffey 1978 Prelimary assessment of the sources control and population to airoorne polycyclic organic matter as indicated by benzo(d)pyrene Prepare( by lnergy and Environmental Analyses Inc for US Environmental Protection Agency Office of Air Quality Planning and Standards Resedrch Triangle Park North Carolina
Simmonds PG et amiddot1 1974 Distribution of atmospheric halocarbons in the air over Los Angeles basin Atmos Envir Vol 8 pp 209-216
Singh HB LJ Salas and RE Stiles Distribution of Selected Gaseous Oryanic Mutagens and Suspected Carcinoyens in Ambient Air in Proceedings Annual Meeting Air Pollution Controrl Association 82-651 June 20-25 1982 New Url eans
Smith WM 1979 Evaluation of cok oven emissions in Yearbook of the American Iron and Steel Institute pp 163-179
SRI International 1980 Asessment oi human exposures to atmospheric benzene SJ Mara and SS Lee finai report to US Environmental Protection Agency EPA-4503-78-031
Taback HJ 1978 Control uf Hydrocarbon emissions from stationary sources in the California South Codst Air Basir Prepared by KVB Inc for the Cal1fornid Air Resources Bod-rdmiddotKVB No 5804-714
191
Trenholm J-f dnd LL )eek 1Y78 11 Assessment of t1azardous organic emissions fror1 slot type coke oven Datteries unpublished paper US Environmental l)rotection Ayency Ernssion Standdrds and Engineeriny Jivision
Turner U~ J970 Workbook of atmusphe~ic dispersion estiimiddot1Jt2s IJS Environmental Protection Agency k~sedrch Tri1ngle Park North Carolina
US Bureau of Mines lYl f-inercls yearbook 197U Vol 11 p 4~3
Van Os dE I l U bull iJ bull lL Ma rs l and et ct 1Y7~1 o En i r onmt~ nta I a s s es smen t of co k e Dy-product recovery plants prepareu by ReseMcn Trictngle lnstitute for US Environmental Protection AgerKy 11cJustrial Environ11ental R~search Laboratory Kesearcll Tri2n~e Idrk NorU1 Caroline EPA-6U02-7J-Ul6 (NTfS PB 293-278)
Westbrool( C W 1979a Level l ass~ssment of uncontrolled sinter plant emissions Prepared by fesearch Triangle Institute for US Environmenta1 Protection Agcncy Industria1 tnvirnnmental fltesearc-1 Laboratory kesearch Trianyle Park North Carolind EPA-bOU2-7Y-112
Westbroollt C~~ 197ltJb Level l css~ssment of uncontro1led l)-BOP emissions Prepared by Researcn Triangle institute for US Environmental Protection Agency Industrial Environmental Hesearch Laboratory Research Triangle Park ~Orth Carolina EPA-6002-79-190
Wetherold rLb LP Provost and CD S111ith 198U Assessment of atmospneric emissions from microetroleu111 refinin~ Volume 3 Appendix l3 Radian Corp Fina Report to US tnv i ron111enta I Protection Agency EPA-600 2-80-07 5c
Ziskind RA OF S1ilith JL Hdhn etnd G Smicroivey (1980) Determinants of Cancer and Carctiovasculdr Uisease Mortality in Asbestos Mining counties of California SAi Report 068-dl-514 l May
192
APPENDIX A
Rapid Screening and Identification of Airborne Carcinogens of Greatest Concern in California
Lawrence W Margler Michael B Rogozen Richard A Ziskind and Robert Reynolds Science Applications Inc Los Angeles California
This paper describes a method for establishing a priority list of airborne carcinogens within a
state jurisdiction In this case it was necessary to identify from among hundreds of potential
candidates the live to ten materials of greatest potential concern In California as airborne
carcinogens
Because no previous Inventory of carcinogens In California existed published lists rankings
and assessments of national scope were used to Identify candidates By systematic manipulation
and comparison of these data sources 47 materials of some notoriety were chosen for closer
scrutiny This selection was pared to 22 candidates largely by eliminating those which had
very little production and use In California (Substances primarily used as pesticides were
excluded from the scope of this study) The remc1ining candidates were then ranked by additive
and multiplicative algorithms and by a panel of experts The results of these rankings were
combined to produce a single selection of 11 priority candidates In alphabetical order they
ethylene dichloride N-nllrosoamlncs perchloroethylene and polycyclk aromatic hydroiarbons
In continuing studies a baseline emissions inventory Is being prepared and a source testing
program is being designed
In recent years concern has grown over su~pected arcinug-ens New Jersey seshythe possibility that certain materials lected ten volatile organic compounds released to the atmosphere through inshy and five htbullavy metals to be examined dustrial and commercial activity may be further and is currently measuring responsible for a significant portion of ambient atmospheric concentrations of the incidence of cancer in the general these substances in a variety of areas In population This concern is manifested the second year of the study the state at the federal level in the National h1s incre1sed the volatile organics Emissiltm Standards for Hazardous Air stt1died to 20 and begun measuring Pollutants (NESHAP) which limit heavier orianics associated with parshyemissions of the known carcinogens asshy ticulatcs2 In California a very different bestos beryllium and vinyl chloride 1 approach-was taken
Only a few states including New Jersey and California have begun efshy Overview of Californias Approach forts to identify airborne carcinogens of concern to the general public for the The California Air Resources Board purpose of setting state emission regushy (( ARB) is sponsoring a three-stage lations for these substances After reshy st 11d~bull of airbornC rmissions of carcinoshyviewing national use data for known and g1middot11s from anthropogenic atlivities The
November 1979 Volume 29 No 11
first stage which is the subject of this paper was to identify roughly five to ten materials which of the hundreds of known or suspected airborne carcinoshygens are most likely to be of greatest concern to Californias general populashytion Also of interest were those which in order to satisfy occupational health and safety regulations might be transshyferred from the workplace air to the outside atmosphere The second stage which is now underway includes pinshypointing of emission sources for each of the carcinogens of importance quantishyfiltation of emissions nnd design of source-testing methods A subsequent stage will consist of source testing and measuring public exposures to those substances for which data are unavailshyable The basis for regulatory action if appropriate will include the results of this program and other related reshysearch
Screening of Candidate Carcinogens
The National Institute for Occupashytional Safety and Health (NIOSH) lists 1905 chemicals which have reported neoplastigenic or carcinogenic effects and 510 which have otherwise received attention for their neoplastigenic poshytential1 The need to select five to ten materials from such a large number of potential candidates dictated that we devise a way to rapidly eliminate from further consideration the vast majority of the substances Given the paucity of published data on most of the candidate substances the screening method was designed to make best use of readilymiddot
C1111yrich1 19i9-Air Pullulion Control A~~ociatlon
1153
A- I
----- ---- ----- ---------------------available information The screening process was as follows (1) eight comshyplications of known and suspected carshycinogens were reviewed and those subshystances which were not used in Califorshynia were highly unstable in air or were very doubtfully carcinogenic were eliminated (2) after more detailed inshyformation was obtained for the reshymaining 25 substances candidates were rated by two different analytical methshyods (3) an expert panel was convened to review dossiers on the candidates and to independeitly nmk them (-i) from the eight to deven substances ranked highest by all three approaches eleven were selected for the emission identifishycation and source-testing design stages of the CAHB effort
lnitiI $cre~ng ot Potential Candidate Carcingtgens
65 compnunds middotwe ire selected from 642 industrial ur~nnic air pollutants comshypiled by MrTRE Corporation for the US Envircmnental Protection Agency (USEPA) in U1at study pollutants were scored by multiplying four explicshyitly defined rating factors annual US production fraction of production lost to the environment volaflity and toxshyicity To adapt this york to our pmpose we first selected the 114 substances listed as being carcinogeni 1~ or neoplasshytigenic Then the scores of each of these compounds under the criteria annual US production fraction of producshytion lost volatility and carcinogeshynicity were multiplied together Seshylected for further consideration were those substances which had a product score above 50 Another 15 substances listed as being carcinogenic but lacking information for one of the other rating factors were also selected This list of 65 was then compar~d with seven other lists of corcinogeris1
-middot 11 Materials comshy
mon to the reduced MITRE list and at least one of the other lists were chosen for further consideration Added as candidates were those suhstances which are regulated as occupational carcino-
gens by the Occupational Safety and Health Administration (OSHA) and certain inorganic carcinogens Finally substances were added which in our judgment should be investigated but had been eliminated at this point Exshyamples of these are bis(chloromethyl)shyether epichlorohydrin and hydrazine
Next the refined list which now contained 47 substances or chemical groups was pared further hy another rapid screening process Eliminated
were all candidates (1) whose production andor use in California was very low (under 10~ lbyr) and wns not thought likely to pose a risk to a localized popushylation (2) which are very unstable in 1ir or CH whilh should not on t lw hnsis of CUtrltbullnL cidlnre liebull considtbullnmiddotd r11rci-
1154
Tahllt I Suh tances n-iewed in letail
Candidate Substances
Arsenic Inorganic lead Ashestos Alkyl lead Benzene Maieic anhydride
nogenic The result of the initial the rating under that criterion and the screening was a list of 22 candidate mashy corresponding criterion weight The terials which is presented in Table I overall rating for pollutant i is then the
sum of the scores under all the crishyteriaRanking Candidates by AddiUve ancll
Mulllpllcatlve AigorUhms The additive approach has several virtues the main one being that it forces
Many screening or ranking systems the user to make all assumptions exshyfaH into one oft wo cat~middotgories additive plicit In the process of setting up such and multiplicative Some systems are a a ranking system new insights into the combination of the two while others problem under consideration may be combine nn ol1jective appronc 11 with g1ined Once the system is set up it is suhjlctive cvnluition of the remiddot ults 1l rdatively easy to use Where data for The L2 substances surviving th1middot initial scoring pollutants are unavailable arshyscreening were ranked independtmiddotntly by tificial scales can be constructed to the tvo approaches If the same subshy quantify subjective factors Finally the stance rated high~middot under both systems sensitivity of the results to the systems its importance to California was judged subjective aspects may be measured For to be more likely than if it had scored example one can determine the effect of highly in only one method changing criteria weights upon the final
Additive Approach In the additive pollutant ranking Similarly an appreshyapproach the user identifies one or more ciation may be gained of the significance criteria and rates each alternatiw subshy of the range of uncertainty for a particshystance against each criterion while sishy ular required data element by varying _ multaneously deciding the relative imshy rating factor values portance of the criteria Eq (1) shows its A fundamental problem with the apshymathematical formulation proach is that there is no logical basis for
adding the individual scores assignedRating for pollutant i = f W1 R11 under the criteria other than the asshy
rbullI sumption that this simulates or even
(1) improves upon the users thought proshyEach criterion or rating factor (1 ) is crss A major operational problem is assined a value for each pollutant and that of weighting the criteria A common each rating factor is weighted (b W1 ) practice is to give all critNia equal -icnrding to its importance reht veto wltgtight hut that is in itstbulllf a st1Ubullm1middotnl lhe otlwr rritnia llw sc11rtbull for 1111 aLnut the 1rcl11tive importance of Ow 1n1 1 undmiddotr nitn1 n j is the prod11middott of n1ttria
Journal of the Air Pollution Control Association
A-2
---------
Multiplicative Approach In the multiplicative approach the rating for each alternative is the product of the rat ihgs under each criterion
Hating for pollutant i --- 11Ill
UJ (~) Jbull I
A multiplicative approach can have ~omc altlvrntnges over additive onrs First in some cases the ratings can be physical parameters such as concenshytrations or volatilities there is then no need to weight the criteria and hence less controversy over subjective judgshyments Second multiplication generally provides a wider range of scores than does addition allowing clearer disshycrimination among alternatives Finally this approach provides results which are more intuitively acceptable As an exshyample of this last point suppose that exposure and harmfulness levels for candidate substance are each converted to values on aO to IO scale and that a certain substance is both extremely toxic and extremely rare An additive approach would give the compound a rating ofO + 10 = 10 which is equivalent to that of a moderately prevalent subshystance (rated say at 5) which is modshyerately harmful (rated also at =-) A multiplicative approach on the other hand would rate the first substance at 0 and the second at 25
Criteria The six criteria used in the additive and multiplicative ranking procedures are defined in Table 11 Asshysignment of values to the Rij was based upon data gathered from published litshyerature personal communications and panel discussion and has been fully documented 13
Because the purpose of this exercise was to determine the relative imporshytance of the suspected candidate carshycinogens R 1 was scaled to the most heavily used candidate substance benshyzene whose annual production and use in California is nearly 109 lb Materials with a use under 105 lbyr would be rated zero for R 1 and rejected Howevshyerbefore rejecting a substance by this criterion we considered whether its emissions could in particular circumshystances result in high exposures to a Joshycalized population
R2 takes into account the fact that the chemical industry is in continual change Substances of concern today may be phased out while the use of others may rise dramatically increasing their imshyporta nee asmiddot poll utan ts Information on developments which could likely result in a change in the growth rate was facshytored into the choice of a value for R2- As an example asbestos consumption in California has been stable in recent years However the pending phaseout of asbestos in motor vehicle friction materials will hasten the decline in asshybestos consumption hence we assigned a value of 1 for R2bull
November 1979 Volume 29 No 11
Ideally one cou1 1 use pouu1c11u
sion as a criterion However in this case emissions were to l e estimated in detail only for the five t1 ten carcinogens fishynally ~elected Thmiddotrtfor(bull a 11w1st1n of tt II iim pol cint ial vus 1~1middotd 1atbulld llpon
knowledge of the s11bstances manufacshyture and USP for R The higlwst rating went to substanc middots which are v idely used especially in consumer products A slightly lower rating went to subshystances which an~ routinely emitted from industrial processes during proshyduction and use Some materials are employed in such a way that emissions are quite low even though tight emission control may not be required by law Materials in this cntegory were assigned a value of two for H3bull Substances which under federal or st ate regulations may not be discharged to the exterior envishyronment but which could be dischJrged by accident received the lowest rating
Each candidate was evaluated (n the basis of its 1ropen-ity to decompbullgtSe in ambient air Materials with half-lives greater than eight hours were considered moderately to highly stable and rated five for R4 bull Low to moderate stability was assigned to substances with halfshylives between zero and eight hours Compounds known to exist in air for only a few minutes would be rated zero
Ta hie II Dcfinitinns of the criteria used
R1 Prbullbullsent use in California
100 of max (109 lbyr) 5 10 1)f max (108 lbyr) 4
1 of max (107 lbvr) 3 01 of max (106 lbyr) 2
001 of max (105 lbyr) l lt001 of max (lt105 lbyr) O
R2 Growth in California use
+ 20 5 +Oolti to +20 4 Positive growth to 0 3 Stahle or unknown 2 DPcline 1 B1middotin~ phased out 0
R Emision potential
Widespread use in ltonsumer products 5 R1middotlatively p1or con rot over emissions 4 Rdatively good control over emissions 2 Tihtly c1int rolled 1
R4 8tabilit~- in ambient Air4 tvlnderale t1 high st 1bility (l 1 middot2 gt 8 hr) 5 Lw to modmiddotrate s1 bility (t1 l -0-8 hr) 3 Unstahle(t2~febull minutes) 0
R Dispcision Potential
Emitted lar~ely as apor or iine 5 particulnte
Emit led largely as rnarse particulate
Rlt Evidence f Carcinu~enici ty
Known or suspected human carcinogen 5 Known mammalian carcinogtmiddotn 4 ~usptctcd 111ammalan carcino~en or 3
known m mmalin muta~tmiddotn Ames ltmiddotst p sitive 2 Prpoundgtcursor 01 co-car inogen
a t11 is the half-liftmiddot
A-3
tion state or anion associations may change in the atmosphere metals do not degrade and were considered stable Asbestos is likewise itahlc Many of the clcc-ompo~ilion rtbullnctions oi orinnic molecules are mediated by light Such substances if released at night would have several hours to disperse in the surrounding area
A rapid way of assessing the relative potential of different substances to spread from a release point is to note their physical state under normal amshybient conditions Accordingly we scored materials emitted as vapors or fine particulates the highest for R5 and coarse particulates the lowest Intershymediate values are possible for varying amounts of fine and coarse particulate emissions from the same source or from different sources
There is as yet no widely agreed upon measure of the relative potenciemiddots of carcinogens although some ranking systems have been proposed14 Extrapshyolating data from in vitro techniques such as the Ames bacterial mutagenicity test and from laboratory animal studies to humans is problematic Therefore a less quantitative measure of the carcishynogenic potential of each candidate substance was used The candidates receiving the highest scores for Rs would be those for which there is strong evishydence of carcinogenesis in humans Exshyamples are asbestos which is implicated in mesothelioma vinyl chloride which has been identified as the agent of liver cancer in exposed workers and bisshy(chloroinethyl)ether shown by epideshymiological studies to cause lung cancer in resin workers The next highest rated substances are those for which human carcinogenicity is unknown but which have produced cancer in one or more mammalian species in laboratory tests Next are those which have not been shown to be carcinogens but which have proven to be mutagenic in test animals Substances for which the only knowlshyedge of carcinogenic potential is a posishytive Ames test (producing mutations in histidine-requiring strains of Salmoshynella) are rated 2 Finally substances which are implicated only as precursors or co-carcinogens would be rated lowest
Substances unequivocally associated with carcinogenesis were considered as carcinogens in this study Conditions of emission and exposure including the presence of co-carcinogens were facshytored into the evaluation of each candishydate where possible Carcinogenic subshystances derived from the metabolism of a precursor were considered as carcinoshygens However ubiquitous substances which have been hypothesized to be precursors (eg secondary amines nishytrous acid and nitric oxide which combine under certain circumstances to
1155
form N-nitrosoamines) were not conshysidered because of uncertainties in the importance of their link to the carcinoshygenic compound and the practical conshysiderations demanded by the scope of the study
It was beyond the scope of this study to jud~e the validity and interpietation of the experimental ad epidemiological evidence upon which the carcinogenicity of candidate substances has been esshytablished We accepted the conclusions about carcinogenicity drawn by the Inshyternational Agency for Research on Cancer or the National Cancer Institute and did not consider dosage or route of administratioll of tested substances However considerations of test validity did enter into the subjective evaluation by the panel of experts
Weight lr1 the sdditive ranking scheme each rating criterion is weighted according to its limportance relative to the other criteria Little precedent exists for assigning these weights In our judgment and as generally agreed by the panel of experts (see below) R 1 RJ and R6 are more important than the other criteria Evidence of carcinogeshynicity was considered to be the most important criterion of aB so W 15 was assigned a valtze of 3 W 1 aid W 3 were set at 2 and W W-i and W5 were set at 1 In order to discern the potential senshysitivity of the rankings to the weight assignments the candidates were also ranked using equaily weighted crishyteria
Ranking Candidates by Panel o~ Experts
Anine-member panel of experienced governmental industrial and academic scientists whose disciplines include~ -organic and physical chemistry indusshytrial hygiene toxicology epidemiology and iegulatory control of toxic sub~ stances was convened to provide addishytional data for our ranking algorithms
to discuss our candidate --ubstanc -sand rejections to suggest pos-ibie ne suushystances for consiceration and Lt rank the candidates indepen11mtly ( f our own ranking Tvo v-ee ilts he fore the meeting pa-d uember-s were jven one-to three-page dossiers on eac I canshydidaie substance
At th~ start of the meeting- before any group discussion the pand was asked to rate each candidate suhtance with a score from Oto 5 Next cmiddotach candidate was discussed at length e provided ~n ovrview and summarized critical issues identifi2d up to that Pbullgtnnt Through materials brought to thtmiddot meetin and their persoril experience panel memshyber Veimiddote able to provide much useful information on the candidates and adshyditional insight in to our ~a ting criteria
At the encl of the two-d1y sc~sion the pa-iel ugain -ated the cai 1didates
Flnal SelediDn
Tabe III show- the highest-scoring substance a~ det(rraine1 by the 1ddishytive and mlltiplictive approaches and by th~ panel in ~e additive approach 21 single r1nking as obuined by avershyaging the two ran kings resulting from using equal and uneguai weights The ran kings ltif most candidi tes were unafshyfected hut carbo11 tetralt hloride chloshyroform chromium and inorganic lead changed more tban thmiddotee positions_ Because uncertainties in the data base predude imputing signifiance to small differences in the Lnal ordering the lists in Table III are prtmiddotsented in alphabetishycal order However it is of interest to point out that benzene consistently ranked hihest
Becausimiddot some cindidates had equal rating scores we ltould nut choose exshyactly ten candidat ~s from the additive and multiplicative rankings Instead the
top nine and eleven were selected from the two exercises respectively We also considered the ten substances scored highest hy the panel at the end of the session The final consensus selection cornsistt-d of the 11 candidate substances appearing on at least two of the three lisL- For the substances included in the cun~ensus ranking a baseline emissions inventory is being conducted and a source testing program is being deshysigned
1ejected Substances
Some comments about certain subshySiF1ces not appearing on the final list are in order inasmuch as they include known carcinogens and compounds which hae received considerable atshyllttion in recent years as occupational c11c111ogens
rouisionally Rejected Substances Appended to the consensus ranking (T1ble III) were vinyl chloride gasoline and engine exhausts tobacco smoke and pesticides No further action by the CRB is recommended at this time for vinyl chloride because it is already subject tu the USEPA emissions stanshydard a CARB ambient air quality strndard and an OSHA standard
Gasoline and tobacco smoke were appcndlmiddotd to this list because each ocshycurs very widely and contains several of the candidate substances reviewed in this study some of which are in the final listing For example gasoline contains benzene ethylene dibromide ethylene dilmiddothloride and alkyl lead compounds the last three being in leaded grades only Both gasoline and diesel combusshytion products include PAHs Tobacco smoke contains among other neoplasshytigenic substances nitrosamines PAHs nickel arsenic cadmium and other heavv metals Manv individuals are inshyvolu~tarily and i~ many situations
Table HI Hi~hest ranked candidates from each rnnkin~ methodbull
Highly ranked but no inventory
l I ighesl consensus recommended Additive Multiplicativ(~ Pan I of eqwrts rnnkmg at this time
1156 Journal of the Air Pollution Control Association
A-4
lirtualy unavoidably exposed to t11-k1ltco smoke Because the sources of gasoline its combustion products and tobacco smoke emissions are well known no specific action was recomshymended for these materials during the tbullmissions innnlory and -ource testing design stages of the present study It was considered importint ho-vemiddoter to draw attrntion to the general publics exposhysure tot hrse substances PrsticidltS are listed for the same reason thou~h ad(bullmiddot tailed examination of pesticides w1s beyond the scope of this study Many pesticides are widely used and some of them are known to be carcinogenic
Other Rejected Substances Acryshylonitrile and vinylidene chloride were placed in a provisionally rejected group because of the panels suspicion that imports of these compounds to Cnlifornia from Japan may be appreshyciable yet are hard to substantiate Should such imports be verified in the future these two compounds would take on greater importance Formaldehyde was also provisionally rejected because the preponderance of evidence indicates that it is not carcinogenic and that bisshy(chloromethyl)ether is not formed from formaldehyde in appreciable quintities in industrial environments 15 The ocshycupationally controlled carcinogens ethyleneimine and beta-propiolactone were rejected in part because of their reactivity in air At the time of this study DBCP a pesticide was no longer heing middotproduced in California and was therefore rejected from further considshyeration
Occupational Regulations and Community Exposure
A question of interest to the CARB was whether the regulation of acknowlshyedged occupational carcinogens adshyverselv affects the ambient air outside the v~rkplace Our general findings can be illustrated by the example of asshybestos
Asbestos is a very widely used mateshyrial for which no ambient air standard exists Concentrations in the workplace are limited to an eight-hour timtmiddotshyweighted average concentration of two fiberscm=1 of air and a ceiling concenshytration of ten fihcrscm 1bull In meeting this standard exhausting air containin~ asshybestos to the ambient air is not reshystricted except by the USEPAs reshyquirement that there shall he no visible emissions containing asbestos piut ides from s11eh foci lit i1bulls (bullxd11din~ hrak(bull shops 1 Considning that undc-r certain conditions 101 asbestos fiherscm1 could ht Pmitted without being visihl( 11 this standard mamiddot allow considerable asshybestos emissicns It is unlikely however that emissions would actually approach these levels First since the OSHA standard cannot generally be achieved
November 1979 Volume 29 No 11
by ventilation n ine the generation or asbestos partid in the workplace must be greatly reltlu ed Second the air is usually filtered 11gt prevent recirculation of asbestos to 1 workplace Asbestos waste must bed 1sposed of in sealed imshypermeable bag~ or containers Thus under current onupational regulations the amhient air 1enerally appCars to be afford(bulld greatn protection than it would without s11ch regulations 1
Conclusion
The screenin and ranilting methodshyology presented in this paper proved to be a feasible approach to establishing a priority list of Jirborne carcinc)gens in California We ftel that it is an efficient means of focu~ing further efforts on emissi()ns inventories source testing and ambient measurement for it not only identifie all the carcinogens of potential concern but it also permits the state regulatory agency to direct its reshysearch resourCtmiddots toward those subshystances of partilmiddotular interest within its jurisdiction
References l National Elllission Standards for Hazshy
ardous Air P llutants A Compilation as of April 1 19 3 PEDCO Environmenshytal Inc for US Environmental Proshytection Ag1middotncy EPA3401-78008 (NTIS No I B 288 205) 19i6
S Gray Ne Jersey Department of Enshyvironmental l 1rotection Trenton NJ private com111unication 1979
1 H E Chrislllsen E J Fairchild and R J Lewis Sr middotmiddotSuspected Carcinogens A Suhfile of th1middot NIOSH Regii-try of Toxic Effects of Chemical Substances 2nd Ed National Institute for Occupational Safety and HPalth NIOSH i7-149
4 B 8 Fuller J Hushon 1 Kornreich R Ouellette L Thomas and P Valker Scoring of rganic air pollut1nts for US Environrnental Protection Agency Mitre Corpt ration T1middotchnical Report MTR-6248 I 9i6
S F R Lozan11 Toxicants Selected for Priority En- 1ronmental Assessment corresponde11ce with EPA Region VI Administrat r 19i7
1gt Survey of inyl Chloride Levels in the Vicinity of Keysor-Century Saugus California US Environmental Proshytection AJet middoty National Enforcement lnvestiiatim Center Denn-rand San Francisco E A-n02-77-017a 19i8
0 Bardin l middotstimon~middot in Hearings on Helationshi1 Bet wfen Cancer and the Environmei 1 US llltuse of Hepreshysentatives bull th Congress 2nd Session Committee (In Interstate and Foreicn Commerce ~crial No 94-141 1976 pp 7~i0
middot T H Ma11middoth Carcinogens in the workplace lwrE to start d1middotaning up Sc-imiddotrmmiddot l17 1-1110) lfi8 (197)
I CH ar1111 Classif1r11t1on of the Exshyposure Haza d Prtbullstgtnled Ii~middot Thirt~middot-Six Clwmirals 011 I llixllmmiddots lternorandum from Clllm1middot I Offi(middote of thtgt Assmmiddotiate lgt1rector for 1 middotarcinoltnesis Division of Cancer Cau~ l and Prevention lo SushyJwrmiddoti-or~middot ( 0 Auditor Gtmiddotneral Acshycounting Oft lmiddote19ii1
I0 Chemicals ith Sufficient Evidence of ( arcinogenit ty in Experimental Ani-
A-5
ma1s Jnlernauona1 Agency 1or 1c-curcn on Cancer IAHC Working Group Reshyport Lyon France 1978
11 Threshold Limit Values for Chemical Substances and Physical Agents in the Workroom Environment with Intended Changes for 1977 American Conference of Governmental Industrial Hygienists 1977
12 M H Rogozen D 1 Hausknecht and R A Zisk ind 1976 A Methodology for Ranking Trace Elements in Fossil Fuels According to Their Potential Health Impact by Science Applications Inc for Electric Power Research Institute 1976
1t L Margler R Ziskind M Rogozen and M Axelrod An Inventory of Carcinoshygenic Substances Released into the Amshybient Air of California Vol I Final Reshyport Screening and Identification of Carcinogens of Greatest Concern by Science Applications Inc for California Air Resources Board 1979
14 T H Maugh Estimating potency of carcinogens is an inexact science Science 202 (4363) 38(1978)
15 C C Yao and G C Miller Research Study on ais(Chloromethyl)Ether Forshymation and Detection in Selected Work Environments by the Bendix Corposhyration for National Institute for Occushypational Safety and Health Cincinnati Ohio Contract No 210-75-0056 1979
16 Occupational Cancer Control Unit Calshyifornia Department of Industrial Relashytions Los Angeles California private communications with staff members 19i9
This paper was developed while Dr Margler was a Staff SriPntist in the Energy-Environment Systems Division of Science Applications Inc (SAIi 1801 Avenue of the Stars Suite 1205 Los Angeles CA 90067 and Mr Reynolds was a Project Manager with the California Air Reshysources Board Sacramento CA 95812 Dr ltar~ler is currently emshyployed as DirEgtrtnr of Bnvironmental Scitbulllltcs for WinzlN nnd Kelly Conshysultini EnintNi t1 Third Stretbullt EurekH CA 9gtgt01 and Mr Heynolds is Air Pollution Control Diretbulltor for the Lake County Air Pollution Conshytrol District 55 N Forhes Street Lakeport CA 95-13 Dr Rogozpn is a Staff Scientist and Dr Ziskind is Deputy Manager of the Energy-Enshyvironment Systems Division of SAi
1157
Appendix f~
_l)ispobullition of Miscellaneous Sites
Gould Inc Ver_n)_~ St~condary Lead S~_n_elter
The emissions of arsenic from the four large secondary lead smelt~rs
in Cali forni) J1~1 ~ bull~tiilnted in the program This estimate was based upon a
uniform fugitive emission f1cJr nt WPll supported by measurement inforriation
Therefore source monitoring was r2cq11)nded 1nd Gould Inc Vernon being
~he largest was singled out It was however proposed to monitor both Gouli
and RSR Corp (Quemetco) in the City of Industry since analysis of the plants
revealed significant differences in plant equipment and engineering The
latter being more typical of 1 nodern facility
As a r2sult Jf pr-~-middott -iiscussions and plant inspections we became
aware that Gould was actively in the process 1)f constructing a new facility
which would completely replace the existing one We have monitored progress
on the new site and concluded it would he inappropriate to utilize program
resources to conduct field tests at Gould ~rr1issi 1rns from the new Gould
faci 1ity wi 11 be bas~rl 11p11) test results from RSR
PG ampE - Pittsburg PG ampE - Salinas a~d So Cal Edison - Long Beach Oil Fuel Power Plants
It was app ro p ri ate tJ cons i dP r the (~mission of arsenic from power
plants during the initial study stage inc~ imiddotrace quantities in the fuel oil
are known to be ernitted Because of the population distribution in tile
vicinity of three plants and some 11realistically conservative estimat2s of
erlissiJn c-1ctors tht~ facilities appearc~d among the top seventeen stationary
sources of potential c1Jn-r-n HovJ~ver as a result of a reexamination of
emission data it was concluded that an error ~dj ~een made in the material
balance of arsenic - resulting in tn emission factor equivalent of a 300
release ~io lit~rature was found to justify 1reater than a 30 transfer
function Thus we estimated at the outside th~ emission factor should be
reduced from 013 lb per 1000 lb to approximately 0013 lb and resulting
ars9nic 2missions for the entire state w~r~ cJ11servatively estimated to be
1760 lbs (from 17600) divided amongst 311 the states power plants Clearly
then the four secondary lPad smelters estimat~ of S9410 lbs of arsenic(
B-1
emissions per year makes consideration of poer pl int ernissio1s if ffsenic a
very low priority
In a literdture review by SAI - Methodokgy __f_or Ranking Trace
Elements in Fossil Fuels Accordi~9__t_~_their Potential Health Iripact (Roqozen
1976) - emissions estimates f0r 15 trac~ substances were researched for c11l
and fuel oil pomiddot1er plant c 1J1v---r-sion Source content combustion pr-)~~
transfer functions and contrcl methodology were co1~idered to develop e~ission
factors The output to input ~atio computed and utilized n -~i-i~ study for
arsenic was 002 to 03 (ie tra1sfer function betw~en 2 11d JO~) reflecti~q
the wide variety of fuels processses and controls nationwiJe The upper end
reflecti1g both high arsenic coals and poor e11iss10n contro 1 devices Clearly
neither of those conditions accurtely apply to th2 thr~~ si~~~ r1nrl thus
substantiate the rlecision not to perform emissions neasur-i-~n~s
Calaveras Asbestos Company - Copperopo l middotis Asbestos Mini n(i and Mi 11 i nq
In the past (late sixties and early seventies) ~h2 ~3laveras
Asbestos Company came under heavy criti-is1 1fter imiddot1sp~ction rieasurements
revealed serious problems However significant reduction of e1nissi0ns have
occured prompted by NESHAP regulation and occupcttional standards SAI
inspected the site in f)~c~1b-2r 1979 under an EPA contract An missions
inventory was published under that work (Ztskind 1980) and the bottom line
conclusion is that currently no significant ernissioi1 are gteing released as a
r~sult of hl-isting which would reach the publ le i 1le 0pen pit is some 900
feet deep with blasting at the bottom 0ver 80 percent of emissions in the
CARB Emission Inventory System were attributej to pit blasting In fact at
its current depth blasted 1aterial does not reach the mine surface with tile
explosive implacement used Additionally the site is more remote than might
be realized The ~ituation is vastly different today than in the past and
currently attention should be paid to the issues of occupational exposure and
breakdown of ventilation syste111 controls in t1e inilling operationsmiddot He
recorrrnended that no furtw~ corsideration be Jiven to this site for the
purposes of tl1is w)JJil ie identification )f silt~middot1ificant releases vhich
might he responsible for causi1J middot1 fmiddot middotgtbullgt middot 1= 1(11 1bull11 tinn Pxposure
8-2
Vctri0J5 Refineries
It was est=tblished early in the anrilysis prmiddotJr111 V1at gtenzene
emissions from refinery operations coulrl in th~ 1guregate constitute a
significrnt source Since there ilre a large rrnrnber 0f refineri~s in
California (46 in Los Angeles County itself -it t(~ i 1~ middot)f r-1ltariination) with a
wide variety of types sizes ages etc their eva l uat 11) 1 i =r~sent -3~ i)7
monumental task It was noted immediately t 13t ~~it hi n the total scope of the
study the design and conduct )f c r~f i r1~ y testing program which would develop
a complete benzene emissions in1~ntory for 01e site was impractical let alone
to d1aracteri ze emissions fro111 three ie those listed among the 17 rnost
s i g n if i ca n t potent i a l st a t i on a r y sou r c e em it 1-_ l rs bull
The three refineries were singled out for special attention in the
previol1s phase ~ecause they uriiquely had components which process materials
containing 10 or more per~nt hy middot1-bulli1 11t )-~112~ne (46 FR 1165 Jan 5 1981 pg
1491 ) Est i mates by E P A ( F e d bull r l Rr-~ g i s t ~ r 1q81 ) i n d i cate that 90 p e r cent o r
more of the total benzene fu 1Jitive emissions arise from such components
Therefore attention is appropriately focused on the three henzene production
consumption refineries Chevron (Richmond) Arco (Crson) and Chevron (El
Segundo) SAI staff considered the possibliJ )f -1 testing program at one of
these refineries and cone l url 1middotd it would not 1)e cost-effect i v2 for o number of
reasons
California is a minor producer and consumer of benzene with
approximately 15 of the 114 hill ion pounds produced and
consumed nation1lly in 1977 This can 1)e riY1pared with the fact
that California has approximately 10 of th~ ~ations population
Benzene exposur~ to the generil nJmiddot1lation has been partitioned
among the various sources Aproximately 90 of exposure is
estimated (SRI 1978) to be caused hy gasoline distribution
activities and 1ehiclEmiddot emissions in 1 ir~an areas with nearly all
of the balance )Y benene hrndl ing operations (refineries and
chemical plants) In the case middotlf California this percent1g~ Jill
be even JT1ore di middotpruJor-tionate b~cause of its greater s1are of the
population and ~esser share of the henzene ha~dling Furthermore
since the three refin 1bullry cites ire heavily urbanized no new rural
population cegmmiddotnt is beinq exposed
B-3
Californias most heavily urbanized Air Quality Management
Di st r i cts have a l ready adopted t en zen e f ugit i v e em i s s i on
staridards comparable and vith S( 1ine features VJr stringent than
the proposed national (bullmission lttandarrl (4f-i FR 1165 Jan 5
1981) Thus the concmiddotlusion rleri1ed above (Le refinery emissions
are secondary cause of exposure to the urb~n microopu~ation) is
further reinforced since H is irojected that re1eases from
components in benzene service (gt10 benzene) will be reduced by
73 by the proposed F~deral Standards T~~ ~istrict rules (eg
South Coast Air Quality Management District 456 and 4fi6l) are
not restricted to berz~ne per senor to only components in
benzene service The rjLs 1lso i-lclude flanges in addition to
the components ca i led uut in the propose nat i 01a l standard
The EPA estimated that if the proposed emission standard were adopted the
1~aximum annual benzene concertr0Vioi1 -~01middot 1 plant would be 36 ppb at a
distance of 0 1 kilomeL~~ avay Cor1paring this ground level concentration
with the general urban background i1 California of 19 ppb shows the latt2r t1
dominate In recognition of the secondary im0ortance of these thr~~ i~~s i~
was decided to utilize program resources to d~velop field data at other sites
1111here little emissions informati1)i1 1as cll-1aila1)le
8-4
APPENlJIX C
US tn v 1 ronrnen ta I Protection Agency
middotJc k
METHOD 108 - UETERMINATION OF PA~TICULATE ANlJ GASEOUS ARSENIC EMISSION
FRUM NONFE~RUUS SMELTERS
1 Applicaoil ity and Principle
11 Apmicroli1ability This rnetnod amicropliltS to tt1e determination of
i nor~an i c d rsen i c ( s) e1n is sions trom non f errou Sm(~ I ters dnd other sources as
smicroecified in the reuulations
12 Principle Particulate and gaseous As emissions are withdrawn
isokinetically from the source and collected on a glass mat filter and in
water The collected As is ther1 analyzed by means of atomic aosorption
spectrophotometry
2 Apparatus
21 Sa111pliny Train A schematic of tt1e sampling train is shown in
fiyure 421 it is similar to tt1e Method 5 train of 40 CRF 60 Appendix A
Tne sampliny trdin consists of tt1e followiny components
Note H1i s and a11 subse~uent nferences to ottler methods refer to the Methods in 40 CFflt 60 Appendix I
3 Annual Book of AST11 Stdndarctbull Part 31 Water Atmospheric
Analysis American Society for T~sting and 1-1aterials Philadelphia PA
1974 micromicro 40-4~
C-15
APPENDIX D
~f-~__d_y_ e s f o r Pr o c e s s i ng a n d An a l y s i s of PA H i n
Co~e Oven Emissions from Kaiser Steel
Global Geochemistry Corporation
10 Samples expected
l 1 Connective tubes between section welded to oven port cover and remainder of sampling system - These are to be stainless steel tubing disconnected and capped off for each sample They will be about 10-15 long
1 2 Impingers in series first two water-filled third dry - The center tubes will be removed and wrapped in clean foil and the impingers stoppered as is for shipping One set for each sample Protect from light during shipping
20 Extraction -procedure
Sample fractions are to be covered with foil or in opaque containers with inert gas flush during storage in freezer Solvents used in extraction will be BampJ glass distilled pesticide grade low residue) Working bench area will be screened with amber plastic and light bulbs
2 l Connective tubes will be rinsed with dichloromethane by partial filling capping tilting several times and draining into a beaker This will be repeated until the washings are colorless but at least three times
22 Impinger contents will be transferred to a separatory funnel The impingers and center tubes will be rinsed inlo the funnel using a measured amount of dichloroshymethane in portions The funnel is shaken well and the extract drawn The rinsing and shaking is repeated with two more portions of solvent The extracts and washshyings are combined over Na 2 S0 4 bull
A separate portion of the water used for impingers is extracted in the same way
For each liter of water 100 ml dichloromethane will be used per extraction ie 300 ml total
Internal standard is added to the combined exshytracts at this p0int
D-1
_rmiddot
30
40
Concentration of extracts
Bulk solvent is removed on a rotary evaporator in a water bath at not over 40deg stopp~ng short of dryness The res~dual volume should be just less than the f i n a l v o l ti 11 e o wl1 i ch the s o l uti on i s to be d i l u t e d (accurately) The object is a final solution with roughly five mgml solids concentration in dichloromethane After the vol w1 e ~ s made up to the mar k i n a vol umet r i c flask an aliquot is removed and evaporated on a tared disk in a nitrogen stream and weighed to determine yields The remaining solution is stored in a ial or ampoule sealed with a Teflon-lined septum A split for SAI QA may be made at this point (see Section 43 also)
Gravity column c1eanup chromatogram
4 1 Into a 10 min 1 d column with co2rse sinter and Teflon stopcoc~ 6 g 120 mesh silica gel (activated at 120deg) 1s sluiry packed in pentane cooling the column by evaporation from a tissue wrapped around it and wet with acetone) This minimizes vapor gaps in the column A cap of 3g sodium sulshyfate is added to the top The column must not dry out before the chromatogram is complete
42 A sample of the extract concentrate is taken This is to be based on experience of expected PAH conshytents but initially would be 5-10 mg This is exchanged with pentane by adding 10 ml portions of pentane and evaporating over about 100 mg silica to small volume Three to five portions of pentane are used The final residue is rinsed onto the column with pentane
4 3 The chromatogram is developed with four solvent steps
l 25 ml pentane 2 l 0 ml 2 0 ~~ dichloromethane in pentane
II II ii II3 l 0 ml 50 u II II4 l 0 ml 5 0 ~~ methanol
The column volume is approximately 8 cc This is allowed for while collecting fractions During the first chromatogram the graduated cylinder collector is observed for visible solvent changes and a more precise estimate of the column volume
T h e f o u r f r a c t i o n ~ a r e e v a p o r a t e d i n a n 1 t r o g e n stream at not oveimiddot 40 to l 0 ml then stored under nitrogen in the freezer The principal constituents
0-2
of the fractions will be
l Aliphatics 2 Lower molecular weight PAH
11 113 Higher 11
4 Polar materials
Fractions 2 and 3 are to be analyzed by HPLC the others retained for future interest or in case of an incomplete or defective separation Splits for SAI QA may also be made from these fractions
50 HPLC Analysis
5 l The column is a reverse phase partition column C18 bonded to microparticle silica 4 x 250 mm The solvent system is a gradient from 6040 acetonitrile water to 100 acetonitrile initial slope setting l 5min The flow rate is one mlmin the injecshytion sample volume one to ten microl depending on conshycentration Detectors are UV absorbance and fluorescence in series The absorbance is set routinely at 296 nm but may be varied in certain instances to aid in identification of peaks The fluorescence filtars are narrow pass excitation at 360 nm and cut-off emission above ~10 nm Both d e t e c to rs a re rec o rd e d s i 1i ul ta n e o u s l y
52 Reference calibration is based initially on indishyvidual chromatograms of single compounds and roushytinely interspersed chromatograms of multicomponent reference mixtures The proposed calibration stanshydards are the following at minimum
Phenathrene Pyrene Chrysene Perylene Benzo(a)pyrene Benzo (ghi)pcrylene I n d e n o ( l 2 ~ c d ) p y r e n e Coronene
The internal standards (added in Section 22) are 910-Dimethylanthracene
9-Phenylanthracene 9 10-Diphenylanthracene
60 Peak Identification
Significant peaks not recognizable as present in the reference standard will be investigated by alternate methods for identification
D-3
6 bull I Stopped-flow examination of UV absorbance for maxima
by manual wavelength varmiddotiation allows checking against known reference spectra (published or measured) This is usually sufficient confirm a component already suspected to be present
62 Collection of the fraction contairino the unknown peak allows later determination o~ t~e full absorbance andor the fluorescence spectrum for comparison with reference compounds
63 If not already run on GCMS the sample can be sent to SAI for that purpose possibly allowing the comshyponent to be identified Since the HPLC elution order is not the same as the GC retention sequence it may be necessary in affibiguous cases to collect the HPLC fraction and run it separately by GCMS
The~e may be numerous minor constituents not identified The decision on how many of these to track down and identify will be made in consultation with CARB adv~sors a~d with SAI since at some point the returns ute not worth further expense
_64 IdeAtified peaks are quantified by comparing their areas to the areas of the known concentrations of reference compounds A recovery factor based on the added internal standard nearest in elution order to the sample component is then applied the amount of internal standard originally added to the sample is divided by the amount found in the HPLC analysis This ratio is the multiplier used to correct for losses in concentration and chromatography steps
70 Quality Assurance
71 Transmittal of sample ~plits to SAI provides for external QA This should be done with 15 (or one in seven) of all samples
72 Blank water extracts are carried through the same analytical procedure (10 of total)
73 The cleanup and HPLC analysis are duplicated on 15 (or one in seven) of all extracts
74 The multicomponent reference is run first every day then elution volumes and peak area for each comshyponent are plotted on a control chdrt to pinpoint developing problems At least evetmiddoty tenth chromato-gram will be this standard It wi 11 also be rerun after any service on the instrument~ such as a column change
D-4
APPENDIX E
Gas chromatograph 1mass spcmiddotctromlt~try protocols used by SAI Trace
Organic Chemistry Laborator1
Aliquots of sampl=s received from Global Geochemistry Corporation
were analyzed by combined c11s chromatography mass spectrometry (GCMS)
Liquid extracts were spiked with an internal standard compound (deuteriumshy
labeled phenanthrene) priot to analysis by gas chromatography mass spectroshy
metrymiddot A Finnigan model 4( 21 Quadrupole GCMS system including an Incas data
system was used for all analysis Liquid samples (1 microliter) were injected
into a 30-miter x 025-mm ID SE-54 fused silica open tubular 9as chromatoshy
graphy column (JampW Scientific) A programmed GC oven rate of a 4 minute hold
at 30degc followed by a 40degCmin temerature increase to 160degc and a s0 cmin
temperature increase to 270degc was used for sample analysis The detector
system was a quadrupole mass spectrometer operated in the electron ionization
mode at 70eV The quadrupole mass filter was scanned from 35 to 475 atomic
mass units in 095 seconds A hold time of 005 seconds was used to stabilize
the electronics prior tote next scan Data were continuously acquired using
a Finnigan Incas Data system which writes time intensity mass spectral data
to a magnetic disc The mass spectrometer was operated under computer control
during acquisition Figures 33-7 through 33-9 show the total ionization
chromatograms for each of the three samples that were analyzed
Following analyses the data files were searched for priority
pollutant polynuclear aromatic hydrocarbons (PNAs) In addition to the
quantitative data reduction used for priority pollutant PNAs a survey analysis
was performed on all significant GC peaks (ie signal to noise of greater
than 101) The survey analysis was based upon a computerized library search of
individual background subtracted peak maxima scans for qualifying GC peaks
The library used for qualitative identification AJas a combined NIHEPA
National Bureau of Standards library 1hich contains approximately 26000
entries
Tables 33-11 through 3-13 present the results of this survey analysis As expected a series of al~yl Slbstituted PNAs were observed in addition to heterocyc l i c compounds con ta i ni n~ nitrogen iind sulfur moieties Further
investigation of the mass spectre of the nitrogen substituted identifications
confirmed the presence of benzonitrile isoquinoline andor quinoline acridine
carbazole and alkyl substibted ~yridines Ion specific searches for N-nitroso
substituted amines were neg1tive
E-1
APPENDIX F
Detailed Fiber and Mass Concentration Data - Johns Manville
=-- -i l L - [- - =-- -= f- - =-- -l J middot - -=-
i-- L =-- - ~ ~ L ~ ~ =--shy r---- --
- 7 l -=----7- -- - - - -- =-- - - ~ =-- - 1- -~- l
g~~~~ r_-middot~ ~-i [-- L ~ L Ldeg
middot--~ l - --------=-- r- L r- r-
~~~ =-- -- r- = - -r---
a- - - - - _ - -I~=-- L -- l 7 -~--
gtZlt _
- _(_ -lt lt =- = ZJZ~
-
__ _
~ -- __ middot- -~ ~~~ = L ~f~ Z ~ C
middot -
__ ~ lt lt =- _~~ijE
gt = z lt gt
rbullshy__
F-2
SI I HMllLfc J H 1 I OBr I lll-H ANI) NS~ coHcrrrllliTJ ON CUfllllTI VImiddot SIJMMIIY
TIITIL JnI~ SfiNtlEO TOTtl Ill- or FI ITETI TOTI Illlt ~111-11 11 ltllOT FIIACT I Ori 1111 OF ~middotWI MUllllNJbull lOTI 01IIMImiddot 01- JII TOTI FI lilmiddott~ UgtIJH nD
= t nwo 00 so MI CllONH = t I 1JOO SO (M = II 900 SO CM t bull (0000 l1T~
I I JO SO CM 70000 CUil JC Ml-TEHS
= ~lt O V l lWfl~
Manville D-2 Baghouse
723 347 -633
CIIHYSOTI LE JfllII I nou J11BICUOUS NO PTTFllN NON-SBI-STOS tLL FIIIEIIS
FI 111-11 COllflT ltFIIWll-
ITllClmiddotrir r lllrf
2gt()
) ( I f S bull1 bullbull
I 0
l ~-Irbullmiddot
00
fl JI)(~
00
0 OUU~~
00
0 OflO
~60
1110 000~~
FI IWII ( ()[IClmiddotNTHT I ON ltI- I mn ll11 ~41 GI OF FI LTEH) (FIBlmiddotn~ ll-11 CIJB f-11bullTl-H OF IH)
I (2fi71bull+0G (aG(l-+05
( gt027 F+04 OfmiddotHH
OOOOOE+OO OOOOOEtOO
OOOOOF+00 o uuoul~middotH)O
0 O(OOE+OO o 0000I+oo
J bull f 1)07 E+O( fi r71HE+OG
I
w
TcfLI ~~lllllmiddotCE 1lllmiddot lt~O Ctl lFII ~~O UI OF FIiTEi) t q UI 11-11 Cll II fllmiddotTlbullll OF J I H)
I 590)1+07 6 t 90llbull-HH
( HMOE+O 2 40411bull-1 o
0 OIHHlF+OO o 00001+oo
0 OUO(I 1 00 OOOOOJi+00
() 10110~ Ji)
OOOOOE+OO 1 lHHllJ i 00
0 00001middotgt1 00
M ~~~ middotor-ic Irrrn TI or
( PGRAMS 1bull111 ~c UI OF F J LTEH) (PGRAMS 1111 UJI illI Lil ()j Ill)
IUitTII Ml-Ml ltMI tJIOfl~~) ~Tlgt lgtEV
Ml- H Lot c1middoton nr COlmiddotF I II
I) I MWTEll Ml-N ltMI c11or1~~) STI) tHV
MEtl IOC cImiddotor1 r1N CCWF V11
AIt lbullI rt 1bulli I~ (Sil MIC) STIgt 1gt1-V
Ml- N 101 c1-ori r1u COlbulltmiddot V1ll
I 0 I ~if1+07 bull tJf I (11bull-HH
r 1gta 1 1) bull 2 (10()
0 lt700 l 1Hi4~ 2G44
() ~( 1 027B
- l bull 6 I 1bull1middot 0 J l)J 09J~fgt
1a 1071 1~ 09rn
0 2(jlt)fi l ~HVil 4040fi
~ HltF+O~ U (H~61bull+o I
o rooo 0()()00
-0GJOB 0 (000 0 (000
OOfiOO 00000
- )()57
OOGOO 00000
0 0 1)02 00000
- I J 0 0 0 1W 00000
00000 0 ()000 00000 00000 00000
00000 00000 00000 0 (WHO 00000
0()000 0()()()()
0 ()()()) 00000 00000
00000 00000 0000() 00000 00000
00000 00000 00000 00000 00000
00000 00000 00000 00000 00000
00000 00000 00000 00000 00000
00000 00000 00000 00000 00000
00000 00000 0 (WOO 00000 ()0000
( ~1701 l B bullJl~H 0 lt~( I BIJI ~ ~i Ill
o~-~ 0 ~j
- I ult-fc 0 I Hbull)I o bull)abullgt~
12 (1 41 17bullgt7
0 1 l I H) 4 ltmiddotI 1
VOUJME (ClJB MIC)
MFiN ~Tl) 111-V ru-1N 10C Ct-OM MN COE- Vtll
2lHH7 gt HUB
-2 7m 0 060 1)
( IJl)()C)
0 001 00000
-ltj 74a 1gt 0001~ 00000
00000 0()000 00000 00000 00000
00000 00000 00000 00000 00000
ooono 0 1)000 00000 00000 0 ltWOO
2 7 0 1J lJ I J J
-2J+H () ()~(
B 77gtB
lt ~
lt
C
J c
~ lf
_
ii
(]j (fJ
J I 0
a1 c r--- l)-- Clltr (Y) r- ro C)(V) gt deg CN C1 (0 I ~ C M
N --- r--
w J
s rJ gt-
c
~
LJ
t-
~ - -lt-[-bull
0- ~ l L L l~ -18~~
L a- - ~ I-
-I - ---
-
-- -- -- - - ~l
-1_ - - --
--~--- [-middot-~--- t-- -- [- -_ 3J ~I r-
z g ---
l
t--[-
l l c +-shyc
~ Ci - - I
-1--r-middot~- ~
L l- L ~
L
II V
C
c
t--
-= -- - - ---- - _
middot= - middot-= l J l~ - ~ ~ L - ~ L ~ [--
II II II II II II II
_
---
--
z
- I
==== lt _ --middot --~ 7 _
- ~~- _
z ~ ~ =lt ~~~ti3
f
gt =2 lt_
2 z _ lt=lt i~iS
-- z ~ lt
~~~t3
~lt~ =--~~~sect~ _ -- -middotImiddot_
L -_ -
F-4
(lJ II) I
~ Or--M
ltlJCsr-M ~ O) bullbullbullbull ~ lCMIO bull- ci gt (= N lC I M ~ 0 N
_
J J
---r--co - - ~l r-j = ~i
~~ 2-
88 rJ--_-ltI -
lt ~lt
z
Cl J ~ _ e-l ~~
- ~ middot~ e-- l ~ l~ [- - ~ -~=----~
_ L
L
--o---
-o-~shyc L~ ~ - 0 - Ll-C d o~-
11middot II 11 11 II II II
-_ 0- _
--
-- lt== ~~ ltZ z-== -~~=E~8C~l~-=rI - lt ~ fJ
ltZ lt---=lt===~~~- _ ltltlt=gt ~~=~~~~~ -
~ ~ lt -=z~ z _ -shy middot--z =-
-middotmiddotmiddotmiddot middot-middot -----
r bull bull _
- _ - ~-middot __ _
~ _
middot- - --- middotmiddot
_
- l
~=~r -~= ==
lt =~~~ti
F-5
FIHJ-ll Nil MAS~ CONCFNTllJTlOi SJ I SMllLE I l) Al OJ CUMJIATI JIbull t-llHMHY
TOTL iHf- SCNNFD lOTiL Illmiddot OF FILTFTI TOJI Ill- JSlllmiddot~ll I I OIIOT FlltCT [ ON 1111middot OImiddot ~1-M iIHlllllJNE TO 1 1 I ()lIJMImiddot ()I t I H TOTI V 1111-11~ COUNTED
= 1080000 SQ MI~RlozNS = 11 900 SQ CM = 1 l bull )00 SO CM = 1 00000 liHlS
11 IH) so CM 2G70000 CUBIC METfiHS
1 ) fi FI JllIIS
Man vi l1 e D-2 Baghouse
723 0 347 -633
Cl Ill Y-OJ l LE MIPII I BOLF Jf1BI(UOUS NO PATTEnN NON-ARHFSTOS ALL FI BFllf~
F 11l-ll f()llrlr ( I 1IFH-- l 17 G o 00 00 00 19fi
Tl fL P1 Sl~JNNFD Tl lI ltl- OF 1bullrn~n H 11I HF ~IJlbullI) lll 1 ll(IT 1-ILCTON IL CW ~U-1 HErlllllNE
THl1 (11llrJI~ OF 1IH Tt1T-d 11111middotH COIJrfflI)
= 697~oo so MICRONS l l 1)00 so CM t I JOO t-o CM
= 1 00(00 lHT~ = I l 1)0 ~o GI
ant()Oo cun1r NETEns = 7 0 I l lWllS
Manville (upwind)
CIIIIYsOllY AMPII I nou JMfl[CUOUS NO PA1vrrHN NON-A~~HFSTOS ALL FIBFHS
v1mI1 Cttllffl l 1-- I II I fl~ 1 () () oo l 0 00 40 70
1middot1111unmiddot ToT1L f J BFllS o ooo~ ooow- t OTT~ ( 01rn 57 I 1~m too ooo~
1middot 1111--11 1or111rnnT I ON I Imiddot I 111 II~~ 1111 ~o CM OF f I LTFTl) 1 11111 1middot111 CUil middot11bullnn Illmiddot J l ll)
O OOOlJEmiddotH)O UUUUOL+OO
0 00001-1-00 0 00011 lbull-Imiddot00
G 1 I nn+OG 1 HJ llmiddot~+OG
00001+oo 0001101-1-00
r n~+H+uG l 7llHmiddotlE+Ofl
l bull l 1)1-n+O(i I J ~ 1)(11-middotl()i
T11T I ~Ull FiCImiddot 1 ll f ( ~ l lrl ll11 tn Ii OF F 1ITFll)
f ill I I I i 11 I U l ( i I Ill J
0 OOl)Ol ll() () U()0 I Imiddot()()
0 UOOOlmiddotgt1-00 O OOlgtUlbullgt100
l 0 1) ) I 1 lJ J ~ tl( 1H-HH
0 UUUOlmiddotH)I) o 00001-+oo
~ lt1Hlt1(E+OG 7(H(Ml-middotHM
0 OOOOJbullH)O 0 0000lbullgtHlO
77 I
N ~
rl gt- 10NCJfffll1T I ON
r orR AMS ITll ~q lfI (II- FIT n~ll) - LIi LIIll 111middotTU OI 1lll)(PGRAMS II i1 Ill MlmiddotN
bull I t bull I 1 f I ~ middotr1 Ill lmiddotl 1- [~ IOC ( Hl1I f r~ Ull-F ll
0 0000 Imiddot bull-110 0 0000 Ibull HO
0 ltHgtOO 000110 t) (HJ()()
0()()()()
0()000
llOOOOF+Oo 0 00001bull-100
0 ()000 OOtlOO 00000 0 (WOO 00000
0 (((7
0 ~0-i -0 middotlbullribull)bulll-
0 ltl l lt 0 oHi
00000 00000 00000 0(0HO 0000()
077[0 0 -1117
-0 ~bull)I)()
0 7middot1middot1 (i () L~bull)7
0 7~B( 0 I I
-o alt17n 0 (1)~1 0 17lt1 I
11 I 11ITlt I j i 1l 1 111bull I
f 11middotgt N gt I ii ii I V 111 ii IOC tHlfl flf1 Clll-tmiddotmiddot VAIi
00000 0()000 0000() 00000 O 0000
041000 ()0000 () 0000 o 0(1()0
0000()
0 llfl(i7 OltHJ
-~ lt-11 I 0 07 I 1 () I)lt)
0 (H)ll()
00000 00000 0 ()()()() 0 0(100
0 0 Omiddotlmiddot l
-c07 1)
0 I 1 I( 0 1101
0 l l j--
OOGli -~bullI) ( (J
0 1lt11 I 077G7
I 1 (71 rJI C)
tll N -r11 111-V rmiddotllfl trn CI llf I 111bull1 COIT rll
() ())(l(J
00000 0 ()(1()()
0000() () (1()(10
00000 00000 00000 0 (I()(()
00000
0 ~ I Ifdeggt 0 I Imiddot~ I
- I H 1)7
O I GOO I I i ) l)
0 0001) 00000 00000 00000 00000
0 1 1)17 o~ViH
-1 0 1)gtG 01lbullH 0 7 I till
0 l l lt G 0 ~r1
- J bull J1 1) I 0 17 I bull at-lt
lllllrmiddot lt Clll ii IC)
~I Imiddotmiddot N ~middot1 D 1)1bullI f-1 Imiddotgt f IOC ClltHI rm uw-- i ll
O(lOOO 00000 00000 () ()(J()(J
(l(t(()()
0()()00 00000 00000 (1(1(1(()
00000
000fi7 OOOGO
-G Jll7l 0011G 1 H20
00000 00000 00000 000()() 00000
0 0 I bulllCi 001w
-4 iHd 00101 I ~700
OOIOH 0 0 I Ilt
-(1 j l)~G
000(i 1tHH
--
--
L c~
U c t c
(
J
J -
C
C
c g 0 0
+ + e--c~ =~ - r~ C- l~
--
+ +---
- -C 0 o- -c--1
I
lt t w
lt
C
ta J ~ ltCi
lt ~
a---~ -- c gt C 0 rtl 5 -
VJ C f-rJJ c
[J_
ltI ~ C z
z _ ~
~ lte 0 ~
C
tJ
e
C
~ c C
L N
~~ C 0
C
l~ 1 0
t CI ~ ----~ -
i-
cc cc + + t=~= g~ --
-~ C + + t~ ---[w (~ lC
-
ccc ~t c5 ~ ee 00000
co--oc I
- - - - -_____
------ _ - _ _--__ ---__ ---- _ - _
ooocc
rr ~ c-~ E lt
C
-
~ C g L-
gt o + + ~~ ( L C- lN o -l
t l - -I
~ - L~ [ti e- [
l _ 1
- - ~_ C
gg -- -- ~ -0000-tr--- ____
----C __
- __ - _ --ggggg
- -
C
+ + -- --- -~_shy-----_ _ -----___ ooo~
-----__ ___ _ - - - _ - _ _- - - - ___ 0~000
--__ ---_ __ -- --c5 ooco
C cc
1 II II 11 11 II II
z
-
r ~ c
z
~ t lt i C z __ -- -~tt ~ middot~ ~
-middot middotmiddotmiddot _ - -----=___
_ =~ - _ -=shy ~
-- -~== 2~= ~ () V)
_middot ~ c( c(
0 er ~- ~ ~ lt 0 0___
-
- amp lt~ ~~~t8
F-21
) ) ) )
TOT1L AllE SCiNrmn TIIT1 L ll lbull1 tW FI 1lFll TIIL1 11bulli 1~lllbullll lllIIOT lIIJClION -11- OV TUI Mlbullrllll1Nft TOT1 I VOIIHIIbull 01- I Ill TOTmiddot I Imiddot I I 11-11~~ COi l NTlbulln
_LI__l_ ___l 4 bull P 7 bull P A R l n A 4 R 1 n 4 4 ___ 7 A ___ _RA____ 1 A 1A 8 1 SL__ 1 R 5 1 ~ 3 _1] ________ 111~____JJ__) 1 ~---~-~ 4 9 1 ----1___11
A0 1A4 JR 17 lAR A 41 4C 174 A ll~ JnR ll7 9 5 R12
----- --------- - ---- ---- -- middot-middot --- ---- -- ----middot-- ----middot ----~ 7 c ~ bull n 1 1 2 l 9 -s 1 -i s s ~ c c__ bull A____02 1 1 2 1 r n 1 c 0 r-- l ~ 1 o 1 0 1 t 1 o c- ----~--- -Pl Ol OP lA2 11A 61 c 0 ~ cq03 l 5 4 l 4 4 1 i r 7 -i r) c q 1 deg 1n
qA RO 107 ~_12_7 SA c 7 A04 _l _ s_ _________l 7 l RA____ 7 ______l _o 4 __ l nn_________ 0 _7________ R_l __ o ~ _____ J(__7 11__r ___ _t-__ o ____ ln ___ h_ ______ An____ c-_n __05 ]lA 140 170 7cA __2n1 qc- 7q 1n1
c 2c 7 lR 1 O 406 11 c A 0 4 n 3 4 7 l 4 R ~ 7 7 5
___ ___~ c Z bull A ~R_ ---2_h 2 bull bull 4 e R bull C07 l_ H l_ ri ------- R1 --- c ----------- AQ _middotbull 4_p____ r 7 ___1[_2_ _r--_______ _ A____________ )_3-____ _____ __t 9 ______ ~_____ _3 __ --~___()_______ __I__08 C7 lR9 9A 1n~ ll7 S AS 84 ~Q 33 11 11 1n R 3A 3309 7RJ 1R lA 14-1 17fI 77 4R 7(_
---~- L~ 1 7 -~ 1 7 4 1 l 1 9 S 3 9 C) 4 9 A ~ 9 q10 __ 7 c _ ___ _ 2 RA l A ____J__6_R 1 C 4 7 P C 1 rq I
-----~~a-R___-=4bull2_1 e __ middot-middot middot bulls __ - __ IL 9 1 ~_1_ ___3 7 40__19 1 ~~- gt q6_ (- l Q 144 43 _37 _ A f ---middot ~ bull r bull q ~ 9 IL ~ 1 _0 -~ bull 1 A 3420 l r -- I- I q4 l qo l c c 0
sectJ~ BLVDII EJ SEi~D~ 1 nu tmiddot-middotmiddot-middot J _
tl~ r-middotn ----i J middot--1_ ___ ~1-1
A f gtiff_ ffmiddotc YYt)II ~ r bull Ja------ ijlttQ~QW~
___ ~Q -~imiddotbull1 ~H fl(--middot
-lt) bullbull
1 middot
middot--middot middot-- _
middotmiddot_ u - -_-~~middot_1r
~-- ~ _
~ -middotmiddot bullY
j ~ ~ L ~ ll c_l
~ pf-
111-r ~ (i middot) ~
bI middot_FT )-I Imiddot ~middotmiddot () --
I filfltf) J ~3 ilj-t ~ C _G~ llli_A T1 AN B~ACI
I~
603 middot Alondro Pork
l U (
Goll Coorll
~ htyen 1~J-(_
d n t U Wik c rr 1 i t_y of Indus try
(
H-7
Plant Site - Dow ch em i cal US A Pittsburg
I bullbull middot1 Sfl Wi - bull AUIIUlllJLII+
BELMOr r l -- -- 7iSHOfgt ITHACA lN I
N[VAOA lN TGERS llltI
Q LN
GfORC~
72copy~~~~-0AYTON l
I I+
__ middot
1- - -~ middot
I I
- ----------
Plrint lite - ilu rnnt Antioch
) j
(
H-9
Plant Site - Johns rlanville ctockton
a a
LEGEND
CENSUS T1ACT1
51 )I 51 07
1980 CENSUS TRACTS SAN JOAQUIN COUNTY
37
100
SOUfCE TESTS - STAUFFER CHEMICAL COMPANY
101 SITE OVtRVItw
Summary
Stauffer is located in the Carson area of Los Angeles County It is
a manufacturer of polyvinyl chloride from tl1e monomer (VCM) ll11hich is produced
on sit2 Stauffer is the only California producer of ethylene dichloride
(EUC) which is converted to VCM The VCM is regulated by several standards
including Cal OSHA EPA (emission standard) and CARB (ambient)
During Phase I the plant was inspected and possible emission
sources were identified Th~se cans~sted of EDC storage ta11ks fugitive
emissions from valves flanges and pumps process vat2r content gas
incinerator effluents and th~ loading of outbound tankers
Since the completion of Phase I several events have transpired which
a ffectebullj the test program to deterni ne foe i 1i ty emissions These wer2
~ the completion of a vent ~ldS incineration system tied to all significant EDC storage tdnk breathers
$ the completion of a comprehensive study by SAi staff to develop a nationwide material balance for EOC
bull inactivity in EOC importation for the plant and elimination of EDC exportation from the plant
Based 11pon this input and the plant inspection of February 3 1981
it was expected that a plant emission factor for EDC can be determined with
relatively little uncertainty It is further expected that si 1Jnificant
atmospheric release of EOC du~ to plant operation may not occur at the plant
itself but rather offsite These vmuld drise fro11 two sources - the process
water discharge from the plant and severcil off site EDC storaye tanks
Facilities Description
Ethylene dichloride is being produced at Stauffer by two processes
direct chlorination and oxychlorindtion In the former EDC is produced by
d i rec t ch I o r i nd t i on uf et 11y l t~ n e i n the pr 1~senc e of cHl Fe C 1 J crl t d lys t I n th e
latter process EUC is microroduced by the oxychl)rindtion of ethylene with
hydroyen chloride and oxyuen in tne prescrlCe of a cdtdlyst typically Cuc1 bull2
132
lhe microldrrt fluriti1~s tuC after tne microrimar y reactions by separation and digt ti l I dt i ur1 fgtr-ocJuLL 1 s ctured ctncl umiddot 2d ctS foed for VCM production All
sourclS agre~ tl1dt c1ission of EDC is (f concern in its productionstorage
process stages ctnJ O little importance in VCM production steps (J~B 1980)
At the timt of plant inspection some storage of EOC existed at three
ledsed storage tanks located at tne Port of Los lngeles 1-iaterial has not
been 1-Jithdravm from these tanks during the 1dst year and there had been
discussion to consolidate materictl into a single tank
All microrocess components handling EDC storage are now tied to a closed
ventilation incineration syste11 It is expected that virtually all
chlorinated hydrocarbons includiny EDC will be etfectively destroyed since the
system must demonstrdte VCM concentrations are reduced below 1 ppm Firebox
tempera tu re is aoodegF and residence ti me greater than 1 5 seconds under
heaviest flow conditions Note that by way of comparison combustion perforshy
mance datct for PC8s are 99995 destruction at 1832degF and 1 second residence
time and 9Y9Y9YY4 at 2 seconds dnd 2372degF (N Flynn SAi Personal
Communication)
Fugitive emissions from valves and flanges are monitored by the
plant according to the requirements of SCAQMD Rule 4661 Pumps and
compressor follow ~ule 466 Emission and control requirements for vinyl
chloride are specified by ~ules 1005 and 10051 EOC concentrations in
wastewater are monitored several times daily in tne primary EDC steam stripper
stream and once daily in the composite plant outflow stream Daily water
disct1arge limits are 2~ ppm wit11 typical montt1ly averages being in the 8 ppm
vicinity The discharge limit is ~mbodied in the discharge permit (number
5061) vJith tt1e Los Angeles County ~anitation District
Tnere are a number of points at which tt1e rnc can be released to the
dtmospnere and they are similar for both dir~ct chlorination and oxychlorinashy
tion These include the following along with their emissions factors for direct chlorination
A chlorinator vent 2xliJ- 5 IOdSS per unit mass EDC produced
I) I ight end column vent -52x lt)
C dist i I I at i un column vent lXlU-5
u storage tank breatning 7xlu- 5
133
E storaye tankworkiny loss 27x 0-u 5F fugitiv~ emission bxll -
-JLi ildSteJdter l gtX] U
Em~ssion factors are si~~lJ fer oxychlorination processes
Enlission factor vau-5 (~1rbulle Tdilt2~ frcn tne literature (JR13 198U) and
assume 9b iciency for incineration (A-E) It is also assumed above that
EIJC emissions to water are Lf~ uf ~rihsion to citir ard that in wastewater
treatment 100 of EOC discharged tc1 vater is rel easl~d to a i 1middot ~ These emission
factors 1bull1e1middote used to prioritize r-2ieaies iJut were cldrly uude approximations
to tre planL l)ecause of thlt ncine-xt~on system it was anticipated that
fdctors A-E would be ri2dtKcdft 1-middotactor - 1nii~t1t be reduced since a monitoring
program had been in force al20st o~e y2ar Factor G and the off site storage
tanks~ vni~cn are not tied into lt1n ircinerator system rniyht -iominate emissions
Emissmiddotion limits and concomitant regulations embodied in Rule l00S of
the SC-l)MU have necessitctted tlw incinera-cion system Ninety gas chronatograph
probes are located throughout t11e plant including the stack of the primary
incinerator Con cent rations of vc1v1 are reported es sent id I I y be I ow tt1e r~yu la -
tory limit of 10 ppm and in fact below the limit of detection somewhat less
tl1an 01 micropra
lUl MEASUK~MENT APPROACH
Tne objective of tt1e proyram is to determine a pl ant emissions for
EOC It is not accemicrotabl~ to dtmiddotvelop sucll informatmiddot1on basect upon published
indlllstry Jide esti11dtes of plant control efficiencivs Fortunately it was
possible to desiyn a monitoring and calculational progra1n to determine EUC
emissions for tne site and not ctbsorb a disproportionate share of program
resources
Hie re are tnre~ modes ot re I ease from tt12 micro Iant dnd d fourt11
offsite
bull Post-Process 1nci11er-dtion
Processes A-E of ecc ion lll 1 dre a 11 1ented into the pl ant
incinerdtio11 syste11 rt1e conceritrcttion of VCfmiddotl contir1uously iectsured in till=
stctck cts SJdS output remicroresents d rectsondble UjJper li11it to diJply tor EDC
con cent rat i ons s i nce i ts ef f i c i ency of i nc i nera t i on 1s d t 1 east c l1 u a I to tr1 at
134
ot VCibulll Tlerefore knowledge of the system dirflow rnd VCM concentrcttion are
tt1e necessary and sufficient conditions for determining the bounds of EDC
reledse fhe pldnt expn~sse I wi 11 inyne~)s to provide tnese data in order to
support the calculations
bull Fugitive Emisions - Valves Flanges Seals and Other Sources
Fugitive emission from the valves pumps and flanges can be
determined more precisely tnan was anticipated since Stauffer has completed a
comprehensive leak inventory of all such components in compliance with Rule
466 466 1 and 1005 The Inventory delineates all leaks uncovered by their
three man crew throughout tne year and indicates the screening level in ppmv
and component identity In consultation with Stauffer we will be able to
identify the total number of components associated with EDC handling systems
( 700) the distribution ot substance composition streams the distribution
and incidence of leaks by hardware component type and leak rate We will
utilize this information to provide historical data to compliment our Foxboro
OVA sampling at the site dased upon this monitoring we predict an EDC mass
emission rate for the fugitive releases The plant has agreed to provide the
necessary information We propose to meet witt plant personnel prior to the
start of monitoring and finlize the sdmpliny strategy to accomplish 100
coverage of the lines of hiuhest EDC composition The sampling approach and
calculation of mass emission are described in Section 72
bull ~~as tewater
From examination of the emission factors derived from published
literature (see Section 10~) it is clear that EDC release from wastewater to
air is potentially several 11rders of magnitude higher than any other plant
source l1ased upon microl ant ml~a sured concentrcJ ti ons of 8 ppm EDC in water a more
realistic emission factor wgtuld b 24xlo-4 1-iassunit mass EOC produced
Clearly this could still be the dominant source Furthermore the release
point would be expected to middotgte locctted between the plant and the sanitary
district treatment site whi1 his~ kmfrom the plant at 24501 S Figueroa We
believe it is necessary to ndemicroendently confirm the average 24 hour EDC
concentration in the dischar-qe valer by obt1ininy the refrigerated composite
sample We t1ave identified ttie I 1nes of int~rest and received agrt~ement to
sample and analyze for EDC in the stream ~e propose to draw a duplicate
sample from the compositor nd ani1lyze for its rnc content This will be
135
compared with Stauffers para11eI tnalysis It is not reliable to obtain
direct readings of EDC by survey instrument in the air above the water flow since concentrations at crny single point will be i 1 the near dmbient range
Emissions will be cc1lcJLtecl 1_sing the plants vulumetric daily flm
and ass~m~ng that complete degass40~ of the effluent will eventually occur
This assumption is based umicroon the relative volatility uf EDC and tne distances
involved However it should be noted that no direct experimental or
monitoring data is available with which to confirn this Th2 composition of
the discharge stream is unknomiddotwn since it merges into a larqe multisource flow
Conversation with the LA County Sanitation l)istrict (J f1ilne) reveals the
stream to be both exposed and covered Vents exist where me measurements
could De made to assess gross leaka~e at key points
We will obtain samp1es of several plant process discharge streams in
40 ml bottles witn no head space Transit time fr0m sample collection to
analysis will be minimized and wi l I not exceed 24 nours This time frame is
conservat i v e al though ED C i s a v o l ct ti l e mater i al ar1 ct vJi 11 undergo
concentration degradation and outgdssing lnalysi) will b(~ performed in tt1e
SAI Trace Environmental Chemistry Laborcttory utilizing pur 1Je and trap analysis
FID gas chromatography A trial a11alysmiddotis was conducted and the EOC
characteristic peak was distinctive down to the ppb level Therefore the
analysis should easily confirm concentrations in t1e 8 ppm range
Offsite Storage ranks
Three EOC leased storage tanks are located offsite at the Port
of Los Angeles and are owned by annther firm Annual emissions from the tanks
were very roughly estimated based un AP-42 JRB 1980) as 44 kkgyear
based on preliminary estimates of stored quantites
Considering tne magnitude of this source these storage tanks must
be investigated We will gather information about their configuration and
control i n order to ca I cu l ate their e111 i s s i on s bull
136
lU3 UEfERM I NAmiddot 10~~ OF MISS IONS
1031 Incinerator
Vinyl chl 11ride co11centration is monitored dt approximately 90
locations throughoul the pl trlt (Lanyner 19Bl) including the incinerator
output Conc~ntrations are reported by Stauffer dS less than 0l ppm at a
flow rate between 10000 ant 12000 SCFM Although no direct monitoring of
EDC is conducted it is posible to conservatively bound the concentration of
EDC at U 1 pp111 Tl1at conce 1ltrat ion is a suitable ci10i ce s i nee it represents a
conservative bound on tile VCM levels measured near the stack output
Furthermore the molar volume will be taken as 224 L rather than the higher
value it would nave because of the sl i~thtly elevated temperature at the
detector location
Using 12000 SCFM one has thE annual emission of EDC as 12 x 103
SCFM x 283 LSCF x 526x105 minyr x lmole x 97gmole x 10-7vv 224L
Therefore incinerator emissions of EOG are thought to be bounded by 773 kg or
170 1byear
10J2 Fugitive Emissions
Seven hundred (7UlJ) sources were surveyed comprising nearly 100 of
EDC service All accessibl~ plant areas with streams containing greater than
1 EUC were screened except for a small number of devices located in areas
11lere active maintenance was being condJCted It is believed unnecessary to
p2rf-gtrm any emission factor adjustment since it is estimated that gredter than
95 of the requisite components were screened
Three leaks above an arbitrary OVA reading threshold of 20 ppm were
parameters and mass emission numbers fhe upper bound leak rate was
determined to be close to twice the nominal leak rate which accounted for the
response fltlctor of amicroproximamiddotely 05 by Radirn for TLV detection of EDC
( 13 rown 1980) bull
Stauffer found and reported approximately 10 leaks in the EDC
service during 9 months previous to the pLrnt testing Assuming no111inal leak
137
) )
Table 103-1
STAUFFER CHEMICAL MASS EMISSIONS PREDICTION FROM OVA SCREENING VALUES
OVA SAI Actual Upper Device Response Response Cone Source Leak Rate Bound Location (ppm) Rae tor (ppm) a b Se I Re Function lbyr lbyr
Gate Valve Unit P1559A
Valve Heavy Ends Column Unit 14439
Compressor Seal EDC Storage Unit Tl062
I-- w co
300 11 273 -035 096 0 17 153 D 6
11 I500 114 439 II 241 C 1
r III II II o-is lUI bull2000 vl 1 1818 A 9~
All emissions 100 EDC (12-Dichloroethane)
11
values in the nei~hborhood of 2000 lbs~ total could be emitted annually
Therefore it appears ttiat major reductions in the mass emissions were achieved
by the company run inspection prugram and the fugitive emission source has now
become of seconqary importance as a fraction of total plant emissions
lUJJ Wastewater Discharge
Wastewater samples were collected in duplicate from four sites
within the plant for determination of ethylene dichloride (EDC) concentration
The samples were collected in EPP standard 40 ml VOA vials on July 1 1981
Analyses were performed using standard purge and trap techniques coupled with flame ionization detection gas chromatography The results are given in the
table below
Sample description EDC conc~ntration range (microJml)
EDC concentration average ( 1-19ml)
Approximate fl ow ( gpm)
PVC Interceptor Box 82 - 108 95 200
EDC Stripper number Chlorination area C1404
2
011 - 014 012 25
Final collection site for pH adjustmentP663 349 - 3B2 366 350
Final discharge site sanitary sewer 6 - 254 158 500
The water in the PVC lnterc1~ptor ~ox was warm and represents washings from the
PVC reactor which travels t 1J the Interceptor Box in a concrete drainage ditch
Water from the EDC ctripper was vPry t10t and because of its heat was difficult to collect The ~1eat ot trie wat11 r may in pdrt account for tne
relatively low concentratio11 of ElC here as the EDC would out-gas from the hot
~vater more readily The fir1ctl collection site tor pH adjustment is a large
concrete container middotvith mixers in it located just prior to the final discharge
site Water at the final dischdrg~ site is being constantly aerated due to
the speed that it flows thru~yn the concrete drainage trough This aeration
could account for tne relatively large range in the rnc conotent found here
The average EDC concentraticmiddot1 at tne final discharge site is middotbelow the daily
discharye requirement of 25 19ml EDC
139
Plant oersonnel indicdted monthly iverage readings are typically on
the order of 8 ppm but daily averages can re 1ch greater than 20 microgml In
addition LA County Sanitation Uistrict staff (J Milne 1981) indicated that
an on-site impoundment pond can become laden with EOC during certain abnormal
operating periods and discharge Vdriances arl requested by Stauffer This
mi 9t1t Jccur on the order of O1c0 aCii ecH c1 1d t1erefore is no-c expected to
signifcantly impact averag~ cnnJJI discharg ~ values lt s not expected at
this t1ime tEit evaporative eriss uns from this pond are significant except
infrequently during upset conditions and spi 11 control operations Program
staff ~~ 1er2 una111Jare of any periods ~h 1~n EDC cimtent in the ponds could be
appreciable and therefore no sampling was performed
Utilizing the average uf the two final discharge concentration
1readinJs (158 micrograrnsm~ 1 and 1G 1Ji gpm fl)W one has 34600 lbyear of EDC
released into the sanitary sEwer For the pJrposes of calculating population
exposures from plant releases it will be assJmed that the EDC is locally
emitted from the wastewater streams Tl1ere ire no monitoring data available
with which to develop a more accurate release profile
However emissions of me fron the plant wastewater discharge can be
calculated according to the method of Mcmiddotckay (197S) as modified by Dilling
(1977) It should be noted that this mLst be considered an estimate since
conditions of fl ow and the presence of other substances will influence
emission rates
Using Uilling (1977) for non~erated flow first recalculate
Henry 1 s law constant (dimensionless-my of chmiddot~mical per liter of air divided by
mg of chemical per liter of water) dS
1604 P M 1 Wl
TS l
)
wnere
H Henrys law constant dimensionlessl
Pi tile compounds pur1~ componen- vapor pressure in mm Hg at T
Mwi the 11olecular weiglt
T tile absol ~te temperature of he wastewater in K
Si tt1e compounds ~uluhility in myliter at T
Then for UJC
(l6u4)(7~)(9Y)H = = U0447 with 1
(2lJ4)(8690)
LIii bullul1Jl)i I 1 1y 1 middot UL i11 wt1Ler yiven -1L L0degC is 8u4JU UHI tile vapl)r pressure
1 4 psi ( 7 2mm) at 7 0 degF bull
Tl1e overall liquid mass-transfer coefficient Kil is given by Dilling (1975)
as
(2211)(06) in mhr
-f-0-4~ (M _) 12+ 100 Wl-H-
1
=U108111tH
Then from Mackay the percent desorption is given by
c 1 = exp (-Kil tL) where
c 0
c = the concentrd ti on at time t of EOC 1
co = tile initial concentration
L = the Ii quid depth (Ill)
t = tt1e retention time (in hr) of the liquid in the wastewater
system
Retention time is bc1s~d 11pon a flow velocity range of 3 ftsec as estimated by
the Los Angeles County Sdnitatii)middot1 Dibulltrict (1J Milne) over a middotdistance of
ctmicroprox i 1nate I y S km to the Joi n_t Wdter Po 11 ut ion Contra l Pl ant Sewer fl ow
depth throughout the entire route have been estimated by J Milne as typically
1 foot 030m) to the Davison Pump Plant and 3 feet (09m) to the Joint
Tredtment Plant distances assumed to bt~ 1 mile and 2 miles respectively
Trdnsit tiine l)eco111es between 05 tnd LU hours sequentially Then computing
the net emission reduction as the product of each leg
Therefore 26 of the EOC i erni t ted bdween the pl ant and the Joint Water
Pollution Control Plant Rtgtsfdence ti111e and conditions at the plant account
141
for additional releases and residudl content is forther emitted between the
plant cnd the ocean discharge poiiit ~-Herever the flow is r2tained aerated
or shallow emissions will be accentuated
Offsite Storage Tank Emissions
There are three storage tanks leased by Stauffer for EDC storage
1ocated at 22nd and Gaffey Sts ~ bull San Pedro These tan ks a ~e used for long
term storage rather than providing feed on a routine basis Therefore
breathing loss rather than v1orkir 1J lass is )f concern An tanks are white
two being 67 ft in diameter while t~2 third is 57 ft All are 40 ft 3 inches
high and have capacities of e~thi l0~10000 (2) or 840000 gallons The
tanks are cone roof type and are not tied into vapor recovery systems The
South Coast Air Quality r1dna~J~ment u~ str4 ct has based their estimate of
emissions on the specification of 12 foot vapor level Using the AP-42
formula for breathing loss with the vapor pressure of EDC at 20degc and an
average diurnal temperature variation of 26degF one hds for each of the two
larger tanks
l) 173 H bull51 T bull 5 LlI = (619 x 10-S) M( _P_ tH F CKcp 147-P
68 173 51 5 = (619 X 10-S) (9119~ 116 bull (67) ( I 5) (26) (1) (1)
14 7-1 16)
Thus for the two larger tanks LB= 472 lbday
For the third tank 0=57 and = 178 lbdayL8 The total emissions become 23724 lbyear ~ince he plant is in the process
of shutdown it is not clear what the liquid levels currently are Note that
if the material in the three tanks were combined into one totamiddot1 -~1nissions
would be reduced markedly Exposure to a population from this site was
determined separately since it is lbullgtcated over six miles fron tl1e plant
142
110
ASSESSMENT OF PUPULATlUN EXPUSUR~
111 OVEKVIEW
A major program goal wa~ to compare emissions from the various
sources and identify and rank any 11 tlot spots in California where the general
population was exposed to elevated concentrations of carcinogens A simple
Gaussian dispersion model was therefore used to obtain order-of-magnitude estimates of exposure of the genermiddotal population surrounding each source
Since this was essentially a screening study use of more sophisticated models
was not appropriate
It cannot be emphcsized too stronyly tnat most California residents
are exposed to emissions of hazardous substances from a variety of natural and
rnanmade sources Urban dwellers dre typically exposed to greater concentrashy
tions than rural residents however all are subjected to so called 11 background 11 levels from multiple sources In order to place stationary
source exposures in perspective the typical ambient levels of each substance
were identified from the literature and compdred with the concentrations due
to the emissions from each ~lant Exposures were thus expressed both as
absolute quantities and as increments above background
Comparison of plants presents a further difficulty in that various
substances are being consict~red No attempt was made to evaluate the relative
importance of exposure to two different substances such as chloroform versus carbon tetrachloride other than by ambient concentration
112 DATA SOURCES
1121 Meteorologicdl Llata
The dispersion model to be descritgted in Section 113 required input
of annual average wind speed and frequency of occurrence of wind from each
compass direction These data were obtained for most of the sites from the
South Codst Air wuality Management District (SCAQMD) In other cases (Kaiser
Jotms-Manvi l le l)ow and OuPlt)nt) only clai ly cJVerage wind speed and frequency
data were available In al I cases we used data from the meteorological
station nearest the modeled e11issrnn source Table 112-1 summarizes the
141
Table 112-1 METEOROLOGICAL DATA BASE
Observation Site Data Period of Pldnt (Distdnce Source Observation Remarks
from Plant)
A11 i ed Che11 i Ca 1 Lennox SCA~MD 1Y71-1974 Averaged hourly readings available ~ 1 Segundo (3 111iles)
S ff~r-- Uemical Long l3each SCA()MD 1962-1974 Averaged hourmiddoty leadir19s avai1able offsite Carson (3 miles) storage tanks appruxi1ntL~ly 5 miies from
Figure 114-3 Cumulative Population Exposure to Carcinogenic Polycyclic Aromatic Hydrocarbons From Coke Ovens at Kaiser Steel Corporation Steel Mill Fontana
CV)
0 -I X-V) V) a _J
s 0
C 0 micro co s micro C a u
l ~
14
1 --
J- t_bulli
_ C
uw m 0
1 -0 a bullbull micro
microdeg ()
0 micro tbullmiddot C QJ V)
0
X LLJ
C
middot4middot~middot C 0 micro rt1
-- - 0 ~ 0
0
n __ _______bull~ -______ --al_ - ~_
I bull
I I - bull
middot 1
-~1
l i l ~l
~ 0 I 1
Li bull I t_~I
i bull -1 _1-J
0fmiddotbullmiddot-middot+-middott-t-r+-t-+-+middotr+-t--+-1--t+ t f-t-J-t--r-+tmiddotmiddot~~bull+-t--1~~bull-+-1 Id middot t 4 E1 (
I ncrementa 1 Concentration (~gm3)
Figure 11 4-4 Cumulative Population Exposure to Asbestos From Johns-Manville Plant Stockton
I
)
--~ ~w t _ ---~- ----------~ --____~-=~ M
~
bull I bull I I I I ii i Q t I i E e- I Ii tl I I I 6 I a I I J I f t t bull I G fl I
a I e 1 1 a 1 1 1 1 iii bull bull 1 1 bull 1 1 bull 1 1 bull bull bull bull 1 1i bull bull 1 1 bull a bull 1 1 r ~ middot u ~ ~ I0
+H~ ~ ~~~~-~~~~-rmiddot-~~~~~~~~~-~--~~-~~-~t~~~~~-~J C
~1_ c bull 1 t~ bullbull bullabmiddot 11 l bullmiddot II bullbull C- I u~
0 rl
gtlt I I I- 20 () ()
CJ -1 ~_J
J Cbull s
I If I I I I0
C bull I Ii I I I I I I I I0 1E
bull-+l I t I I ro s Lower+l
I I I G I iii I bull
() 14 I I IC
u I0 I I II Iu
C
1-_~I fi
CJ ~ I-- (j) CJ I G I I p +l
ro +-l I I I1~3~V)
0 +) UDf)Pr u I I 11 I I
CJ 1iII
Vi I 0 I I It- I a X
0 ~
w I I I I bull
Lower C 0
middotr-
- 4+)
(1j
J Q_
0 bull1
ij a 3- ti I
D d I I I I
I I I bull
II B I t
i ] i amp I
1 ei ~ amp 3 ii
I I I I
e ft a a I bull G
I I I bull I
I I bull I I t If
UpPrf
I I
I I I-_L~ lI I I f
I I I I I
- I I
j I I II I -
lI ImiddotbullII
l I
l Li
Incremental Concentration gm3)
Figure 114-5 Cumulative Population Exposure to Perchloroethylene and Carbon Tetrachloride From Dow Pittsburg (Solid Lines) and to Carbon Tetrachloride From DuPont Antioch (Xs) Results For Upper Lower Bounds on Emission Estimates Are Shown
1i-1 - middotbullth~middot t =lttF--i-+-middotr+-r+middot~middoti-middot-~-4-4middoti t -~--is -~middot-rmiddotibull--+ u ~~ i--t4~ t-bull~~311 li-i1middoti- C - JJ _ CL
-- I 1 t middot1 c7 Ibull ) ~ I-middot bulli C bull 1-_ 11 11gt1l11 ~bull -a bull-bullbull t a- II bullbullbull bullI lt_I Iii _- I
Incremehtal Concentration (microgm 3)
Figure 114-7 Cumulative Population Exposure to Carbon Tetrachloride From Allied Chemical El Segundo (Curves Corresond to Upper and Lower Bounds on Emission Rate Estimate)
--- I
~__~-unt2wbull--o1WCUatJ41ia1
16 0amp (V)
0 ~
gtlt
14~11l 1l QJ _J
~
1 middot-0
c t1middotmiddotmiddot 0
micro ro
_ c micro ~ 1euro1pound1
0- QJ u-J c 0 u O =-middot ~-= QJ
c_1 micro ro +- V)
0 micro ~ t1 O QJ 1l 0 a 40gtlt w c 0
-
ro --
micro
20middot l ---a I ~ I a 0
0
pound1r-t k
Figure 114-8
8tWI-S~1111111N~-
a I __ I wi~ i awLi--~~ 2 I I
1--+ I ~--if bull =~~--bullJ-+--f-1-----middot+--+middot-middot--r9c- 1 c- J
bull _ bull -~
Incremental Concentration (microgm 3)
Cumulative Population Exposure to Carbon Tetrachloride and Chlorofonn From Allied Chemical El Segundo Assumingsix-months Operation With Each Feedstock (Curves Correspond to Upper and Lower Bounds on Emission Rate Estimate)
i u---J--bull-+ i i--+ --4--+-imf-bullbullQ~-+-+-+~bullbull-4 171 i R = Fi middot1 1bull1 r1__ bullltaa ~a
Incre0ental Concentration (microgm 3)
Figure 114-9 Cumualtive Population Exposure to Carbon Tetrachloride From Allied Chemical El Segundo During Offloading From Midnight to 6 am middot-middotmiddot------middot--middot
--bull ft ampLA 1111-~lliaIN_WPl1A 1JIU_Mt91 MJ1WWlaquoPI-W1111~ --------__----Et1
M gt~middotI
0 -I
-X
~ t~ bullM
V) middot-middot ~ V)
QJ _J
f 0
0 -
C -4 0middot~middotmicro re f micro C a u
10 Cdeg u 0 3 0middotmiddotmiddotbull~ -0 a
+-gt ro
+J V)
+-gt 0
2E1-0 QJ V)
0 a X
LLJ
C 0 1pound1--middot+-gt ro
-- Q_ 0 0
I I
~ -~middot bull I -bull-_ bull I
middot lbullgtbull___I I
frac14 llk I
I bull
-~
I
-1_ I
Ibull II__ JI
I~
(1 f---1-+ bull-r_--f bull middot~-bull +-- bull i -bull 1 -1-+--+bull-+-t-+bull--t1 ~ _ 4middot 6 E 1t1 1middot2
1 3 5 I middot~ _ 11
Incremental Concentration (ngm3)
Figure 11 4-10 Cumu1ative Population Exposure to Ethylene Dichloride From Stauffer Plant Carson
_~ ~ t M
C X E k1~
-
- c- 1-1 ~ C middot-middot -_J
I--- -J 4t1middotbullmiddotmiddota
D
-+-shy --middot liiit1J
r ~
ltL
t-1 _ V
0 -+-1
middotr -bull ~1
X
C
+- bull1 ~1 atr l -=
~~lll(k~IHmiddot~~-Wa~middot~nL1middot~~~ta KF u n P middot~dE-tveRS ~-tl~~~-u-~~~tmiddot_-~~~~liil-~-~~~ tc r
I bull
bull -r~l-yen--~1l~a l fl __ A4o bullbull bullbullbull ~ 111bull--middotbull
8 NJ industrial sites 28 samples in industrial areas with benzene consumption facilities Torrance CA Tuscaloosa AL National Forest 1000 sarip Ies-Torontt) Azusa CA El Monte CA Long Beach CA Upland CA Los Angeles CA Riverside CA
bull Assumes all year operation on either feedstock If plant operates 50 on each feed the population exposed to greater than 50 increment over background goes to 16025 - 19377 and 4044 - 16025 for 100
Ambient concentrations of benzene vdry over one order of magnitude in the literature and therefore make tnis calculation questionable For Kaiser therefore the carcinogenic PAH assessment was used to evaluate incremental population exposure above background
+ The partition of emissions from Oow are 78 carbon tetrachloride and 22 perchloroethylene by weight
173
1142 Comparison of Incremental and l3ackground Concentratiors
Those sites whicn elevate background concentrations greater than SO~~
City Kentucky 6600 The latter two are associated wi t11 EOC pl ants Data
taken in senFice s~ations ano lrctffic areas ir various cities range from -~
300-3640 ngm~ As previous~J mentioned the Stauffer plant is discontinuing
operations These two sources should be examined as part of any possible
start--up permitting activity
11422 Kaiser Steel Corporation
The Kaiser steel plant polycyclic aromatic hydrocarbon (PAH)
emissions cffmiddotise from the coke oven omicroerations Comfdrison of the cumulative
population exposure Figure 114-3 and Table 114-4 with the specified
background levels for urban areas illustrates the breadth of the exposure
distribution The five known PAH carcinogens that were isolated in the coke
oven emissions were quantified in tile ambient air of Los Angeles by Gordon
1Y76
Benzo[a~pyrene 046 ngm J
t3enzo~e]pyrene osio Benzaanthracene U lo
Chrysene U60
lndeno Cl23-cd~pyrene 134
Althouyn these concentrations ctr~ low co1npared with 1 number of other cities
cited in the literature and repres2nt a very I imited data base tt1e predicted
concentrations from the Kaiser plant ~enerally exceed these levels by d
174
significant margin ie greater t11an 30000 are predict2d to be exposed to
4redter than l()-lt t 1tbull 1111tlit~nt background of 3j nqm 3 It ic 1 iklly tlat
lHrllerte exmicroo~urt 11 Ilie drld dre also 11evatPd ovr r11nbient I1owever since
backyround concer1tr1tions of t)enzlmiddotne show large variation no population
calculation was specified
11 4 bull 2 bull 3 Dmv
Carl1on tetrdcnloride releases from Dow constitute approximately 78
of tl1e total CT plus perc emissions in Table 114-6 and were found to elevate
urban background concentrations greater than 50 in five census tracts Emissions are predominately from storage and check tank working and brething remiddot1 eases
1142~4 Allied Chemical
The Al lied plant was modeled several ways since the plant can
operrttt~ wiUl chlorotor111 dr cc1rbon tetrctlt hloridt fo(~d The c1bull-i pr~middot11ntlid in
ldlgtlt~o ll4- drld llil-B represent dflrlUil 01Hrdtion with eitt1er teed Partial
yedr operation with each feed can be scaled from the individual annual modes
and is presented for equal hltllf year operation in Table 114-9 As with the
Du Pont plant carbon tetrachloride emissions arise from feed tank working
loss fable 114-10 illustrates peak exposures predicted to arise during an
off loading cycle As expected concentrations in that six hour period far
~xceed annual average values Insufficient data were available to contrast
levels with backyround transient concentrations
175
l~O
AUHLAlHJ~ CONTROL n CHNil)UES
Alternative control approacnes for the most significant emission
sources dinmig ~he various pl1ants n ees fitie(J n the follo11 ric suJsections
EmpiiasVi nas Deen placed in dedlinq ~riU1 Uiose sources of 91reatest absolute
inaynHud2 (lbyr) and those const1 ut 0ing trie largest increm121middot1t middot] bulli~he
bdc kgrnund con cent ration of Hie em~ tted subs tcince i~o ctt tention Wi 11 be ~i ven
to the s~iondc1tY sources witt1middotn cK1 microant or to trlL case of J011ns-Manvi I le
since -UH imiddotajor source is less tran LOO 1byr and is not microrejicted to raise
bctckgro 1wd exposure levels to the ~~ntr21 population Furt12rmore a11
emission sources dt Kaiser Ste 1 1iC(1 v1 1~n directly dedlt -1ith in this
program r1re elltlteci ro t11e cc1lt1bull ovcmiddotr nperatmiddotions These faci 1ities are to be
closed du1rm and all primary reel 11i~ oH~rdtions discontinmiddot 2d Kaiser
forecasters have predicted further deterioration of the plant economics and
the phased closure has been accelerdted for primary steel making operations
Note that this closure is essentially irreversible cince differential
exmicroansion and contraction of the coke oven structur1 occurs in the cooliny
process and it would be improbable that ovens could be reheated without
extensive reuuildiny at major expense
12l STORAGE TANK EMISSIONS - fAUFFrn IHJW OUPUNf ANU ALLIED
At c1ll four sites emissions from storage tanks constitute either the
primary or near dominant (Stauffer) source of carcinogen release andor
genero1i iJOpulation exposure Curreritly the tanks of interest at each site are
permitted by the local Air Quality IJistricts hJwever they do not require
emission ontrol systems for vdrious reasons Tl1e estimated releases grounds
for exemption and other pertinent information are given in Table 121-1
In order to amicromicroreciate tth~ practicdl dlternatives for emission
controls the provisions of SCAWMU Rile 463 ar~ listed below which specify the
acceptable alternatives for tdnks r4uirin9 controls ie tanks 11aving
capacities greater than 39630 yal with suostances of true vapor pressure
176
Table 121-1
Site
Stduffer Uff-Site Tanks (J) (Sdn Pedro)
Allied (El Seyunrlo) - KdJ
1bulllctLer1 d I reld iturctjc
-J -J
Dow (Pittsbury)-Product Check Tanks (4)
Uow - microroduct sturag~
DuPont (Antioch) - Raw Materidl Feed Stordge
Approx Capacity~
6lUS x 1g (c) 84 X 10 (1)
lt3Y630
18000 each
70000L) 3UUOUO
30000 working 42636 liquid full
STORAGE TANK
Substance
Ethylene Uichloride
Carbon Tetrctci1 I Jr i J1
Perchloreshythylene and Carbon Tetrashychloride
CT perc
Carbon Tetracr1 l ori de
SUMMARIES
Emission Mode
Tank breattli ng
Tank v10rk i ng Tank breatttiny
PERC-~JOrk i ng PERC-b r2a t ~ i ng CT-working CT-breattling
Tank breathing Tank breatt1i ng
Tank workin9 Tank breathing
Vamicroor Pressure
014 at 70 F psi
1 63 psi at 68 F
16d psi at 68 F(CT) UJ9 psi at 68 F(perc)
168 i)Si dt 68 F
Grounds for Exemption
Exempt from control re4uirements to SCAQNJ Rule 463 since vapor microressur-2 dues nut iXC~ec
l~ io at sturdJ~ conditions
txe11pt TlJil fule -+b3 since tank capaL ~J ~ s u~der 39630 gal (lSO 11 )
Exempt from Regulation Rule v~~vu u~~~~~shytank capacity is under 39630 ga 1 andor substances are not classified as organic 1i4uicts
Exempt f ro1n Ru l ~ 85300 because subsLctnce is not classified as dn organic li4uid accordin to Regulation-Rule 8120
exceeding l~ microsi at storage conditions
~ floating roof tanks
~ fixed roof tanks with an internal floating type cover
bull a vapor recovery system with vapor collection and retllrn (or disposal) proce5sinJ lXctbullerlin~ 95 efficiency
ln a cllemicai plant a typical vapor recovery systein (KR Evans
SCAQMU) migm consist of a collection inanifold to d recovery system (suctl as a
vapor sphere) to a compression syst2m dnd subsequently to absorption and
recovery systems The absorpt~on st~n c~nsisting possibly of (rubber
stripper or activated carbon
In dealing with speciti( carcinogenic substances such as vinyl
chloride hiyhly efficient vapor control system perf0rmance has sometimes been
stipulated necessitating incineration systems
Fer th~ cases of corccrn realistic alternltltives are as follows
Stauffer Off-Site TanKs - Tnere are a number of options which can 1~arkedly 3
reduce ei1issicns from these tanks from th~ calculaL~d value of over 20 x 10
lbyr One alternative is to consolidat~ the material in tne three tanks
One of the large tdnks can contain the currently stured material and reduce 3emissions to apmicroroximately 13 0 x 10 lbyr Anotl1er alternative is to
transfer dll E0C to a single floating root tank Various types of such tanks
exist dlld are reviewed in the EPA report Lrganic Cht~mical Manufacturing Volume
J Storage Fugitive dnd Secondary Sources EPA-450J-80-025 eg internal and
external floating roofs and a variety of seal desiyn configurcttions Generalshy
ly such designs would be expected to reduce emissions to the order of
one-fourth to one-fifth the current level Cost of 1nstalliny ct contact
single seal internal floating roof wds estimated by BB Lumquist Pentrex
for EPA as $27770 in 1Y8U dollars for 7U ft diameter tdnk Cost to build an
external floating roof tank was estimated by G Stilt of Pittsburg Oes Moines
for EPA as ~14UUUO for D=67 H=4Uft Anottler com111on approach is to utilize
cdroon ctdsorption This works we11 witn nonpolar hydrocarbons dS VvC is
removed from the vapor phase A basic syte111 consists ot tvo carbon beds and
a regeneration system Regeneration is typically performed with steam or
vacuum Figure 121-1 illustrattS rhe systems (Basd2kis 1~110) Steam raises
tne VOC vctpor microressure The resulting sti~am-VOC mixture is condensed and
173
AC-TJA~D CA-e0-l
-----~---
LOW- ffiESiJ~C ~T~H----
PiltOCESS COOL-J~ At-JO oLOWER DRY t-J~ 2gtLOWE~
COgtJDENSER
J
COOLIN(a - --------
WATi=-A
RECOVeRE SOLVENT
WASTE WATER
Fi l 1 1 middot gure middot - Activated-c arbon Adsorption System
179
rout2d to c ~epordtor decanted and returned to storage In vacuum regenerashy
tion VOC vapor is desorbed by pu1 ling a vact1um on the carbon bed then condens-
ed a1d returned System efficiency ~ ~ est ii1a ted at 96 f(duc middot i m from fixed
roof levels (EPA 4503-80-025)
H~fri 9ctated vent concicr1sers dre une of the most cu1mon emission
reduction prJcesses -for conttol ~ ng hx2-d-- storage tai vUC Figure
12 1-- 2 i l 1ust rates a u n i t ( Er i scmiddot n 19HU ) eff i c i en cy of middot middot~ ~ c middotlt middotmiddot _y i s rate ct
bet1Jee~middoti 60-90 for the vapor pressure range of concern For such a large tank
the capital costs would be high ~igure 121-3 illustrat~s EPA estimates
(EPA l ~~SOb) for the condenser section Condenser sys tern a ea mu l d be in
excess of 1000 ftL
It should be noted t -at Jtc1 uf fer Uiemi cc1 l has arH1ounced the c 1 osu re
of its VCPVC plart PrevfoJsly tt12) had planned to purge tlle off-site tanks
of EUC Since any future po~sible micro1ant stdrt-umicro will necessitate a compreshy
hensive SCAQMD review this ltoument can assist in evaluating proposed control
measures
Allied and DuPont Feed Tanks - In bott1 of these cases the nore significant
quantity of emissions arise from tank workinltJ ie during the off-loading
activity rather than tank breathing Contnl ~easures taken for working
emissions are thus of primary concern Ther~fore no detailed discussion will
be provided on the alternatives for control gtf bredthiny emissions Additionshy
ally both tanks are in the range of 20U00 gallons which is a capacity where
floating roof tanks are almost nonexistent Out of 670 floating roof tanks
surveyd by EPA less then 15 were smaller than 30000 gallons in capacity
Therefore the utilization of any floating roof concept and seal combination
will not be considered
Commmerc i a11 y it is estimated by Ou Pont that carbon tetrachloride
feed costs approximately $017lb (E Taylor personal communication)
Ttierefore less than i2500 is lost to l)uPont annud l ly as a result of working
l eve l em i s s fo ns and 1es s fr om Al l i ed bull Thus i t is u n l i kel y that any
appreciable economic incentive exists to dev1~lop ci vapor recovery system
However a Cdndidate system could be pattern~d after the chloroform
feed-storage unit currently at Allied This is a dedicated vapor balance
system Chloroform is off-lndded from tank cars and stored in a closed
unvented tdnk As the 5tordye tank is filled the c1ir spdce d1spldced is fed
Figure 121-3 Installed Capital Cost vs Condenser Area for Various Materials of Cunstruction for a Complete Condenser Section
182
back to the tank car Also in this case the storage tank is not vented in its
normal breathing mode and is part of a closed feed system to the reactor
Vapor recovery system alternatives which are particularly effective
in loading and handling include r~frigeration adsorption andor absorption
Control efficienci~- 1re est1mateJ in the rlt1nge of 90-95 (EPA 1980b) and of
course depend on the speci tic su1)Stance and equipment used Carbon adsorpshy
tion systems were discussed above The smallest capacity carbon adsorptionmiddot
system priced by EPA (EPA 1980b) has 2 vertical beds of carbon (900 lbs - 4ft
diameter by 3 ft depth) witl1 an ilstalled cctpital cost of $135000 based on
December 1979 dollars
Dow Tanks - Dow has two pair of p~oduct check tanks which alternately are
filled dnd off-loaded Emhsions from th~sl tanks were calculated based upon
operating cycle (3 day fi 11) satubull-dtion vapor pressure and physicdl
chardcteristics Direct hec1d-spa1e testing was planned however it could not
be accomodated because of r 1stri c ed access due to unscheduled mai ntendnce on
the field test day
Dow has indicated that chey are studying the option of installing a
vapor controlrecovery systt~m in 1hese and their largerproduct storage tanks
The extent of their engi neri ng ind assessment work is unknown as are their
current plans It may be possibk to incorporate a vapor balance design into
the system whereby the disp1aced apor is trdnsferred to another process point
within tl1e system Alterna1ively a vapor r1middotcovery system such as carbon
adsmiddotorption is fedsible Hot-1ever since emissions for t11e check tanks are
primarily due to working loss tht~ use of a conservation vent or an adjustment
of its operational differential p~essure would be ineffective for tr1e
reduction of the bulk of such emissions Furthermore since check tank size
is re1at i ve 1y sma 11 no cons i derctt I on was given to conversion to a fl oat i ng
roof configuration for those tanks Conversion would be possible for
the large storage tanks
12~ WASTEWATE~ EMISSIONS - STAUFFER
Emissions of me throug11 ~-a~)te-1dter discharge are the largest single
source identified at the plctnt sites The discharge limit was set at 25 ppm
by the Los Angeles County Sc1nitation iistrict and was estdblished with the
occupational limit in mind of 50 ppm JVer an 8 hour work shift dt the District
18~
tredtrient center~ T11ere hctd been suine di~cuss1on between pdrti~s of possibly
raisiny tne discharge limit since it is felt by Stauffer that dilution of the
stream is sufficient to d 11 ow it Itmiddot shou Id he noted t10-1ever tna NIUSH has
recommended the permissible exposure limit be reduced to 5 ppm averaged over a
work period of 10 nours microer ddy 40 ours per Ieek witt1 a cei l iny level of 15
pmicroin dVeraged over a 15-mi nute microet~ v(J
It is not certain at 1~hrt rdte me 1rill be reledscd from the
wastewater stream At the microlant discharge points wastewater is both hot and
aerated thus favoring release~ No measurements have been ta~en downstream of
the plant after considerable dilut1on has occured The distance to the water
treatment pa11t is approximately 5 kin at 1i1ic~1 point anerobic diyestion is
conducted lt is presumed that a11 ELlC will be released before final ocean
di scnar~Ji
A possible emi ss i D~ cont ro process for nmiddotduct ion of the EDC in the
effluent is by ~recess adjust~Pnts or additio~s For example process
modifications to the EUC stripping stage coula dramatically reduce discharge
lev~ls A control alternative is the use of octivdted carnon or XA0-2 resin
to recover EOC in tne discharge stream Tests of Gulf South Researd1
Institute on XA0-2 and activated carbon (Coco 1980) show high recovery yields
for nonpolar organic carcinogens unaer a range of concentrations Viable
suggested alternatives oy Srnitn included regeneration of Ue trapping
materials dnu even consideration of burning tt1e concentrate carbon media (at
greater than 1000 pmicrom)
Control approaches to reduce emissions from so called secondary
sources in general and waste water emissions in particular fa 11 into four
categories - waste source control resour(e recovery alternative disposal and
add-on controls Alternative control pro(~esses which were considered but
appear to be inappropriate to this case i11clude chtmiddotmica means eg
neutralization precipitation coagulatior1 and chemical oxidation thermal
destruction of tne unconcentrated ~dste s1ream is ii1practical biological
treatment eg aerdtion and biomass-wasteater contcct ~ienerally relates to
the treatment of soluole degrddable organics in the concentration range
bet~~een U01 and l~~ terminal storagt eg lanctfilliny urface impoundment
and deep well injection dre either indppl 1cable or 1mmicroractical Therefore in
summary the poss i b I e contra l approac ties fur this ca e inc 1ude the improvement
r
184
of semicrodration efficiencies in steam stripmicroing tl1e internal recycle of ~aste
stredms ctnd the ddsorption b activdted carbon The design configuration
~fficiency and cost )bviously depend upon numerous microlarit specific factors and
their determination vJOuld rec-Jire detailed analyses
n1e ~~dsrewctter systbull-m of a model cl1emical production plant based
upon the average properties uf a composite of 30 chemicals was evaluated for
EPA by IT Enviroscience (EPA 1980b) Included prominantly among the 30 was
tUC with the hignest uncontrolled secondary emission wastewater release rate
ie ~ percent of tt1e production and 34 perc~nt of the emission Cost and
impact analyses were evaluated for alternative control systems to reduce
secondary voe emissions from wastewater Four systems were considered a
carbon adsorption system (eAS) for recovery of the voe from the wastewater a
cover to reduce secondary VOC emissions from the wastewater clarifier a cover
for the clarifier plus a carbon desorption system and a cover for the
clarifier plus ct eAS system usinu a fume incinerator The scale of the model
system was greatly in excess of the Stauffer plant thus further making
detailed comparison i mpract i cal bull However emission reduction factors ~-1ere 6diven as ~9 and cost effectiveness per 10 g reduction genera1middot1y ranged
between $450 to 1733 These factors would likely grossly underestimate the
system cost if scaled down to the range of the Stauffer plant ie the order
of 34600 lb annual discharge
The alternative approaches of improved steam stripping efficiency
and internal recycling of the stripper iischarge stream with a reduction in
makeup re4uirements could decreas~ net ~DC wastewater content release
123 CONTROLS FUR SECONDARY LEAD S~1tLTER STACK EMISSIONS
Given our findiny of low (16 kgyr) ~missions of arsenic from the
reverberatory furndce at RSR it middotJOuld ippear that the arsenic content of themiddot
lead feedstock is low andor that the microants system for reducing lead
emissions is dlso quite effective for arsenic ~SR it will be recalled from
Section 32 uses a quenchinu cnariber ctnd Dayiwuse filters to remove
pctrticuldte matter and a carbonate scruooer to remove sulfur dioxide
There are no federal new source microertormance standards for lead
eini ssions per se t10wever the Standdrds of Pbullrformance for Secondary Lead
SmeHers (40 CFR 6Ull2) limit total particuLte emission from a blast furnace
185
or reverberatory furnace to SU mgm 3 According to Augenstein et al (1978)
who reviewed the techno1ogy for control ing lead ~missions from ttwse sources
baghouse filters or wet scrubbers are yenerally used to contro1 particulate
emissions When fabric filters are used to control blast furnace emissions
they are nurma11y preceded by an afterburner which inciner~tes hydrocarbons
that bull~GU Id otnerwi se b 1 ind t ne f o1 ur-i c After-burners a re nor necessary for
reverb2middotmiddotatcry furnace emiss-ion smiddotnce the excess air and temperature are
usually sufficient to oxidize tn2 hydrocarbons
According to Augenstein et ali 11 sr1aker--type baghouse filters are
the 1nost effective means of controlling ead fume emissions from secondary
furnace opErations 11 Collection etficiencies can exceed 99 oercent One
advantage to this control approact is that l~ad oxide dust can be recovered
easily cmd recycled in the smeller Flue gases must be coated to below 300 degF
for dacro~ bags and to below ~00 degF for fiberglass bags (Hi~h et al 1977)
Although wet scrubbers are effecti1e under some circumstances in
controlling lead emissions it is more difficult to recover the lead oxide for
recycling In addition sulfur dioxide presi~nt in the flue gases becomes
oxidized to sulfuric acid and can cause corr1sion problems For this reason
sodium carbonate or other basic reagents ar~ added to the scrubber solution
Although it concerns a yold smelter an approach described by
Marchant dnd Meek (1980) provides an exampli~ of an arsenic control
alternative which might be apmicrolicatgtle to secmdary lead smelters At the
Campbe1 1 Red lake Gold Smelter in Balmerton Ontarion Canada Smelter gases
are first passed throuyh a hot electrostatic precipitator (ESP) The ESP is
heated so that the arsenic (which is principally in the form of As ) remains2o3 gaseous and is not yet collected This exclusion of arsenic allo~~s the ESP to
recover particulate gold rnore easily The ESP exhaust is then quenched with
ambient air to condense the arsenic trioxide Baghouse filters then remove
the arsenic along with other particulate matter
124 CUNT~OLS FUR STEEL MILL EMISSIONS
Given the imminent and irreversibli~ cessdtion of coking activities
at Ka 15 er Stee 1 Corpora t i on a rev I e~ o r t e c Imo 1 o gi es for ( on t r o 1 l i ng
~missions from the coke ovens rnd he Clt1ke byproduct recovimiddotry pldnt WdS not
deemed to be necessary Tnis is t1e only primary steel mill in California
1Pi6
KEFUENCE)
Al lf~n Jr Ll~ llJBU1 J 111odel to 1middotstimite hi1drdous (bullmissions from coke oven Juurs pr~pdrelt1 DY 1-kSecircn Tr1dnye7nrffut fur U~ Environmental Protection Agency Uffi ce of Air ~ua 1i ty and Standards fltesearch Tri ang I e Park Nortn Cdrolina KTl17362-UL
Jl len Jr CC 1~8Ub tstination of chdr in(i emissions for by-product coke oven microrepared by fltesedrch Tr1anglt Insc1tute for US nvironmental Protection 1-yency Uf f ice of Air l)ua l i ty and Standards Ke search Triangle Park North Cdrolina RTI17362-0
Allen Jr CC 198Uc 11 Environmeritdl c1ssess1ent of coke by-product recovery plants Proceedings First Symposium on Iron c1nd Steel Pollution Abatement Technology Ctncayo Illrno1s--Tin1ctober - 1 November 1979 EPA-600-9-80-012 pp 7~-dd
Augenstein ur1 T Cornin et al 1978 Con_trol techniques for lead air emissions prepared by PEDCU Envircnmental Inc for US Environmental Protection Agency Research Triangle Park North Carolina EPA-4502-77-012-B ( NT IS P88U-197551)
jarber WC 1977 Interim air pcllution a~sessment for eight t1igh-volume industrial organic chemicals 11 US Environin~ntal Protection Agency memorandum to Air and Hazardous Materials Uivision Kegions I III-X and to Uirector Environmental middotPrograms Uivision Region II
t3drrett KE WL Margard et al 1977 Sampling and analysis of coke-oven door emissions Prepared by ~dttelle-Columbus Laboratories for US Env 1 ronmenta I Protection Agen y Industri a I Environmental Research Laboratory Research Triangle Park North Carolina EPA-b002-77-213
t3 a r ret t flt bull r bull and P bull K bull ~~ebb 1Y78 bull 11 E f f e ct i venes s of a wet el e ct r o stat i c precipitator for controlling PUM emissions from coke oven door leakage Presented at 71st Meetiny of the Air Pollution Control Association Houston Texas 25-29 June 1978
l3asdekis HS 1980 Carbon Adsorption Control Uevice Evaluation IT Enviroscience Inc report in prepdration for US Environmental Protection Agency ESEU
8ee fltW et al 1974 Coke oven charging emission control test program Vol I c1nd II US Environmental Protection Jgency EPA-6502-74-062
~eimer RG 1978 Air pollution emission test Analysis of polynuclear aromatic hydrocarbons from coke oven effluents Prepared by TRW Uefense and Smicrodce Systems Groumicro for US tnvirorunental Protection Agency Office of Air wuality Planning and Standards Kes~arch Triangle Park North Carolina EMB Kemicroort 8-CKU-12
187
Braman RS 1976 Application of the arsine evolution methods to environmental analyses presented at the International Conference on Environmental Arsenic Ft Lauderdale Florida 5-8 October
Bratina J E 1979 11 Fabric filter applications on coke oven pushing operations J Air Poll Cont Assoc 29(9) lll6-920
tiuonicore AJ 1980 Environmental assessment of coke quench towers in proceedings First Symposium on Iron and Steel Pollution Abltlternent Technology Chicago~ Uirnois JO October - f N~vernber 1979 EPA-6009-~1b-62~ pp 112-142
t3usse A D and dR Zimmerrao1 1973 User 1 s guide for tdegh~ Climatological Uisper~ion Model US Environmental Protection Agency Resea~ch Triangle Park Nortl Carolina) EPA-R4--73~0024
California Air Resources 8oard 1976 Coke oven emissions miscellaneous emissions and their control at Kaiser Steel Corporations Fontana Steel making facihty Enforcement l3ranch) S2crcmento California L amp E-76-11
Calvert~ CA 1976 Envir __ Sci_ i( __ Technol 10256
Coco L~i 2t aL~ 1979~ 11 02c10pmert of trec1tment and control technology for 11refractory pet rochemi co was tt=1s r middotl 1d l Report Gu 1f South Research In st it ute
to U S tnv i ronmenta l Protection Agency EPA-ti002- 79-080
Colucci~ LM and CR Begemi 1971 Palynuclear aromatic hydrocarbons and other pollutants in Los Angeles air In Proceedings of the International Clean Jir Congress Vol 2 Academic Press pp 28-35
Cooper JA and JG Watson Jr 1980 Receptor oriented methods of air partkulate source apportionment 11 J Air Pollution Control Assoc 30(10) 1116-1125
Coutant R W JS McNulty and RD Grammar 1975 Final report on determi 11at ion of trace elements in a combustion sys tern Prepared by Batte11 e Columbus Laboratories for the Electric Power Research Institute EPRI 122-1
11 11Dilling W 1977 lnterphase transfer processes Envir Sci TechnoL 11 4 pp 405-409
Oow1irgl) MP JO Jeffry and AH Laube 1978 11 Reduction of quench tower emissions Presented at the 71st Air Pollution Conotrol Association Conference Houston Texas 25-30 June APCA No 78-92
EPA 1980a Review of Criteria for Vapor Phase Hydrocarbons 11 US Environmental Protection Agency ~eport 6008-80-045
EPA 1980b Organic Chemical Manufacturing Volume J Storage Fugitive and Secondary Source EPA report 450J-80-025
EPA 1981 Compilation of Air Pol lutdnt Emission Fdctors (Including Suppleshyments 1-7) Third Edition Suppl1111ent 12 Utfice of ir l)uality Planning dnd Stdndlt1rds AP-42-ED-3-Supp I - lt
188
Erikson Uli 1~80 11 Control Uevice Evaluation Condensdtion 11 IT Envioscience lnc report in prepdration for US Environmental Protection Agency ESEU
lrtel liL 197Y Quench tower particuldte emissions J Air Poll Cont Assoc 2Y(9) 913-916
Ev a n s K bull P bull 1981 Pe r son a l Commun i cat i on v i th R bull Zi s k i n d bull
Goodwin DR 1980 11 Air pollution emissions standards in Proceedings First Symposium on Iron and Steel Pollution Abatement Technology Chicago 11 lrno1s 30 October - 1 November 1979 EPA-6009-80-012 pp 37-45
Gordon GE 10 Receptor Models Environ Sci Tech 14(7)792-800
Gordon KJ 1976 11 lJistribution of airborne polycyclic aromatic hydrocarbons throuyhout Los Angeles Envir Sci Technol 370-373
Great Lakes Cdrbon Corporation 1977 Study of coke side coke-oven emissions~ Vol 1 Source testing of a stationary coke side enclosure EPA-340l-77-014A
lirimsrud EP and HA Rasmussen 1975 Atmos Envir Y 1010
Hartnan MW 1~80 Source test c1t US Steel Clairton Coke Ovens Clairton P~nnsylvania Prepared by rRW Environmental Engrneerrny D1v1s1on for US lnv1ronrnental Protection Agency Emissions Standards and Engineering l)ivision Research Trianyle Pdrk North Carolina EM~ Report 78-CKU-13
Hendriks Ilt V et a I 1Y7Y Urgani c air emissions from coke quench towers 11
Presented at 2nd Arrnual M(etiny ot the ir IJol lution Control Associdtiont Cir1cinndti Utlio June 1~7lJ l-dmicroer NU 9-391
Higt1 MU ME Lukey dnd TA Li Puma llJ77 Inspection manual for secondary lead smelters premicrolt1red b Enyineeriny Science lnc for US lnvironmentdl Protection Ayency Office 0f ~nforc~nent EPA-3401-77-001
Holzer G et al 197 11 Collection amp analyses of trace organic emissions from natura I sources 11 Journal of Cnromatogra flhY 142 pp 755-764
Science Applications lnc Final lemicroort to US Environmental Protection Agency tPA-560lJ-80-UU2
J bull Mi l n e l lJ 81 Person a I Cornm un i cat i on wi t t1 bull Zi ski nd bull
Kemner WF 1979 Cost tffectiven2ss model for pollution control at coking facilities Premicroarect oy PEDCo Environmental lnc for US Environmental Protection Agency Industrial Environmental Research Laboratory Research Tridnyle Park Nortll Carolind EPA-GU02-7l-1U5 (NTS PB 8U-118706)
Kessler T AG Sharkey Jr and kA Friedel 1973 Analysis of trace ele111ents in coal by smicroark-source mass spectrometry US 8ureau of Mines Report ot I n v es t 1 gd t 1on s No bull 77 14 bull
189
Kolak NP J Hyde and R Forrest~r 19 1 9 Particulate source contributions in tt1e Niltlgara Frontierffi Prepared t)_y Nr2bulll York State IJepartrnent of Env i ronm~ntc i Con serv d ti on for US En vi r inment ct l Protect ion Agency Region 11 EPA-9024-79-006
Mackay U and PJ Leinonen llt175 Hate of Evamicroordcion of low - solubility contaminants from vJater bodies to c1tmospmre 11 Envir Sci Tecnnol 9 13 pp 1178-libU
Magee tiv HJ Hctll and GM Vc~rlt_a Jr 1973 Potential Doimiddotiutants in fossi1 f1Jel Prepared by ESSO fbullt~scarcr middotnd Engineeriny Co tor US Enviromihm~ct Protection Agency Ufl-R2-7 --2l1~L (NTIS Pt3 as u~9)
Marcriant middotG0H ard RL Meek 1~gu Eva uation of technology for control of arsenic tmissfons at the CdlmpJel Hd Lakmiddot G~ld Smelter prepared for the US
1Env i ronmentd l Protection Agenc)-l-l(lU-sfrT l lnv i ronmenta l kesedrCiI Laboratory Cincirrn2ti Uh70 lYA-60U2-8U--141 (NTIS 8 oU-21Y36J)
Margler L$ HA Ziskind M8 lcgozen (Ind M Axelrod 1979 11 An inventory of care i nog)n i c substances re i ea sect into ne ambient air of California Vol une I - Scn~enin~ and id2nt~fication o-f r~1rci1ogens of greatest concern Science J- pp 1i ca middott~ L J s In c bull Fi nd l kcmicro~ rt ~ i I - U 6 C) - U- 5 0 4 to U1 e Ca I i ~- o r n i a Ai r kesources 130ard
Milne J 198L Persond1 C~mmunicdtmiddotion middot1itt1 R Ziskind
llurchio JC WrC Cooper and A ue Leo11 1970 Asbestos fibers in ambient air of California 11 Cal Horn i ct Air lesourc2s l3oa rd ARl3 4-0S4- l EHS Report 73-2 (March)
National Academy of Sciences 1Y72 Part1culate polycyclic organic matter Washington~ DC pmicro 4-ll
Nati ona I Kesedrcn Counc i I 197b Arsenic prepi1 red by Subcommittee on Arsenic Co111mittee on 1bull1edical ctnd t)ioloyicai Effects of Environmental Pollutants washin~ton uc
Oliver JF and JT Lane 1979 11 Control of visible emissions at CFampl 1 s coke plant-Pueblo Colorado J Air PolL Cont Assoc 29(9) 920-925
Pe I I i z z a r i E bull 0 bull anct J bull E bull l$ u n c h 1s 19 1 nbi en t ai r ca r c i n o gen i c v a po r s Improved samplin~ and analysis tectrnology EPA Contract 68-02-2764
Pellizzari E U 1977 11 Tle 11easure11ent of carcinogenic vamicroors in ambient middotatmospheres EPA-b007-77-J55
Phelps RG 1Y79 lSI-CPA-oatLlle c-ilte oven door sealing proyram J Air Pol 1 Cont Assoc 29(J) JlJ8--JlL
Porter RA 1976 Uismicroersion ~yuation s lucions Dy calculator I guide for air pollution engineers ctnd sci12nthts S cond edition rexas Air Control tioard ~MT1S P~ 262 lJ~)
Redburn T 1980 Kaiser battles aginst l~gacy of woes 11 Los Angeles Timesmiddot (7 Semicroternber irno) Part VI pp 111
Research Triangle lnstittJte 1977 11 Qudntification of benzene in 15D -1bullnoient air samples 11 for US Environmental 11rotection Agency Office of Air Quality Planning a11d Stitndards
Rinkus SJ and MS Legator 1979 Chemical characterization of 465 known or suspected carcinogens and their correlation with mutagnic activity in the Salmonella typhimurium system Cancer Research 393289-3318
Koberts R M 1980 11 An inventory of arc i nogen i c substances r~ leased nto the ambient air of California Tasks 11 and IV KVB Final Report KVB 26900-836 to Science Applications Inc
Rogozen MB LW Margler P Mankiemiddott1icz cind M Axelrod 1978 Water pollution control for coal slurry pipeiines Prepared by Science Appl1cat10ns Inc for US Department of Energy (NTIS SAI-068-79-516
Rogozen MB DF Hausknecht and RA Ziskind 1976 Methodology for ranking elements in fossil fuels according to their potential health impact Science Applications Inc Final Report 260-77-539 to the Electric Power ~esearch Institute
Russel 1 JW and LA Shadoff The sdmpl ing and determination of halocarbons in ambient air usin~ ~rncentration on porous polymers
Siebert PC CA Craig and Eb Coffey 1978 Prelimary assessment of the sources control and population to airoorne polycyclic organic matter as indicated by benzo(d)pyrene Prepare( by lnergy and Environmental Analyses Inc for US Environmental Protection Agency Office of Air Quality Planning and Standards Resedrch Triangle Park North Carolina
Simmonds PG et amiddot1 1974 Distribution of atmospheric halocarbons in the air over Los Angeles basin Atmos Envir Vol 8 pp 209-216
Singh HB LJ Salas and RE Stiles Distribution of Selected Gaseous Oryanic Mutagens and Suspected Carcinoyens in Ambient Air in Proceedings Annual Meeting Air Pollution Controrl Association 82-651 June 20-25 1982 New Url eans
Smith WM 1979 Evaluation of cok oven emissions in Yearbook of the American Iron and Steel Institute pp 163-179
SRI International 1980 Asessment oi human exposures to atmospheric benzene SJ Mara and SS Lee finai report to US Environmental Protection Agency EPA-4503-78-031
Taback HJ 1978 Control uf Hydrocarbon emissions from stationary sources in the California South Codst Air Basir Prepared by KVB Inc for the Cal1fornid Air Resources Bod-rdmiddotKVB No 5804-714
191
Trenholm J-f dnd LL )eek 1Y78 11 Assessment of t1azardous organic emissions fror1 slot type coke oven Datteries unpublished paper US Environmental l)rotection Ayency Ernssion Standdrds and Engineeriny Jivision
Turner U~ J970 Workbook of atmusphe~ic dispersion estiimiddot1Jt2s IJS Environmental Protection Agency k~sedrch Tri1ngle Park North Carolina
US Bureau of Mines lYl f-inercls yearbook 197U Vol 11 p 4~3
Van Os dE I l U bull iJ bull lL Ma rs l and et ct 1Y7~1 o En i r onmt~ nta I a s s es smen t of co k e Dy-product recovery plants prepareu by ReseMcn Trictngle lnstitute for US Environmental Protection AgerKy 11cJustrial Environ11ental R~search Laboratory Kesearcll Tri2n~e Idrk NorU1 Caroline EPA-6U02-7J-Ul6 (NTfS PB 293-278)
Westbrool( C W 1979a Level l ass~ssment of uncontrolled sinter plant emissions Prepared by fesearch Triangle Institute for US Environmenta1 Protection Agcncy Industria1 tnvirnnmental fltesearc-1 Laboratory kesearch Trianyle Park North Carolind EPA-bOU2-7Y-112
Westbroollt C~~ 197ltJb Level l css~ssment of uncontro1led l)-BOP emissions Prepared by Researcn Triangle institute for US Environmental Protection Agency Industrial Environmental Hesearch Laboratory Research Triangle Park ~Orth Carolina EPA-6002-79-190
Wetherold rLb LP Provost and CD S111ith 198U Assessment of atmospneric emissions from microetroleu111 refinin~ Volume 3 Appendix l3 Radian Corp Fina Report to US tnv i ron111enta I Protection Agency EPA-600 2-80-07 5c
Ziskind RA OF S1ilith JL Hdhn etnd G Smicroivey (1980) Determinants of Cancer and Carctiovasculdr Uisease Mortality in Asbestos Mining counties of California SAi Report 068-dl-514 l May
192
APPENDIX A
Rapid Screening and Identification of Airborne Carcinogens of Greatest Concern in California
Lawrence W Margler Michael B Rogozen Richard A Ziskind and Robert Reynolds Science Applications Inc Los Angeles California
This paper describes a method for establishing a priority list of airborne carcinogens within a
state jurisdiction In this case it was necessary to identify from among hundreds of potential
candidates the live to ten materials of greatest potential concern In California as airborne
carcinogens
Because no previous Inventory of carcinogens In California existed published lists rankings
and assessments of national scope were used to Identify candidates By systematic manipulation
and comparison of these data sources 47 materials of some notoriety were chosen for closer
scrutiny This selection was pared to 22 candidates largely by eliminating those which had
very little production and use In California (Substances primarily used as pesticides were
excluded from the scope of this study) The remc1ining candidates were then ranked by additive
and multiplicative algorithms and by a panel of experts The results of these rankings were
combined to produce a single selection of 11 priority candidates In alphabetical order they
ethylene dichloride N-nllrosoamlncs perchloroethylene and polycyclk aromatic hydroiarbons
In continuing studies a baseline emissions inventory Is being prepared and a source testing
program is being designed
In recent years concern has grown over su~pected arcinug-ens New Jersey seshythe possibility that certain materials lected ten volatile organic compounds released to the atmosphere through inshy and five htbullavy metals to be examined dustrial and commercial activity may be further and is currently measuring responsible for a significant portion of ambient atmospheric concentrations of the incidence of cancer in the general these substances in a variety of areas In population This concern is manifested the second year of the study the state at the federal level in the National h1s incre1sed the volatile organics Emissiltm Standards for Hazardous Air stt1died to 20 and begun measuring Pollutants (NESHAP) which limit heavier orianics associated with parshyemissions of the known carcinogens asshy ticulatcs2 In California a very different bestos beryllium and vinyl chloride 1 approach-was taken
Only a few states including New Jersey and California have begun efshy Overview of Californias Approach forts to identify airborne carcinogens of concern to the general public for the The California Air Resources Board purpose of setting state emission regushy (( ARB) is sponsoring a three-stage lations for these substances After reshy st 11d~bull of airbornC rmissions of carcinoshyviewing national use data for known and g1middot11s from anthropogenic atlivities The
November 1979 Volume 29 No 11
first stage which is the subject of this paper was to identify roughly five to ten materials which of the hundreds of known or suspected airborne carcinoshygens are most likely to be of greatest concern to Californias general populashytion Also of interest were those which in order to satisfy occupational health and safety regulations might be transshyferred from the workplace air to the outside atmosphere The second stage which is now underway includes pinshypointing of emission sources for each of the carcinogens of importance quantishyfiltation of emissions nnd design of source-testing methods A subsequent stage will consist of source testing and measuring public exposures to those substances for which data are unavailshyable The basis for regulatory action if appropriate will include the results of this program and other related reshysearch
Screening of Candidate Carcinogens
The National Institute for Occupashytional Safety and Health (NIOSH) lists 1905 chemicals which have reported neoplastigenic or carcinogenic effects and 510 which have otherwise received attention for their neoplastigenic poshytential1 The need to select five to ten materials from such a large number of potential candidates dictated that we devise a way to rapidly eliminate from further consideration the vast majority of the substances Given the paucity of published data on most of the candidate substances the screening method was designed to make best use of readilymiddot
C1111yrich1 19i9-Air Pullulion Control A~~ociatlon
1153
A- I
----- ---- ----- ---------------------available information The screening process was as follows (1) eight comshyplications of known and suspected carshycinogens were reviewed and those subshystances which were not used in Califorshynia were highly unstable in air or were very doubtfully carcinogenic were eliminated (2) after more detailed inshyformation was obtained for the reshymaining 25 substances candidates were rated by two different analytical methshyods (3) an expert panel was convened to review dossiers on the candidates and to independeitly nmk them (-i) from the eight to deven substances ranked highest by all three approaches eleven were selected for the emission identifishycation and source-testing design stages of the CAHB effort
lnitiI $cre~ng ot Potential Candidate Carcingtgens
65 compnunds middotwe ire selected from 642 industrial ur~nnic air pollutants comshypiled by MrTRE Corporation for the US Envircmnental Protection Agency (USEPA) in U1at study pollutants were scored by multiplying four explicshyitly defined rating factors annual US production fraction of production lost to the environment volaflity and toxshyicity To adapt this york to our pmpose we first selected the 114 substances listed as being carcinogeni 1~ or neoplasshytigenic Then the scores of each of these compounds under the criteria annual US production fraction of producshytion lost volatility and carcinogeshynicity were multiplied together Seshylected for further consideration were those substances which had a product score above 50 Another 15 substances listed as being carcinogenic but lacking information for one of the other rating factors were also selected This list of 65 was then compar~d with seven other lists of corcinogeris1
-middot 11 Materials comshy
mon to the reduced MITRE list and at least one of the other lists were chosen for further consideration Added as candidates were those suhstances which are regulated as occupational carcino-
gens by the Occupational Safety and Health Administration (OSHA) and certain inorganic carcinogens Finally substances were added which in our judgment should be investigated but had been eliminated at this point Exshyamples of these are bis(chloromethyl)shyether epichlorohydrin and hydrazine
Next the refined list which now contained 47 substances or chemical groups was pared further hy another rapid screening process Eliminated
were all candidates (1) whose production andor use in California was very low (under 10~ lbyr) and wns not thought likely to pose a risk to a localized popushylation (2) which are very unstable in 1ir or CH whilh should not on t lw hnsis of CUtrltbullnL cidlnre liebull considtbullnmiddotd r11rci-
1154
Tahllt I Suh tances n-iewed in letail
Candidate Substances
Arsenic Inorganic lead Ashestos Alkyl lead Benzene Maieic anhydride
nogenic The result of the initial the rating under that criterion and the screening was a list of 22 candidate mashy corresponding criterion weight The terials which is presented in Table I overall rating for pollutant i is then the
sum of the scores under all the crishyteriaRanking Candidates by AddiUve ancll
Mulllpllcatlve AigorUhms The additive approach has several virtues the main one being that it forces
Many screening or ranking systems the user to make all assumptions exshyfaH into one oft wo cat~middotgories additive plicit In the process of setting up such and multiplicative Some systems are a a ranking system new insights into the combination of the two while others problem under consideration may be combine nn ol1jective appronc 11 with g1ined Once the system is set up it is suhjlctive cvnluition of the remiddot ults 1l rdatively easy to use Where data for The L2 substances surviving th1middot initial scoring pollutants are unavailable arshyscreening were ranked independtmiddotntly by tificial scales can be constructed to the tvo approaches If the same subshy quantify subjective factors Finally the stance rated high~middot under both systems sensitivity of the results to the systems its importance to California was judged subjective aspects may be measured For to be more likely than if it had scored example one can determine the effect of highly in only one method changing criteria weights upon the final
Additive Approach In the additive pollutant ranking Similarly an appreshyapproach the user identifies one or more ciation may be gained of the significance criteria and rates each alternatiw subshy of the range of uncertainty for a particshystance against each criterion while sishy ular required data element by varying _ multaneously deciding the relative imshy rating factor values portance of the criteria Eq (1) shows its A fundamental problem with the apshymathematical formulation proach is that there is no logical basis for
adding the individual scores assignedRating for pollutant i = f W1 R11 under the criteria other than the asshy
rbullI sumption that this simulates or even
(1) improves upon the users thought proshyEach criterion or rating factor (1 ) is crss A major operational problem is assined a value for each pollutant and that of weighting the criteria A common each rating factor is weighted (b W1 ) practice is to give all critNia equal -icnrding to its importance reht veto wltgtight hut that is in itstbulllf a st1Ubullm1middotnl lhe otlwr rritnia llw sc11rtbull for 1111 aLnut the 1rcl11tive importance of Ow 1n1 1 undmiddotr nitn1 n j is the prod11middott of n1ttria
Journal of the Air Pollution Control Association
A-2
---------
Multiplicative Approach In the multiplicative approach the rating for each alternative is the product of the rat ihgs under each criterion
Hating for pollutant i --- 11Ill
UJ (~) Jbull I
A multiplicative approach can have ~omc altlvrntnges over additive onrs First in some cases the ratings can be physical parameters such as concenshytrations or volatilities there is then no need to weight the criteria and hence less controversy over subjective judgshyments Second multiplication generally provides a wider range of scores than does addition allowing clearer disshycrimination among alternatives Finally this approach provides results which are more intuitively acceptable As an exshyample of this last point suppose that exposure and harmfulness levels for candidate substance are each converted to values on aO to IO scale and that a certain substance is both extremely toxic and extremely rare An additive approach would give the compound a rating ofO + 10 = 10 which is equivalent to that of a moderately prevalent subshystance (rated say at 5) which is modshyerately harmful (rated also at =-) A multiplicative approach on the other hand would rate the first substance at 0 and the second at 25
Criteria The six criteria used in the additive and multiplicative ranking procedures are defined in Table 11 Asshysignment of values to the Rij was based upon data gathered from published litshyerature personal communications and panel discussion and has been fully documented 13
Because the purpose of this exercise was to determine the relative imporshytance of the suspected candidate carshycinogens R 1 was scaled to the most heavily used candidate substance benshyzene whose annual production and use in California is nearly 109 lb Materials with a use under 105 lbyr would be rated zero for R 1 and rejected Howevshyerbefore rejecting a substance by this criterion we considered whether its emissions could in particular circumshystances result in high exposures to a Joshycalized population
R2 takes into account the fact that the chemical industry is in continual change Substances of concern today may be phased out while the use of others may rise dramatically increasing their imshyporta nee asmiddot poll utan ts Information on developments which could likely result in a change in the growth rate was facshytored into the choice of a value for R2- As an example asbestos consumption in California has been stable in recent years However the pending phaseout of asbestos in motor vehicle friction materials will hasten the decline in asshybestos consumption hence we assigned a value of 1 for R2bull
November 1979 Volume 29 No 11
Ideally one cou1 1 use pouu1c11u
sion as a criterion However in this case emissions were to l e estimated in detail only for the five t1 ten carcinogens fishynally ~elected Thmiddotrtfor(bull a 11w1st1n of tt II iim pol cint ial vus 1~1middotd 1atbulld llpon
knowledge of the s11bstances manufacshyture and USP for R The higlwst rating went to substanc middots which are v idely used especially in consumer products A slightly lower rating went to subshystances which an~ routinely emitted from industrial processes during proshyduction and use Some materials are employed in such a way that emissions are quite low even though tight emission control may not be required by law Materials in this cntegory were assigned a value of two for H3bull Substances which under federal or st ate regulations may not be discharged to the exterior envishyronment but which could be dischJrged by accident received the lowest rating
Each candidate was evaluated (n the basis of its 1ropen-ity to decompbullgtSe in ambient air Materials with half-lives greater than eight hours were considered moderately to highly stable and rated five for R4 bull Low to moderate stability was assigned to substances with halfshylives between zero and eight hours Compounds known to exist in air for only a few minutes would be rated zero
Ta hie II Dcfinitinns of the criteria used
R1 Prbullbullsent use in California
100 of max (109 lbyr) 5 10 1)f max (108 lbyr) 4
1 of max (107 lbvr) 3 01 of max (106 lbyr) 2
001 of max (105 lbyr) l lt001 of max (lt105 lbyr) O
R2 Growth in California use
+ 20 5 +Oolti to +20 4 Positive growth to 0 3 Stahle or unknown 2 DPcline 1 B1middotin~ phased out 0
R Emision potential
Widespread use in ltonsumer products 5 R1middotlatively p1or con rot over emissions 4 Rdatively good control over emissions 2 Tihtly c1int rolled 1
R4 8tabilit~- in ambient Air4 tvlnderale t1 high st 1bility (l 1 middot2 gt 8 hr) 5 Lw to modmiddotrate s1 bility (t1 l -0-8 hr) 3 Unstahle(t2~febull minutes) 0
R Dispcision Potential
Emitted lar~ely as apor or iine 5 particulnte
Emit led largely as rnarse particulate
Rlt Evidence f Carcinu~enici ty
Known or suspected human carcinogen 5 Known mammalian carcinogtmiddotn 4 ~usptctcd 111ammalan carcino~en or 3
known m mmalin muta~tmiddotn Ames ltmiddotst p sitive 2 Prpoundgtcursor 01 co-car inogen
a t11 is the half-liftmiddot
A-3
tion state or anion associations may change in the atmosphere metals do not degrade and were considered stable Asbestos is likewise itahlc Many of the clcc-ompo~ilion rtbullnctions oi orinnic molecules are mediated by light Such substances if released at night would have several hours to disperse in the surrounding area
A rapid way of assessing the relative potential of different substances to spread from a release point is to note their physical state under normal amshybient conditions Accordingly we scored materials emitted as vapors or fine particulates the highest for R5 and coarse particulates the lowest Intershymediate values are possible for varying amounts of fine and coarse particulate emissions from the same source or from different sources
There is as yet no widely agreed upon measure of the relative potenciemiddots of carcinogens although some ranking systems have been proposed14 Extrapshyolating data from in vitro techniques such as the Ames bacterial mutagenicity test and from laboratory animal studies to humans is problematic Therefore a less quantitative measure of the carcishynogenic potential of each candidate substance was used The candidates receiving the highest scores for Rs would be those for which there is strong evishydence of carcinogenesis in humans Exshyamples are asbestos which is implicated in mesothelioma vinyl chloride which has been identified as the agent of liver cancer in exposed workers and bisshy(chloroinethyl)ether shown by epideshymiological studies to cause lung cancer in resin workers The next highest rated substances are those for which human carcinogenicity is unknown but which have produced cancer in one or more mammalian species in laboratory tests Next are those which have not been shown to be carcinogens but which have proven to be mutagenic in test animals Substances for which the only knowlshyedge of carcinogenic potential is a posishytive Ames test (producing mutations in histidine-requiring strains of Salmoshynella) are rated 2 Finally substances which are implicated only as precursors or co-carcinogens would be rated lowest
Substances unequivocally associated with carcinogenesis were considered as carcinogens in this study Conditions of emission and exposure including the presence of co-carcinogens were facshytored into the evaluation of each candishydate where possible Carcinogenic subshystances derived from the metabolism of a precursor were considered as carcinoshygens However ubiquitous substances which have been hypothesized to be precursors (eg secondary amines nishytrous acid and nitric oxide which combine under certain circumstances to
1155
form N-nitrosoamines) were not conshysidered because of uncertainties in the importance of their link to the carcinoshygenic compound and the practical conshysiderations demanded by the scope of the study
It was beyond the scope of this study to jud~e the validity and interpietation of the experimental ad epidemiological evidence upon which the carcinogenicity of candidate substances has been esshytablished We accepted the conclusions about carcinogenicity drawn by the Inshyternational Agency for Research on Cancer or the National Cancer Institute and did not consider dosage or route of administratioll of tested substances However considerations of test validity did enter into the subjective evaluation by the panel of experts
Weight lr1 the sdditive ranking scheme each rating criterion is weighted according to its limportance relative to the other criteria Little precedent exists for assigning these weights In our judgment and as generally agreed by the panel of experts (see below) R 1 RJ and R6 are more important than the other criteria Evidence of carcinogeshynicity was considered to be the most important criterion of aB so W 15 was assigned a valtze of 3 W 1 aid W 3 were set at 2 and W W-i and W5 were set at 1 In order to discern the potential senshysitivity of the rankings to the weight assignments the candidates were also ranked using equaily weighted crishyteria
Ranking Candidates by Panel o~ Experts
Anine-member panel of experienced governmental industrial and academic scientists whose disciplines include~ -organic and physical chemistry indusshytrial hygiene toxicology epidemiology and iegulatory control of toxic sub~ stances was convened to provide addishytional data for our ranking algorithms
to discuss our candidate --ubstanc -sand rejections to suggest pos-ibie ne suushystances for consiceration and Lt rank the candidates indepen11mtly ( f our own ranking Tvo v-ee ilts he fore the meeting pa-d uember-s were jven one-to three-page dossiers on eac I canshydidaie substance
At th~ start of the meeting- before any group discussion the pand was asked to rate each candidate suhtance with a score from Oto 5 Next cmiddotach candidate was discussed at length e provided ~n ovrview and summarized critical issues identifi2d up to that Pbullgtnnt Through materials brought to thtmiddot meetin and their persoril experience panel memshyber Veimiddote able to provide much useful information on the candidates and adshyditional insight in to our ~a ting criteria
At the encl of the two-d1y sc~sion the pa-iel ugain -ated the cai 1didates
Flnal SelediDn
Tabe III show- the highest-scoring substance a~ det(rraine1 by the 1ddishytive and mlltiplictive approaches and by th~ panel in ~e additive approach 21 single r1nking as obuined by avershyaging the two ran kings resulting from using equal and uneguai weights The ran kings ltif most candidi tes were unafshyfected hut carbo11 tetralt hloride chloshyroform chromium and inorganic lead changed more tban thmiddotee positions_ Because uncertainties in the data base predude imputing signifiance to small differences in the Lnal ordering the lists in Table III are prtmiddotsented in alphabetishycal order However it is of interest to point out that benzene consistently ranked hihest
Becausimiddot some cindidates had equal rating scores we ltould nut choose exshyactly ten candidat ~s from the additive and multiplicative rankings Instead the
top nine and eleven were selected from the two exercises respectively We also considered the ten substances scored highest hy the panel at the end of the session The final consensus selection cornsistt-d of the 11 candidate substances appearing on at least two of the three lisL- For the substances included in the cun~ensus ranking a baseline emissions inventory is being conducted and a source testing program is being deshysigned
1ejected Substances
Some comments about certain subshySiF1ces not appearing on the final list are in order inasmuch as they include known carcinogens and compounds which hae received considerable atshyllttion in recent years as occupational c11c111ogens
rouisionally Rejected Substances Appended to the consensus ranking (T1ble III) were vinyl chloride gasoline and engine exhausts tobacco smoke and pesticides No further action by the CRB is recommended at this time for vinyl chloride because it is already subject tu the USEPA emissions stanshydard a CARB ambient air quality strndard and an OSHA standard
Gasoline and tobacco smoke were appcndlmiddotd to this list because each ocshycurs very widely and contains several of the candidate substances reviewed in this study some of which are in the final listing For example gasoline contains benzene ethylene dibromide ethylene dilmiddothloride and alkyl lead compounds the last three being in leaded grades only Both gasoline and diesel combusshytion products include PAHs Tobacco smoke contains among other neoplasshytigenic substances nitrosamines PAHs nickel arsenic cadmium and other heavv metals Manv individuals are inshyvolu~tarily and i~ many situations
Table HI Hi~hest ranked candidates from each rnnkin~ methodbull
Highly ranked but no inventory
l I ighesl consensus recommended Additive Multiplicativ(~ Pan I of eqwrts rnnkmg at this time
1156 Journal of the Air Pollution Control Association
A-4
lirtualy unavoidably exposed to t11-k1ltco smoke Because the sources of gasoline its combustion products and tobacco smoke emissions are well known no specific action was recomshymended for these materials during the tbullmissions innnlory and -ource testing design stages of the present study It was considered importint ho-vemiddoter to draw attrntion to the general publics exposhysure tot hrse substances PrsticidltS are listed for the same reason thou~h ad(bullmiddot tailed examination of pesticides w1s beyond the scope of this study Many pesticides are widely used and some of them are known to be carcinogenic
Other Rejected Substances Acryshylonitrile and vinylidene chloride were placed in a provisionally rejected group because of the panels suspicion that imports of these compounds to Cnlifornia from Japan may be appreshyciable yet are hard to substantiate Should such imports be verified in the future these two compounds would take on greater importance Formaldehyde was also provisionally rejected because the preponderance of evidence indicates that it is not carcinogenic and that bisshy(chloromethyl)ether is not formed from formaldehyde in appreciable quintities in industrial environments 15 The ocshycupationally controlled carcinogens ethyleneimine and beta-propiolactone were rejected in part because of their reactivity in air At the time of this study DBCP a pesticide was no longer heing middotproduced in California and was therefore rejected from further considshyeration
Occupational Regulations and Community Exposure
A question of interest to the CARB was whether the regulation of acknowlshyedged occupational carcinogens adshyverselv affects the ambient air outside the v~rkplace Our general findings can be illustrated by the example of asshybestos
Asbestos is a very widely used mateshyrial for which no ambient air standard exists Concentrations in the workplace are limited to an eight-hour timtmiddotshyweighted average concentration of two fiberscm=1 of air and a ceiling concenshytration of ten fihcrscm 1bull In meeting this standard exhausting air containin~ asshybestos to the ambient air is not reshystricted except by the USEPAs reshyquirement that there shall he no visible emissions containing asbestos piut ides from s11eh foci lit i1bulls (bullxd11din~ hrak(bull shops 1 Considning that undc-r certain conditions 101 asbestos fiherscm1 could ht Pmitted without being visihl( 11 this standard mamiddot allow considerable asshybestos emissicns It is unlikely however that emissions would actually approach these levels First since the OSHA standard cannot generally be achieved
November 1979 Volume 29 No 11
by ventilation n ine the generation or asbestos partid in the workplace must be greatly reltlu ed Second the air is usually filtered 11gt prevent recirculation of asbestos to 1 workplace Asbestos waste must bed 1sposed of in sealed imshypermeable bag~ or containers Thus under current onupational regulations the amhient air 1enerally appCars to be afford(bulld greatn protection than it would without s11ch regulations 1
Conclusion
The screenin and ranilting methodshyology presented in this paper proved to be a feasible approach to establishing a priority list of Jirborne carcinc)gens in California We ftel that it is an efficient means of focu~ing further efforts on emissi()ns inventories source testing and ambient measurement for it not only identifie all the carcinogens of potential concern but it also permits the state regulatory agency to direct its reshysearch resourCtmiddots toward those subshystances of partilmiddotular interest within its jurisdiction
References l National Elllission Standards for Hazshy
ardous Air P llutants A Compilation as of April 1 19 3 PEDCO Environmenshytal Inc for US Environmental Proshytection Ag1middotncy EPA3401-78008 (NTIS No I B 288 205) 19i6
S Gray Ne Jersey Department of Enshyvironmental l 1rotection Trenton NJ private com111unication 1979
1 H E Chrislllsen E J Fairchild and R J Lewis Sr middotmiddotSuspected Carcinogens A Suhfile of th1middot NIOSH Regii-try of Toxic Effects of Chemical Substances 2nd Ed National Institute for Occupational Safety and HPalth NIOSH i7-149
4 B 8 Fuller J Hushon 1 Kornreich R Ouellette L Thomas and P Valker Scoring of rganic air pollut1nts for US Environrnental Protection Agency Mitre Corpt ration T1middotchnical Report MTR-6248 I 9i6
S F R Lozan11 Toxicants Selected for Priority En- 1ronmental Assessment corresponde11ce with EPA Region VI Administrat r 19i7
1gt Survey of inyl Chloride Levels in the Vicinity of Keysor-Century Saugus California US Environmental Proshytection AJet middoty National Enforcement lnvestiiatim Center Denn-rand San Francisco E A-n02-77-017a 19i8
0 Bardin l middotstimon~middot in Hearings on Helationshi1 Bet wfen Cancer and the Environmei 1 US llltuse of Hepreshysentatives bull th Congress 2nd Session Committee (In Interstate and Foreicn Commerce ~crial No 94-141 1976 pp 7~i0
middot T H Ma11middoth Carcinogens in the workplace lwrE to start d1middotaning up Sc-imiddotrmmiddot l17 1-1110) lfi8 (197)
I CH ar1111 Classif1r11t1on of the Exshyposure Haza d Prtbullstgtnled Ii~middot Thirt~middot-Six Clwmirals 011 I llixllmmiddots lternorandum from Clllm1middot I Offi(middote of thtgt Assmmiddotiate lgt1rector for 1 middotarcinoltnesis Division of Cancer Cau~ l and Prevention lo SushyJwrmiddoti-or~middot ( 0 Auditor Gtmiddotneral Acshycounting Oft lmiddote19ii1
I0 Chemicals ith Sufficient Evidence of ( arcinogenit ty in Experimental Ani-
A-5
ma1s Jnlernauona1 Agency 1or 1c-curcn on Cancer IAHC Working Group Reshyport Lyon France 1978
11 Threshold Limit Values for Chemical Substances and Physical Agents in the Workroom Environment with Intended Changes for 1977 American Conference of Governmental Industrial Hygienists 1977
12 M H Rogozen D 1 Hausknecht and R A Zisk ind 1976 A Methodology for Ranking Trace Elements in Fossil Fuels According to Their Potential Health Impact by Science Applications Inc for Electric Power Research Institute 1976
1t L Margler R Ziskind M Rogozen and M Axelrod An Inventory of Carcinoshygenic Substances Released into the Amshybient Air of California Vol I Final Reshyport Screening and Identification of Carcinogens of Greatest Concern by Science Applications Inc for California Air Resources Board 1979
14 T H Maugh Estimating potency of carcinogens is an inexact science Science 202 (4363) 38(1978)
15 C C Yao and G C Miller Research Study on ais(Chloromethyl)Ether Forshymation and Detection in Selected Work Environments by the Bendix Corposhyration for National Institute for Occushypational Safety and Health Cincinnati Ohio Contract No 210-75-0056 1979
16 Occupational Cancer Control Unit Calshyifornia Department of Industrial Relashytions Los Angeles California private communications with staff members 19i9
This paper was developed while Dr Margler was a Staff SriPntist in the Energy-Environment Systems Division of Science Applications Inc (SAIi 1801 Avenue of the Stars Suite 1205 Los Angeles CA 90067 and Mr Reynolds was a Project Manager with the California Air Reshysources Board Sacramento CA 95812 Dr ltar~ler is currently emshyployed as DirEgtrtnr of Bnvironmental Scitbulllltcs for WinzlN nnd Kelly Conshysultini EnintNi t1 Third Stretbullt EurekH CA 9gtgt01 and Mr Heynolds is Air Pollution Control Diretbulltor for the Lake County Air Pollution Conshytrol District 55 N Forhes Street Lakeport CA 95-13 Dr Rogozpn is a Staff Scientist and Dr Ziskind is Deputy Manager of the Energy-Enshyvironment Systems Division of SAi
1157
Appendix f~
_l)ispobullition of Miscellaneous Sites
Gould Inc Ver_n)_~ St~condary Lead S~_n_elter
The emissions of arsenic from the four large secondary lead smelt~rs
in Cali forni) J1~1 ~ bull~tiilnted in the program This estimate was based upon a
uniform fugitive emission f1cJr nt WPll supported by measurement inforriation
Therefore source monitoring was r2cq11)nded 1nd Gould Inc Vernon being
~he largest was singled out It was however proposed to monitor both Gouli
and RSR Corp (Quemetco) in the City of Industry since analysis of the plants
revealed significant differences in plant equipment and engineering The
latter being more typical of 1 nodern facility
As a r2sult Jf pr-~-middott -iiscussions and plant inspections we became
aware that Gould was actively in the process 1)f constructing a new facility
which would completely replace the existing one We have monitored progress
on the new site and concluded it would he inappropriate to utilize program
resources to conduct field tests at Gould ~rr1issi 1rns from the new Gould
faci 1ity wi 11 be bas~rl 11p11) test results from RSR
PG ampE - Pittsburg PG ampE - Salinas a~d So Cal Edison - Long Beach Oil Fuel Power Plants
It was app ro p ri ate tJ cons i dP r the (~mission of arsenic from power
plants during the initial study stage inc~ imiddotrace quantities in the fuel oil
are known to be ernitted Because of the population distribution in tile
vicinity of three plants and some 11realistically conservative estimat2s of
erlissiJn c-1ctors tht~ facilities appearc~d among the top seventeen stationary
sources of potential c1Jn-r-n HovJ~ver as a result of a reexamination of
emission data it was concluded that an error ~dj ~een made in the material
balance of arsenic - resulting in tn emission factor equivalent of a 300
release ~io lit~rature was found to justify 1reater than a 30 transfer
function Thus we estimated at the outside th~ emission factor should be
reduced from 013 lb per 1000 lb to approximately 0013 lb and resulting
ars9nic 2missions for the entire state w~r~ cJ11servatively estimated to be
1760 lbs (from 17600) divided amongst 311 the states power plants Clearly
then the four secondary lPad smelters estimat~ of S9410 lbs of arsenic(
B-1
emissions per year makes consideration of poer pl int ernissio1s if ffsenic a
very low priority
In a literdture review by SAI - Methodokgy __f_or Ranking Trace
Elements in Fossil Fuels Accordi~9__t_~_their Potential Health Iripact (Roqozen
1976) - emissions estimates f0r 15 trac~ substances were researched for c11l
and fuel oil pomiddot1er plant c 1J1v---r-sion Source content combustion pr-)~~
transfer functions and contrcl methodology were co1~idered to develop e~ission
factors The output to input ~atio computed and utilized n -~i-i~ study for
arsenic was 002 to 03 (ie tra1sfer function betw~en 2 11d JO~) reflecti~q
the wide variety of fuels processses and controls nationwiJe The upper end
reflecti1g both high arsenic coals and poor e11iss10n contro 1 devices Clearly
neither of those conditions accurtely apply to th2 thr~~ si~~~ r1nrl thus
substantiate the rlecision not to perform emissions neasur-i-~n~s
Calaveras Asbestos Company - Copperopo l middotis Asbestos Mini n(i and Mi 11 i nq
In the past (late sixties and early seventies) ~h2 ~3laveras
Asbestos Company came under heavy criti-is1 1fter imiddot1sp~ction rieasurements
revealed serious problems However significant reduction of e1nissi0ns have
occured prompted by NESHAP regulation and occupcttional standards SAI
inspected the site in f)~c~1b-2r 1979 under an EPA contract An missions
inventory was published under that work (Ztskind 1980) and the bottom line
conclusion is that currently no significant ernissioi1 are gteing released as a
r~sult of hl-isting which would reach the publ le i 1le 0pen pit is some 900
feet deep with blasting at the bottom 0ver 80 percent of emissions in the
CARB Emission Inventory System were attributej to pit blasting In fact at
its current depth blasted 1aterial does not reach the mine surface with tile
explosive implacement used Additionally the site is more remote than might
be realized The ~ituation is vastly different today than in the past and
currently attention should be paid to the issues of occupational exposure and
breakdown of ventilation syste111 controls in t1e inilling operationsmiddot He
recorrrnended that no furtw~ corsideration be Jiven to this site for the
purposes of tl1is w)JJil ie identification )f silt~middot1ificant releases vhich
might he responsible for causi1J middot1 fmiddot middotgtbullgt middot 1= 1(11 1bull11 tinn Pxposure
8-2
Vctri0J5 Refineries
It was est=tblished early in the anrilysis prmiddotJr111 V1at gtenzene
emissions from refinery operations coulrl in th~ 1guregate constitute a
significrnt source Since there ilre a large rrnrnber 0f refineri~s in
California (46 in Los Angeles County itself -it t(~ i 1~ middot)f r-1ltariination) with a
wide variety of types sizes ages etc their eva l uat 11) 1 i =r~sent -3~ i)7
monumental task It was noted immediately t 13t ~~it hi n the total scope of the
study the design and conduct )f c r~f i r1~ y testing program which would develop
a complete benzene emissions in1~ntory for 01e site was impractical let alone
to d1aracteri ze emissions fro111 three ie those listed among the 17 rnost
s i g n if i ca n t potent i a l st a t i on a r y sou r c e em it 1-_ l rs bull
The three refineries were singled out for special attention in the
previol1s phase ~ecause they uriiquely had components which process materials
containing 10 or more per~nt hy middot1-bulli1 11t )-~112~ne (46 FR 1165 Jan 5 1981 pg
1491 ) Est i mates by E P A ( F e d bull r l Rr-~ g i s t ~ r 1q81 ) i n d i cate that 90 p e r cent o r
more of the total benzene fu 1Jitive emissions arise from such components
Therefore attention is appropriately focused on the three henzene production
consumption refineries Chevron (Richmond) Arco (Crson) and Chevron (El
Segundo) SAI staff considered the possibliJ )f -1 testing program at one of
these refineries and cone l url 1middotd it would not 1)e cost-effect i v2 for o number of
reasons
California is a minor producer and consumer of benzene with
approximately 15 of the 114 hill ion pounds produced and
consumed nation1lly in 1977 This can 1)e riY1pared with the fact
that California has approximately 10 of th~ ~ations population
Benzene exposur~ to the generil nJmiddot1lation has been partitioned
among the various sources Aproximately 90 of exposure is
estimated (SRI 1978) to be caused hy gasoline distribution
activities and 1ehiclEmiddot emissions in 1 ir~an areas with nearly all
of the balance )Y benene hrndl ing operations (refineries and
chemical plants) In the case middotlf California this percent1g~ Jill
be even JT1ore di middotpruJor-tionate b~cause of its greater s1are of the
population and ~esser share of the henzene ha~dling Furthermore
since the three refin 1bullry cites ire heavily urbanized no new rural
population cegmmiddotnt is beinq exposed
B-3
Californias most heavily urbanized Air Quality Management
Di st r i cts have a l ready adopted t en zen e f ugit i v e em i s s i on
staridards comparable and vith S( 1ine features VJr stringent than
the proposed national (bullmission lttandarrl (4f-i FR 1165 Jan 5
1981) Thus the concmiddotlusion rleri1ed above (Le refinery emissions
are secondary cause of exposure to the urb~n microopu~ation) is
further reinforced since H is irojected that re1eases from
components in benzene service (gt10 benzene) will be reduced by
73 by the proposed F~deral Standards T~~ ~istrict rules (eg
South Coast Air Quality Management District 456 and 4fi6l) are
not restricted to berz~ne per senor to only components in
benzene service The rjLs 1lso i-lclude flanges in addition to
the components ca i led uut in the propose nat i 01a l standard
The EPA estimated that if the proposed emission standard were adopted the
1~aximum annual benzene concertr0Vioi1 -~01middot 1 plant would be 36 ppb at a
distance of 0 1 kilomeL~~ avay Cor1paring this ground level concentration
with the general urban background i1 California of 19 ppb shows the latt2r t1
dominate In recognition of the secondary im0ortance of these thr~~ i~~s i~
was decided to utilize program resources to d~velop field data at other sites
1111here little emissions informati1)i1 1as cll-1aila1)le
8-4
APPENlJIX C
US tn v 1 ronrnen ta I Protection Agency
middotJc k
METHOD 108 - UETERMINATION OF PA~TICULATE ANlJ GASEOUS ARSENIC EMISSION
FRUM NONFE~RUUS SMELTERS
1 Applicaoil ity and Principle
11 Apmicroli1ability This rnetnod amicropliltS to tt1e determination of
i nor~an i c d rsen i c ( s) e1n is sions trom non f errou Sm(~ I ters dnd other sources as
smicroecified in the reuulations
12 Principle Particulate and gaseous As emissions are withdrawn
isokinetically from the source and collected on a glass mat filter and in
water The collected As is ther1 analyzed by means of atomic aosorption
spectrophotometry
2 Apparatus
21 Sa111pliny Train A schematic of tt1e sampling train is shown in
fiyure 421 it is similar to tt1e Method 5 train of 40 CRF 60 Appendix A
Tne sampliny trdin consists of tt1e followiny components
Note H1i s and a11 subse~uent nferences to ottler methods refer to the Methods in 40 CFflt 60 Appendix I
3 Annual Book of AST11 Stdndarctbull Part 31 Water Atmospheric
Analysis American Society for T~sting and 1-1aterials Philadelphia PA
1974 micromicro 40-4~
C-15
APPENDIX D
~f-~__d_y_ e s f o r Pr o c e s s i ng a n d An a l y s i s of PA H i n
Co~e Oven Emissions from Kaiser Steel
Global Geochemistry Corporation
10 Samples expected
l 1 Connective tubes between section welded to oven port cover and remainder of sampling system - These are to be stainless steel tubing disconnected and capped off for each sample They will be about 10-15 long
1 2 Impingers in series first two water-filled third dry - The center tubes will be removed and wrapped in clean foil and the impingers stoppered as is for shipping One set for each sample Protect from light during shipping
20 Extraction -procedure
Sample fractions are to be covered with foil or in opaque containers with inert gas flush during storage in freezer Solvents used in extraction will be BampJ glass distilled pesticide grade low residue) Working bench area will be screened with amber plastic and light bulbs
2 l Connective tubes will be rinsed with dichloromethane by partial filling capping tilting several times and draining into a beaker This will be repeated until the washings are colorless but at least three times
22 Impinger contents will be transferred to a separatory funnel The impingers and center tubes will be rinsed inlo the funnel using a measured amount of dichloroshymethane in portions The funnel is shaken well and the extract drawn The rinsing and shaking is repeated with two more portions of solvent The extracts and washshyings are combined over Na 2 S0 4 bull
A separate portion of the water used for impingers is extracted in the same way
For each liter of water 100 ml dichloromethane will be used per extraction ie 300 ml total
Internal standard is added to the combined exshytracts at this p0int
D-1
_rmiddot
30
40
Concentration of extracts
Bulk solvent is removed on a rotary evaporator in a water bath at not over 40deg stopp~ng short of dryness The res~dual volume should be just less than the f i n a l v o l ti 11 e o wl1 i ch the s o l uti on i s to be d i l u t e d (accurately) The object is a final solution with roughly five mgml solids concentration in dichloromethane After the vol w1 e ~ s made up to the mar k i n a vol umet r i c flask an aliquot is removed and evaporated on a tared disk in a nitrogen stream and weighed to determine yields The remaining solution is stored in a ial or ampoule sealed with a Teflon-lined septum A split for SAI QA may be made at this point (see Section 43 also)
Gravity column c1eanup chromatogram
4 1 Into a 10 min 1 d column with co2rse sinter and Teflon stopcoc~ 6 g 120 mesh silica gel (activated at 120deg) 1s sluiry packed in pentane cooling the column by evaporation from a tissue wrapped around it and wet with acetone) This minimizes vapor gaps in the column A cap of 3g sodium sulshyfate is added to the top The column must not dry out before the chromatogram is complete
42 A sample of the extract concentrate is taken This is to be based on experience of expected PAH conshytents but initially would be 5-10 mg This is exchanged with pentane by adding 10 ml portions of pentane and evaporating over about 100 mg silica to small volume Three to five portions of pentane are used The final residue is rinsed onto the column with pentane
4 3 The chromatogram is developed with four solvent steps
l 25 ml pentane 2 l 0 ml 2 0 ~~ dichloromethane in pentane
II II ii II3 l 0 ml 50 u II II4 l 0 ml 5 0 ~~ methanol
The column volume is approximately 8 cc This is allowed for while collecting fractions During the first chromatogram the graduated cylinder collector is observed for visible solvent changes and a more precise estimate of the column volume
T h e f o u r f r a c t i o n ~ a r e e v a p o r a t e d i n a n 1 t r o g e n stream at not oveimiddot 40 to l 0 ml then stored under nitrogen in the freezer The principal constituents
0-2
of the fractions will be
l Aliphatics 2 Lower molecular weight PAH
11 113 Higher 11
4 Polar materials
Fractions 2 and 3 are to be analyzed by HPLC the others retained for future interest or in case of an incomplete or defective separation Splits for SAI QA may also be made from these fractions
50 HPLC Analysis
5 l The column is a reverse phase partition column C18 bonded to microparticle silica 4 x 250 mm The solvent system is a gradient from 6040 acetonitrile water to 100 acetonitrile initial slope setting l 5min The flow rate is one mlmin the injecshytion sample volume one to ten microl depending on conshycentration Detectors are UV absorbance and fluorescence in series The absorbance is set routinely at 296 nm but may be varied in certain instances to aid in identification of peaks The fluorescence filtars are narrow pass excitation at 360 nm and cut-off emission above ~10 nm Both d e t e c to rs a re rec o rd e d s i 1i ul ta n e o u s l y
52 Reference calibration is based initially on indishyvidual chromatograms of single compounds and roushytinely interspersed chromatograms of multicomponent reference mixtures The proposed calibration stanshydards are the following at minimum
Phenathrene Pyrene Chrysene Perylene Benzo(a)pyrene Benzo (ghi)pcrylene I n d e n o ( l 2 ~ c d ) p y r e n e Coronene
The internal standards (added in Section 22) are 910-Dimethylanthracene
9-Phenylanthracene 9 10-Diphenylanthracene
60 Peak Identification
Significant peaks not recognizable as present in the reference standard will be investigated by alternate methods for identification
D-3
6 bull I Stopped-flow examination of UV absorbance for maxima
by manual wavelength varmiddotiation allows checking against known reference spectra (published or measured) This is usually sufficient confirm a component already suspected to be present
62 Collection of the fraction contairino the unknown peak allows later determination o~ t~e full absorbance andor the fluorescence spectrum for comparison with reference compounds
63 If not already run on GCMS the sample can be sent to SAI for that purpose possibly allowing the comshyponent to be identified Since the HPLC elution order is not the same as the GC retention sequence it may be necessary in affibiguous cases to collect the HPLC fraction and run it separately by GCMS
The~e may be numerous minor constituents not identified The decision on how many of these to track down and identify will be made in consultation with CARB adv~sors a~d with SAI since at some point the returns ute not worth further expense
_64 IdeAtified peaks are quantified by comparing their areas to the areas of the known concentrations of reference compounds A recovery factor based on the added internal standard nearest in elution order to the sample component is then applied the amount of internal standard originally added to the sample is divided by the amount found in the HPLC analysis This ratio is the multiplier used to correct for losses in concentration and chromatography steps
70 Quality Assurance
71 Transmittal of sample ~plits to SAI provides for external QA This should be done with 15 (or one in seven) of all samples
72 Blank water extracts are carried through the same analytical procedure (10 of total)
73 The cleanup and HPLC analysis are duplicated on 15 (or one in seven) of all extracts
74 The multicomponent reference is run first every day then elution volumes and peak area for each comshyponent are plotted on a control chdrt to pinpoint developing problems At least evetmiddoty tenth chromato-gram will be this standard It wi 11 also be rerun after any service on the instrument~ such as a column change
D-4
APPENDIX E
Gas chromatograph 1mass spcmiddotctromlt~try protocols used by SAI Trace
Organic Chemistry Laborator1
Aliquots of sampl=s received from Global Geochemistry Corporation
were analyzed by combined c11s chromatography mass spectrometry (GCMS)
Liquid extracts were spiked with an internal standard compound (deuteriumshy
labeled phenanthrene) priot to analysis by gas chromatography mass spectroshy
metrymiddot A Finnigan model 4( 21 Quadrupole GCMS system including an Incas data
system was used for all analysis Liquid samples (1 microliter) were injected
into a 30-miter x 025-mm ID SE-54 fused silica open tubular 9as chromatoshy
graphy column (JampW Scientific) A programmed GC oven rate of a 4 minute hold
at 30degc followed by a 40degCmin temerature increase to 160degc and a s0 cmin
temperature increase to 270degc was used for sample analysis The detector
system was a quadrupole mass spectrometer operated in the electron ionization
mode at 70eV The quadrupole mass filter was scanned from 35 to 475 atomic
mass units in 095 seconds A hold time of 005 seconds was used to stabilize
the electronics prior tote next scan Data were continuously acquired using
a Finnigan Incas Data system which writes time intensity mass spectral data
to a magnetic disc The mass spectrometer was operated under computer control
during acquisition Figures 33-7 through 33-9 show the total ionization
chromatograms for each of the three samples that were analyzed
Following analyses the data files were searched for priority
pollutant polynuclear aromatic hydrocarbons (PNAs) In addition to the
quantitative data reduction used for priority pollutant PNAs a survey analysis
was performed on all significant GC peaks (ie signal to noise of greater
than 101) The survey analysis was based upon a computerized library search of
individual background subtracted peak maxima scans for qualifying GC peaks
The library used for qualitative identification AJas a combined NIHEPA
National Bureau of Standards library 1hich contains approximately 26000
entries
Tables 33-11 through 3-13 present the results of this survey analysis As expected a series of al~yl Slbstituted PNAs were observed in addition to heterocyc l i c compounds con ta i ni n~ nitrogen iind sulfur moieties Further
investigation of the mass spectre of the nitrogen substituted identifications
confirmed the presence of benzonitrile isoquinoline andor quinoline acridine
carbazole and alkyl substibted ~yridines Ion specific searches for N-nitroso
substituted amines were neg1tive
E-1
APPENDIX F
Detailed Fiber and Mass Concentration Data - Johns Manville
=-- -i l L - [- - =-- -= f- - =-- -l J middot - -=-
i-- L =-- - ~ ~ L ~ ~ =--shy r---- --
- 7 l -=----7- -- - - - -- =-- - - ~ =-- - 1- -~- l
g~~~~ r_-middot~ ~-i [-- L ~ L Ldeg
middot--~ l - --------=-- r- L r- r-
~~~ =-- -- r- = - -r---
a- - - - - _ - -I~=-- L -- l 7 -~--
gtZlt _
- _(_ -lt lt =- = ZJZ~
-
__ _
~ -- __ middot- -~ ~~~ = L ~f~ Z ~ C
middot -
__ ~ lt lt =- _~~ijE
gt = z lt gt
rbullshy__
F-2
SI I HMllLfc J H 1 I OBr I lll-H ANI) NS~ coHcrrrllliTJ ON CUfllllTI VImiddot SIJMMIIY
TIITIL JnI~ SfiNtlEO TOTtl Ill- or FI ITETI TOTI Illlt ~111-11 11 ltllOT FIIACT I Ori 1111 OF ~middotWI MUllllNJbull lOTI 01IIMImiddot 01- JII TOTI FI lilmiddott~ UgtIJH nD
= t nwo 00 so MI CllONH = t I 1JOO SO (M = II 900 SO CM t bull (0000 l1T~
I I JO SO CM 70000 CUil JC Ml-TEHS
= ~lt O V l lWfl~
Manville D-2 Baghouse
723 347 -633
CIIHYSOTI LE JfllII I nou J11BICUOUS NO PTTFllN NON-SBI-STOS tLL FIIIEIIS
FI 111-11 COllflT ltFIIWll-
ITllClmiddotrir r lllrf
2gt()
) ( I f S bull1 bullbull
I 0
l ~-Irbullmiddot
00
fl JI)(~
00
0 OUU~~
00
0 OflO
~60
1110 000~~
FI IWII ( ()[IClmiddotNTHT I ON ltI- I mn ll11 ~41 GI OF FI LTEH) (FIBlmiddotn~ ll-11 CIJB f-11bullTl-H OF IH)
I (2fi71bull+0G (aG(l-+05
( gt027 F+04 OfmiddotHH
OOOOOE+OO OOOOOEtOO
OOOOOF+00 o uuoul~middotH)O
0 O(OOE+OO o 0000I+oo
J bull f 1)07 E+O( fi r71HE+OG
I
w
TcfLI ~~lllllmiddotCE 1lllmiddot lt~O Ctl lFII ~~O UI OF FIiTEi) t q UI 11-11 Cll II fllmiddotTlbullll OF J I H)
I 590)1+07 6 t 90llbull-HH
( HMOE+O 2 40411bull-1 o
0 OIHHlF+OO o 00001+oo
0 OUO(I 1 00 OOOOOJi+00
() 10110~ Ji)
OOOOOE+OO 1 lHHllJ i 00
0 00001middotgt1 00
M ~~~ middotor-ic Irrrn TI or
( PGRAMS 1bull111 ~c UI OF F J LTEH) (PGRAMS 1111 UJI illI Lil ()j Ill)
IUitTII Ml-Ml ltMI tJIOfl~~) ~Tlgt lgtEV
Ml- H Lot c1middoton nr COlmiddotF I II
I) I MWTEll Ml-N ltMI c11or1~~) STI) tHV
MEtl IOC cImiddotor1 r1N CCWF V11
AIt lbullI rt 1bulli I~ (Sil MIC) STIgt 1gt1-V
Ml- N 101 c1-ori r1u COlbulltmiddot V1ll
I 0 I ~if1+07 bull tJf I (11bull-HH
r 1gta 1 1) bull 2 (10()
0 lt700 l 1Hi4~ 2G44
() ~( 1 027B
- l bull 6 I 1bull1middot 0 J l)J 09J~fgt
1a 1071 1~ 09rn
0 2(jlt)fi l ~HVil 4040fi
~ HltF+O~ U (H~61bull+o I
o rooo 0()()00
-0GJOB 0 (000 0 (000
OOfiOO 00000
- )()57
OOGOO 00000
0 0 1)02 00000
- I J 0 0 0 1W 00000
00000 0 ()000 00000 00000 00000
00000 00000 00000 0 (WHO 00000
0()000 0()()()()
0 ()()()) 00000 00000
00000 00000 0000() 00000 00000
00000 00000 00000 00000 00000
00000 00000 00000 00000 00000
00000 00000 00000 00000 00000
00000 00000 00000 00000 00000
00000 00000 0 (WOO 00000 ()0000
( ~1701 l B bullJl~H 0 lt~( I BIJI ~ ~i Ill
o~-~ 0 ~j
- I ult-fc 0 I Hbull)I o bull)abullgt~
12 (1 41 17bullgt7
0 1 l I H) 4 ltmiddotI 1
VOUJME (ClJB MIC)
MFiN ~Tl) 111-V ru-1N 10C Ct-OM MN COE- Vtll
2lHH7 gt HUB
-2 7m 0 060 1)
( IJl)()C)
0 001 00000
-ltj 74a 1gt 0001~ 00000
00000 0()000 00000 00000 00000
00000 00000 00000 00000 00000
ooono 0 1)000 00000 00000 0 ltWOO
2 7 0 1J lJ I J J
-2J+H () ()~(
B 77gtB
lt ~
lt
C
J c
~ lf
_
ii
(]j (fJ
J I 0
a1 c r--- l)-- Clltr (Y) r- ro C)(V) gt deg CN C1 (0 I ~ C M
N --- r--
w J
s rJ gt-
c
~
LJ
t-
~ - -lt-[-bull
0- ~ l L L l~ -18~~
L a- - ~ I-
-I - ---
-
-- -- -- - - ~l
-1_ - - --
--~--- [-middot-~--- t-- -- [- -_ 3J ~I r-
z g ---
l
t--[-
l l c +-shyc
~ Ci - - I
-1--r-middot~- ~
L l- L ~
L
II V
C
c
t--
-= -- - - ---- - _
middot= - middot-= l J l~ - ~ ~ L - ~ L ~ [--
II II II II II II II
_
---
--
z
- I
==== lt _ --middot --~ 7 _
- ~~- _
z ~ ~ =lt ~~~ti3
f
gt =2 lt_
2 z _ lt=lt i~iS
-- z ~ lt
~~~t3
~lt~ =--~~~sect~ _ -- -middotImiddot_
L -_ -
F-4
(lJ II) I
~ Or--M
ltlJCsr-M ~ O) bullbullbullbull ~ lCMIO bull- ci gt (= N lC I M ~ 0 N
_
J J
---r--co - - ~l r-j = ~i
~~ 2-
88 rJ--_-ltI -
lt ~lt
z
Cl J ~ _ e-l ~~
- ~ middot~ e-- l ~ l~ [- - ~ -~=----~
_ L
L
--o---
-o-~shyc L~ ~ - 0 - Ll-C d o~-
11middot II 11 11 II II II
-_ 0- _
--
-- lt== ~~ ltZ z-== -~~=E~8C~l~-=rI - lt ~ fJ
ltZ lt---=lt===~~~- _ ltltlt=gt ~~=~~~~~ -
~ ~ lt -=z~ z _ -shy middot--z =-
-middotmiddotmiddotmiddot middot-middot -----
r bull bull _
- _ - ~-middot __ _
~ _
middot- - --- middotmiddot
_
- l
~=~r -~= ==
lt =~~~ti
F-5
FIHJ-ll Nil MAS~ CONCFNTllJTlOi SJ I SMllLE I l) Al OJ CUMJIATI JIbull t-llHMHY
TOTL iHf- SCNNFD lOTiL Illmiddot OF FILTFTI TOJI Ill- JSlllmiddot~ll I I OIIOT FlltCT [ ON 1111middot OImiddot ~1-M iIHlllllJNE TO 1 1 I ()lIJMImiddot ()I t I H TOTI V 1111-11~ COUNTED
= 1080000 SQ MI~RlozNS = 11 900 SQ CM = 1 l bull )00 SO CM = 1 00000 liHlS
11 IH) so CM 2G70000 CUBIC METfiHS
1 ) fi FI JllIIS
Man vi l1 e D-2 Baghouse
723 0 347 -633
Cl Ill Y-OJ l LE MIPII I BOLF Jf1BI(UOUS NO PATTEnN NON-ARHFSTOS ALL FI BFllf~
F 11l-ll f()llrlr ( I 1IFH-- l 17 G o 00 00 00 19fi
Tl fL P1 Sl~JNNFD Tl lI ltl- OF 1bullrn~n H 11I HF ~IJlbullI) lll 1 ll(IT 1-ILCTON IL CW ~U-1 HErlllllNE
THl1 (11llrJI~ OF 1IH Tt1T-d 11111middotH COIJrfflI)
= 697~oo so MICRONS l l 1)00 so CM t I JOO t-o CM
= 1 00(00 lHT~ = I l 1)0 ~o GI
ant()Oo cun1r NETEns = 7 0 I l lWllS
Manville (upwind)
CIIIIYsOllY AMPII I nou JMfl[CUOUS NO PA1vrrHN NON-A~~HFSTOS ALL FIBFHS
v1mI1 Cttllffl l 1-- I II I fl~ 1 () () oo l 0 00 40 70
1middot1111unmiddot ToT1L f J BFllS o ooo~ ooow- t OTT~ ( 01rn 57 I 1~m too ooo~
1middot 1111--11 1or111rnnT I ON I Imiddot I 111 II~~ 1111 ~o CM OF f I LTFTl) 1 11111 1middot111 CUil middot11bullnn Illmiddot J l ll)
O OOOlJEmiddotH)O UUUUOL+OO
0 00001-1-00 0 00011 lbull-Imiddot00
G 1 I nn+OG 1 HJ llmiddot~+OG
00001+oo 0001101-1-00
r n~+H+uG l 7llHmiddotlE+Ofl
l bull l 1)1-n+O(i I J ~ 1)(11-middotl()i
T11T I ~Ull FiCImiddot 1 ll f ( ~ l lrl ll11 tn Ii OF F 1ITFll)
f ill I I I i 11 I U l ( i I Ill J
0 OOl)Ol ll() () U()0 I Imiddot()()
0 UOOOlmiddotgt1-00 O OOlgtUlbullgt100
l 0 1) ) I 1 lJ J ~ tl( 1H-HH
0 UUUOlmiddotH)I) o 00001-+oo
~ lt1Hlt1(E+OG 7(H(Ml-middotHM
0 OOOOJbullH)O 0 0000lbullgtHlO
77 I
N ~
rl gt- 10NCJfffll1T I ON
r orR AMS ITll ~q lfI (II- FIT n~ll) - LIi LIIll 111middotTU OI 1lll)(PGRAMS II i1 Ill MlmiddotN
bull I t bull I 1 f I ~ middotr1 Ill lmiddotl 1- [~ IOC ( Hl1I f r~ Ull-F ll
0 0000 Imiddot bull-110 0 0000 Ibull HO
0 ltHgtOO 000110 t) (HJ()()
0()()()()
0()000
llOOOOF+Oo 0 00001bull-100
0 ()000 OOtlOO 00000 0 (WOO 00000
0 (((7
0 ~0-i -0 middotlbullribull)bulll-
0 ltl l lt 0 oHi
00000 00000 00000 0(0HO 0000()
077[0 0 -1117
-0 ~bull)I)()
0 7middot1middot1 (i () L~bull)7
0 7~B( 0 I I
-o alt17n 0 (1)~1 0 17lt1 I
11 I 11ITlt I j i 1l 1 111bull I
f 11middotgt N gt I ii ii I V 111 ii IOC tHlfl flf1 Clll-tmiddotmiddot VAIi
00000 0()000 0000() 00000 O 0000
041000 ()0000 () 0000 o 0(1()0
0000()
0 llfl(i7 OltHJ
-~ lt-11 I 0 07 I 1 () I)lt)
0 (H)ll()
00000 00000 0 ()()()() 0 0(100
0 0 Omiddotlmiddot l
-c07 1)
0 I 1 I( 0 1101
0 l l j--
OOGli -~bullI) ( (J
0 1lt11 I 077G7
I 1 (71 rJI C)
tll N -r11 111-V rmiddotllfl trn CI llf I 111bull1 COIT rll
() ())(l(J
00000 0 ()(1()()
0000() () (1()(10
00000 00000 00000 0 (I()(()
00000
0 ~ I Ifdeggt 0 I Imiddot~ I
- I H 1)7
O I GOO I I i ) l)
0 0001) 00000 00000 00000 00000
0 1 1)17 o~ViH
-1 0 1)gtG 01lbullH 0 7 I till
0 l l lt G 0 ~r1
- J bull J1 1) I 0 17 I bull at-lt
lllllrmiddot lt Clll ii IC)
~I Imiddotmiddot N ~middot1 D 1)1bullI f-1 Imiddotgt f IOC ClltHI rm uw-- i ll
O(lOOO 00000 00000 () ()(J()(J
(l(t(()()
0()()00 00000 00000 (1(1(1(()
00000
000fi7 OOOGO
-G Jll7l 0011G 1 H20
00000 00000 00000 000()() 00000
0 0 I bulllCi 001w
-4 iHd 00101 I ~700
OOIOH 0 0 I Ilt
-(1 j l)~G
000(i 1tHH
--
--
L c~
U c t c
(
J
J -
C
C
c g 0 0
+ + e--c~ =~ - r~ C- l~
--
+ +---
- -C 0 o- -c--1
I
lt t w
lt
C
ta J ~ ltCi
lt ~
a---~ -- c gt C 0 rtl 5 -
VJ C f-rJJ c
[J_
ltI ~ C z
z _ ~
~ lte 0 ~
C
tJ
e
C
~ c C
L N
~~ C 0
C
l~ 1 0
t CI ~ ----~ -
i-
cc cc + + t=~= g~ --
-~ C + + t~ ---[w (~ lC
-
ccc ~t c5 ~ ee 00000
co--oc I
- - - - -_____
------ _ - _ _--__ ---__ ---- _ - _
ooocc
rr ~ c-~ E lt
C
-
~ C g L-
gt o + + ~~ ( L C- lN o -l
t l - -I
~ - L~ [ti e- [
l _ 1
- - ~_ C
gg -- -- ~ -0000-tr--- ____
----C __
- __ - _ --ggggg
- -
C
+ + -- --- -~_shy-----_ _ -----___ ooo~
-----__ ___ _ - - - _ - _ _- - - - ___ 0~000
--__ ---_ __ -- --c5 ooco
C cc
1 II II 11 11 II II
z
-
r ~ c
z
~ t lt i C z __ -- -~tt ~ middot~ ~
-middot middotmiddotmiddot _ - -----=___
_ =~ - _ -=shy ~
-- -~== 2~= ~ () V)
_middot ~ c( c(
0 er ~- ~ ~ lt 0 0___
-
- amp lt~ ~~~t8
F-21
) ) ) )
TOT1L AllE SCiNrmn TIIT1 L ll lbull1 tW FI 1lFll TIIL1 11bulli 1~lllbullll lllIIOT lIIJClION -11- OV TUI Mlbullrllll1Nft TOT1 I VOIIHIIbull 01- I Ill TOTmiddot I Imiddot I I 11-11~~ COi l NTlbulln
_LI__l_ ___l 4 bull P 7 bull P A R l n A 4 R 1 n 4 4 ___ 7 A ___ _RA____ 1 A 1A 8 1 SL__ 1 R 5 1 ~ 3 _1] ________ 111~____JJ__) 1 ~---~-~ 4 9 1 ----1___11
A0 1A4 JR 17 lAR A 41 4C 174 A ll~ JnR ll7 9 5 R12
----- --------- - ---- ---- -- middot-middot --- ---- -- ----middot-- ----middot ----~ 7 c ~ bull n 1 1 2 l 9 -s 1 -i s s ~ c c__ bull A____02 1 1 2 1 r n 1 c 0 r-- l ~ 1 o 1 0 1 t 1 o c- ----~--- -Pl Ol OP lA2 11A 61 c 0 ~ cq03 l 5 4 l 4 4 1 i r 7 -i r) c q 1 deg 1n
qA RO 107 ~_12_7 SA c 7 A04 _l _ s_ _________l 7 l RA____ 7 ______l _o 4 __ l nn_________ 0 _7________ R_l __ o ~ _____ J(__7 11__r ___ _t-__ o ____ ln ___ h_ ______ An____ c-_n __05 ]lA 140 170 7cA __2n1 qc- 7q 1n1
c 2c 7 lR 1 O 406 11 c A 0 4 n 3 4 7 l 4 R ~ 7 7 5
___ ___~ c Z bull A ~R_ ---2_h 2 bull bull 4 e R bull C07 l_ H l_ ri ------- R1 --- c ----------- AQ _middotbull 4_p____ r 7 ___1[_2_ _r--_______ _ A____________ )_3-____ _____ __t 9 ______ ~_____ _3 __ --~___()_______ __I__08 C7 lR9 9A 1n~ ll7 S AS 84 ~Q 33 11 11 1n R 3A 3309 7RJ 1R lA 14-1 17fI 77 4R 7(_
---~- L~ 1 7 -~ 1 7 4 1 l 1 9 S 3 9 C) 4 9 A ~ 9 q10 __ 7 c _ ___ _ 2 RA l A ____J__6_R 1 C 4 7 P C 1 rq I
-----~~a-R___-=4bull2_1 e __ middot-middot middot bulls __ - __ IL 9 1 ~_1_ ___3 7 40__19 1 ~~- gt q6_ (- l Q 144 43 _37 _ A f ---middot ~ bull r bull q ~ 9 IL ~ 1 _0 -~ bull 1 A 3420 l r -- I- I q4 l qo l c c 0
sectJ~ BLVDII EJ SEi~D~ 1 nu tmiddot-middotmiddot-middot J _
tl~ r-middotn ----i J middot--1_ ___ ~1-1
A f gtiff_ ffmiddotc YYt)II ~ r bull Ja------ ijlttQ~QW~
___ ~Q -~imiddotbull1 ~H fl(--middot
-lt) bullbull
1 middot
middot--middot middot-- _
middotmiddot_ u - -_-~~middot_1r
~-- ~ _
~ -middotmiddot bullY
j ~ ~ L ~ ll c_l
~ pf-
111-r ~ (i middot) ~
bI middot_FT )-I Imiddot ~middotmiddot () --
I filfltf) J ~3 ilj-t ~ C _G~ llli_A T1 AN B~ACI
I~
603 middot Alondro Pork
l U (
Goll Coorll
~ htyen 1~J-(_
d n t U Wik c rr 1 i t_y of Indus try
(
H-7
Plant Site - Dow ch em i cal US A Pittsburg
I bullbull middot1 Sfl Wi - bull AUIIUlllJLII+
BELMOr r l -- -- 7iSHOfgt ITHACA lN I
N[VAOA lN TGERS llltI
Q LN
GfORC~
72copy~~~~-0AYTON l
I I+
__ middot
1- - -~ middot
I I
- ----------
Plrint lite - ilu rnnt Antioch
) j
(
H-9
Plant Site - Johns rlanville ctockton
a a
LEGEND
CENSUS T1ACT1
51 )I 51 07
1980 CENSUS TRACTS SAN JOAQUIN COUNTY
37
lhe microldrrt fluriti1~s tuC after tne microrimar y reactions by separation and digt ti l I dt i ur1 fgtr-ocJuLL 1 s ctured ctncl umiddot 2d ctS foed for VCM production All
sourclS agre~ tl1dt c1ission of EDC is (f concern in its productionstorage
process stages ctnJ O little importance in VCM production steps (J~B 1980)
At the timt of plant inspection some storage of EOC existed at three
ledsed storage tanks located at tne Port of Los lngeles 1-iaterial has not
been 1-Jithdravm from these tanks during the 1dst year and there had been
discussion to consolidate materictl into a single tank
All microrocess components handling EDC storage are now tied to a closed
ventilation incineration syste11 It is expected that virtually all
chlorinated hydrocarbons includiny EDC will be etfectively destroyed since the
system must demonstrdte VCM concentrations are reduced below 1 ppm Firebox
tempera tu re is aoodegF and residence ti me greater than 1 5 seconds under
heaviest flow conditions Note that by way of comparison combustion perforshy
mance datct for PC8s are 99995 destruction at 1832degF and 1 second residence
time and 9Y9Y9YY4 at 2 seconds dnd 2372degF (N Flynn SAi Personal
Communication)
Fugitive emissions from valves and flanges are monitored by the
plant according to the requirements of SCAQMD Rule 4661 Pumps and
compressor follow ~ule 466 Emission and control requirements for vinyl
chloride are specified by ~ules 1005 and 10051 EOC concentrations in
wastewater are monitored several times daily in tne primary EDC steam stripper
stream and once daily in the composite plant outflow stream Daily water
disct1arge limits are 2~ ppm wit11 typical montt1ly averages being in the 8 ppm
vicinity The discharge limit is ~mbodied in the discharge permit (number
5061) vJith tt1e Los Angeles County ~anitation District
Tnere are a number of points at which tt1e rnc can be released to the
dtmospnere and they are similar for both dir~ct chlorination and oxychlorinashy
tion These include the following along with their emissions factors for direct chlorination
A chlorinator vent 2xliJ- 5 IOdSS per unit mass EDC produced
I) I ight end column vent -52x lt)
C dist i I I at i un column vent lXlU-5
u storage tank breatning 7xlu- 5
133
E storaye tankworkiny loss 27x 0-u 5F fugitiv~ emission bxll -
-JLi ildSteJdter l gtX] U
Em~ssion factors are si~~lJ fer oxychlorination processes
Enlission factor vau-5 (~1rbulle Tdilt2~ frcn tne literature (JR13 198U) and
assume 9b iciency for incineration (A-E) It is also assumed above that
EIJC emissions to water are Lf~ uf ~rihsion to citir ard that in wastewater
treatment 100 of EOC discharged tc1 vater is rel easl~d to a i 1middot ~ These emission
factors 1bull1e1middote used to prioritize r-2ieaies iJut were cldrly uude approximations
to tre planL l)ecause of thlt ncine-xt~on system it was anticipated that
fdctors A-E would be ri2dtKcdft 1-middotactor - 1nii~t1t be reduced since a monitoring
program had been in force al20st o~e y2ar Factor G and the off site storage
tanks~ vni~cn are not tied into lt1n ircinerator system rniyht -iominate emissions
Emissmiddotion limits and concomitant regulations embodied in Rule l00S of
the SC-l)MU have necessitctted tlw incinera-cion system Ninety gas chronatograph
probes are located throughout t11e plant including the stack of the primary
incinerator Con cent rations of vc1v1 are reported es sent id I I y be I ow tt1e r~yu la -
tory limit of 10 ppm and in fact below the limit of detection somewhat less
tl1an 01 micropra
lUl MEASUK~MENT APPROACH
Tne objective of tt1e proyram is to determine a pl ant emissions for
EOC It is not accemicrotabl~ to dtmiddotvelop sucll informatmiddot1on basect upon published
indlllstry Jide esti11dtes of plant control efficiencivs Fortunately it was
possible to desiyn a monitoring and calculational progra1n to determine EUC
emissions for tne site and not ctbsorb a disproportionate share of program
resources
Hie re are tnre~ modes ot re I ease from tt12 micro Iant dnd d fourt11
offsite
bull Post-Process 1nci11er-dtion
Processes A-E of ecc ion lll 1 dre a 11 1ented into the pl ant
incinerdtio11 syste11 rt1e conceritrcttion of VCfmiddotl contir1uously iectsured in till=
stctck cts SJdS output remicroresents d rectsondble UjJper li11it to diJply tor EDC
con cent rat i ons s i nce i ts ef f i c i ency of i nc i nera t i on 1s d t 1 east c l1 u a I to tr1 at
134
ot VCibulll Tlerefore knowledge of the system dirflow rnd VCM concentrcttion are
tt1e necessary and sufficient conditions for determining the bounds of EDC
reledse fhe pldnt expn~sse I wi 11 inyne~)s to provide tnese data in order to
support the calculations
bull Fugitive Emisions - Valves Flanges Seals and Other Sources
Fugitive emission from the valves pumps and flanges can be
determined more precisely tnan was anticipated since Stauffer has completed a
comprehensive leak inventory of all such components in compliance with Rule
466 466 1 and 1005 The Inventory delineates all leaks uncovered by their
three man crew throughout tne year and indicates the screening level in ppmv
and component identity In consultation with Stauffer we will be able to
identify the total number of components associated with EDC handling systems
( 700) the distribution ot substance composition streams the distribution
and incidence of leaks by hardware component type and leak rate We will
utilize this information to provide historical data to compliment our Foxboro
OVA sampling at the site dased upon this monitoring we predict an EDC mass
emission rate for the fugitive releases The plant has agreed to provide the
necessary information We propose to meet witt plant personnel prior to the
start of monitoring and finlize the sdmpliny strategy to accomplish 100
coverage of the lines of hiuhest EDC composition The sampling approach and
calculation of mass emission are described in Section 72
bull ~~as tewater
From examination of the emission factors derived from published
literature (see Section 10~) it is clear that EDC release from wastewater to
air is potentially several 11rders of magnitude higher than any other plant
source l1ased upon microl ant ml~a sured concentrcJ ti ons of 8 ppm EDC in water a more
realistic emission factor wgtuld b 24xlo-4 1-iassunit mass EOC produced
Clearly this could still be the dominant source Furthermore the release
point would be expected to middotgte locctted between the plant and the sanitary
district treatment site whi1 his~ kmfrom the plant at 24501 S Figueroa We
believe it is necessary to ndemicroendently confirm the average 24 hour EDC
concentration in the dischar-qe valer by obt1ininy the refrigerated composite
sample We t1ave identified ttie I 1nes of int~rest and received agrt~ement to
sample and analyze for EDC in the stream ~e propose to draw a duplicate
sample from the compositor nd ani1lyze for its rnc content This will be
135
compared with Stauffers para11eI tnalysis It is not reliable to obtain
direct readings of EDC by survey instrument in the air above the water flow since concentrations at crny single point will be i 1 the near dmbient range
Emissions will be cc1lcJLtecl 1_sing the plants vulumetric daily flm
and ass~m~ng that complete degass40~ of the effluent will eventually occur
This assumption is based umicroon the relative volatility uf EDC and tne distances
involved However it should be noted that no direct experimental or
monitoring data is available with which to confirn this Th2 composition of
the discharge stream is unknomiddotwn since it merges into a larqe multisource flow
Conversation with the LA County Sanitation l)istrict (J f1ilne) reveals the
stream to be both exposed and covered Vents exist where me measurements
could De made to assess gross leaka~e at key points
We will obtain samp1es of several plant process discharge streams in
40 ml bottles witn no head space Transit time fr0m sample collection to
analysis will be minimized and wi l I not exceed 24 nours This time frame is
conservat i v e al though ED C i s a v o l ct ti l e mater i al ar1 ct vJi 11 undergo
concentration degradation and outgdssing lnalysi) will b(~ performed in tt1e
SAI Trace Environmental Chemistry Laborcttory utilizing pur 1Je and trap analysis
FID gas chromatography A trial a11alysmiddotis was conducted and the EOC
characteristic peak was distinctive down to the ppb level Therefore the
analysis should easily confirm concentrations in t1e 8 ppm range
Offsite Storage ranks
Three EOC leased storage tanks are located offsite at the Port
of Los Angeles and are owned by annther firm Annual emissions from the tanks
were very roughly estimated based un AP-42 JRB 1980) as 44 kkgyear
based on preliminary estimates of stored quantites
Considering tne magnitude of this source these storage tanks must
be investigated We will gather information about their configuration and
control i n order to ca I cu l ate their e111 i s s i on s bull
136
lU3 UEfERM I NAmiddot 10~~ OF MISS IONS
1031 Incinerator
Vinyl chl 11ride co11centration is monitored dt approximately 90
locations throughoul the pl trlt (Lanyner 19Bl) including the incinerator
output Conc~ntrations are reported by Stauffer dS less than 0l ppm at a
flow rate between 10000 ant 12000 SCFM Although no direct monitoring of
EDC is conducted it is posible to conservatively bound the concentration of
EDC at U 1 pp111 Tl1at conce 1ltrat ion is a suitable ci10i ce s i nee it represents a
conservative bound on tile VCM levels measured near the stack output
Furthermore the molar volume will be taken as 224 L rather than the higher
value it would nave because of the sl i~thtly elevated temperature at the
detector location
Using 12000 SCFM one has thE annual emission of EDC as 12 x 103
SCFM x 283 LSCF x 526x105 minyr x lmole x 97gmole x 10-7vv 224L
Therefore incinerator emissions of EOG are thought to be bounded by 773 kg or
170 1byear
10J2 Fugitive Emissions
Seven hundred (7UlJ) sources were surveyed comprising nearly 100 of
EDC service All accessibl~ plant areas with streams containing greater than
1 EUC were screened except for a small number of devices located in areas
11lere active maintenance was being condJCted It is believed unnecessary to
p2rf-gtrm any emission factor adjustment since it is estimated that gredter than
95 of the requisite components were screened
Three leaks above an arbitrary OVA reading threshold of 20 ppm were
parameters and mass emission numbers fhe upper bound leak rate was
determined to be close to twice the nominal leak rate which accounted for the
response fltlctor of amicroproximamiddotely 05 by Radirn for TLV detection of EDC
( 13 rown 1980) bull
Stauffer found and reported approximately 10 leaks in the EDC
service during 9 months previous to the pLrnt testing Assuming no111inal leak
137
) )
Table 103-1
STAUFFER CHEMICAL MASS EMISSIONS PREDICTION FROM OVA SCREENING VALUES
OVA SAI Actual Upper Device Response Response Cone Source Leak Rate Bound Location (ppm) Rae tor (ppm) a b Se I Re Function lbyr lbyr
Gate Valve Unit P1559A
Valve Heavy Ends Column Unit 14439
Compressor Seal EDC Storage Unit Tl062
I-- w co
300 11 273 -035 096 0 17 153 D 6
11 I500 114 439 II 241 C 1
r III II II o-is lUI bull2000 vl 1 1818 A 9~
All emissions 100 EDC (12-Dichloroethane)
11
values in the nei~hborhood of 2000 lbs~ total could be emitted annually
Therefore it appears ttiat major reductions in the mass emissions were achieved
by the company run inspection prugram and the fugitive emission source has now
become of seconqary importance as a fraction of total plant emissions
lUJJ Wastewater Discharge
Wastewater samples were collected in duplicate from four sites
within the plant for determination of ethylene dichloride (EDC) concentration
The samples were collected in EPP standard 40 ml VOA vials on July 1 1981
Analyses were performed using standard purge and trap techniques coupled with flame ionization detection gas chromatography The results are given in the
table below
Sample description EDC conc~ntration range (microJml)
EDC concentration average ( 1-19ml)
Approximate fl ow ( gpm)
PVC Interceptor Box 82 - 108 95 200
EDC Stripper number Chlorination area C1404
2
011 - 014 012 25
Final collection site for pH adjustmentP663 349 - 3B2 366 350
Final discharge site sanitary sewer 6 - 254 158 500
The water in the PVC lnterc1~ptor ~ox was warm and represents washings from the
PVC reactor which travels t 1J the Interceptor Box in a concrete drainage ditch
Water from the EDC ctripper was vPry t10t and because of its heat was difficult to collect The ~1eat ot trie wat11 r may in pdrt account for tne
relatively low concentratio11 of ElC here as the EDC would out-gas from the hot
~vater more readily The fir1ctl collection site tor pH adjustment is a large
concrete container middotvith mixers in it located just prior to the final discharge
site Water at the final dischdrg~ site is being constantly aerated due to
the speed that it flows thru~yn the concrete drainage trough This aeration
could account for tne relatively large range in the rnc conotent found here
The average EDC concentraticmiddot1 at tne final discharge site is middotbelow the daily
discharye requirement of 25 19ml EDC
139
Plant oersonnel indicdted monthly iverage readings are typically on
the order of 8 ppm but daily averages can re 1ch greater than 20 microgml In
addition LA County Sanitation Uistrict staff (J Milne 1981) indicated that
an on-site impoundment pond can become laden with EOC during certain abnormal
operating periods and discharge Vdriances arl requested by Stauffer This
mi 9t1t Jccur on the order of O1c0 aCii ecH c1 1d t1erefore is no-c expected to
signifcantly impact averag~ cnnJJI discharg ~ values lt s not expected at
this t1ime tEit evaporative eriss uns from this pond are significant except
infrequently during upset conditions and spi 11 control operations Program
staff ~~ 1er2 una111Jare of any periods ~h 1~n EDC cimtent in the ponds could be
appreciable and therefore no sampling was performed
Utilizing the average uf the two final discharge concentration
1readinJs (158 micrograrnsm~ 1 and 1G 1Ji gpm fl)W one has 34600 lbyear of EDC
released into the sanitary sEwer For the pJrposes of calculating population
exposures from plant releases it will be assJmed that the EDC is locally
emitted from the wastewater streams Tl1ere ire no monitoring data available
with which to develop a more accurate release profile
However emissions of me fron the plant wastewater discharge can be
calculated according to the method of Mcmiddotckay (197S) as modified by Dilling
(1977) It should be noted that this mLst be considered an estimate since
conditions of fl ow and the presence of other substances will influence
emission rates
Using Uilling (1977) for non~erated flow first recalculate
Henry 1 s law constant (dimensionless-my of chmiddot~mical per liter of air divided by
mg of chemical per liter of water) dS
1604 P M 1 Wl
TS l
)
wnere
H Henrys law constant dimensionlessl
Pi tile compounds pur1~ componen- vapor pressure in mm Hg at T
Mwi the 11olecular weiglt
T tile absol ~te temperature of he wastewater in K
Si tt1e compounds ~uluhility in myliter at T
Then for UJC
(l6u4)(7~)(9Y)H = = U0447 with 1
(2lJ4)(8690)
LIii bullul1Jl)i I 1 1y 1 middot UL i11 wt1Ler yiven -1L L0degC is 8u4JU UHI tile vapl)r pressure
1 4 psi ( 7 2mm) at 7 0 degF bull
Tl1e overall liquid mass-transfer coefficient Kil is given by Dilling (1975)
as
(2211)(06) in mhr
-f-0-4~ (M _) 12+ 100 Wl-H-
1
=U108111tH
Then from Mackay the percent desorption is given by
c 1 = exp (-Kil tL) where
c 0
c = the concentrd ti on at time t of EOC 1
co = tile initial concentration
L = the Ii quid depth (Ill)
t = tt1e retention time (in hr) of the liquid in the wastewater
system
Retention time is bc1s~d 11pon a flow velocity range of 3 ftsec as estimated by
the Los Angeles County Sdnitatii)middot1 Dibulltrict (1J Milne) over a middotdistance of
ctmicroprox i 1nate I y S km to the Joi n_t Wdter Po 11 ut ion Contra l Pl ant Sewer fl ow
depth throughout the entire route have been estimated by J Milne as typically
1 foot 030m) to the Davison Pump Plant and 3 feet (09m) to the Joint
Tredtment Plant distances assumed to bt~ 1 mile and 2 miles respectively
Trdnsit tiine l)eco111es between 05 tnd LU hours sequentially Then computing
the net emission reduction as the product of each leg
Therefore 26 of the EOC i erni t ted bdween the pl ant and the Joint Water
Pollution Control Plant Rtgtsfdence ti111e and conditions at the plant account
141
for additional releases and residudl content is forther emitted between the
plant cnd the ocean discharge poiiit ~-Herever the flow is r2tained aerated
or shallow emissions will be accentuated
Offsite Storage Tank Emissions
There are three storage tanks leased by Stauffer for EDC storage
1ocated at 22nd and Gaffey Sts ~ bull San Pedro These tan ks a ~e used for long
term storage rather than providing feed on a routine basis Therefore
breathing loss rather than v1orkir 1J lass is )f concern An tanks are white
two being 67 ft in diameter while t~2 third is 57 ft All are 40 ft 3 inches
high and have capacities of e~thi l0~10000 (2) or 840000 gallons The
tanks are cone roof type and are not tied into vapor recovery systems The
South Coast Air Quality r1dna~J~ment u~ str4 ct has based their estimate of
emissions on the specification of 12 foot vapor level Using the AP-42
formula for breathing loss with the vapor pressure of EDC at 20degc and an
average diurnal temperature variation of 26degF one hds for each of the two
larger tanks
l) 173 H bull51 T bull 5 LlI = (619 x 10-S) M( _P_ tH F CKcp 147-P
68 173 51 5 = (619 X 10-S) (9119~ 116 bull (67) ( I 5) (26) (1) (1)
14 7-1 16)
Thus for the two larger tanks LB= 472 lbday
For the third tank 0=57 and = 178 lbdayL8 The total emissions become 23724 lbyear ~ince he plant is in the process
of shutdown it is not clear what the liquid levels currently are Note that
if the material in the three tanks were combined into one totamiddot1 -~1nissions
would be reduced markedly Exposure to a population from this site was
determined separately since it is lbullgtcated over six miles fron tl1e plant
142
110
ASSESSMENT OF PUPULATlUN EXPUSUR~
111 OVEKVIEW
A major program goal wa~ to compare emissions from the various
sources and identify and rank any 11 tlot spots in California where the general
population was exposed to elevated concentrations of carcinogens A simple
Gaussian dispersion model was therefore used to obtain order-of-magnitude estimates of exposure of the genermiddotal population surrounding each source
Since this was essentially a screening study use of more sophisticated models
was not appropriate
It cannot be emphcsized too stronyly tnat most California residents
are exposed to emissions of hazardous substances from a variety of natural and
rnanmade sources Urban dwellers dre typically exposed to greater concentrashy
tions than rural residents however all are subjected to so called 11 background 11 levels from multiple sources In order to place stationary
source exposures in perspective the typical ambient levels of each substance
were identified from the literature and compdred with the concentrations due
to the emissions from each ~lant Exposures were thus expressed both as
absolute quantities and as increments above background
Comparison of plants presents a further difficulty in that various
substances are being consict~red No attempt was made to evaluate the relative
importance of exposure to two different substances such as chloroform versus carbon tetrachloride other than by ambient concentration
112 DATA SOURCES
1121 Meteorologicdl Llata
The dispersion model to be descritgted in Section 113 required input
of annual average wind speed and frequency of occurrence of wind from each
compass direction These data were obtained for most of the sites from the
South Codst Air wuality Management District (SCAQMD) In other cases (Kaiser
Jotms-Manvi l le l)ow and OuPlt)nt) only clai ly cJVerage wind speed and frequency
data were available In al I cases we used data from the meteorological
station nearest the modeled e11issrnn source Table 112-1 summarizes the
141
Table 112-1 METEOROLOGICAL DATA BASE
Observation Site Data Period of Pldnt (Distdnce Source Observation Remarks
from Plant)
A11 i ed Che11 i Ca 1 Lennox SCA~MD 1Y71-1974 Averaged hourly readings available ~ 1 Segundo (3 111iles)
S ff~r-- Uemical Long l3each SCA()MD 1962-1974 Averaged hourmiddoty leadir19s avai1able offsite Carson (3 miles) storage tanks appruxi1ntL~ly 5 miies from
Figure 114-3 Cumulative Population Exposure to Carcinogenic Polycyclic Aromatic Hydrocarbons From Coke Ovens at Kaiser Steel Corporation Steel Mill Fontana
CV)
0 -I X-V) V) a _J
s 0
C 0 micro co s micro C a u
l ~
14
1 --
J- t_bulli
_ C
uw m 0
1 -0 a bullbull micro
microdeg ()
0 micro tbullmiddot C QJ V)
0
X LLJ
C
middot4middot~middot C 0 micro rt1
-- - 0 ~ 0
0
n __ _______bull~ -______ --al_ - ~_
I bull
I I - bull
middot 1
-~1
l i l ~l
~ 0 I 1
Li bull I t_~I
i bull -1 _1-J
0fmiddotbullmiddot-middot+-middott-t-r+-t-+-+middotr+-t--+-1--t+ t f-t-J-t--r-+tmiddotmiddot~~bull+-t--1~~bull-+-1 Id middot t 4 E1 (
I ncrementa 1 Concentration (~gm3)
Figure 11 4-4 Cumulative Population Exposure to Asbestos From Johns-Manville Plant Stockton
I
)
--~ ~w t _ ---~- ----------~ --____~-=~ M
~
bull I bull I I I I ii i Q t I i E e- I Ii tl I I I 6 I a I I J I f t t bull I G fl I
a I e 1 1 a 1 1 1 1 iii bull bull 1 1 bull 1 1 bull 1 1 bull bull bull bull 1 1i bull bull 1 1 bull a bull 1 1 r ~ middot u ~ ~ I0
+H~ ~ ~~~~-~~~~-rmiddot-~~~~~~~~~-~--~~-~~-~t~~~~~-~J C
~1_ c bull 1 t~ bullbull bullabmiddot 11 l bullmiddot II bullbull C- I u~
0 rl
gtlt I I I- 20 () ()
CJ -1 ~_J
J Cbull s
I If I I I I0
C bull I Ii I I I I I I I I0 1E
bull-+l I t I I ro s Lower+l
I I I G I iii I bull
() 14 I I IC
u I0 I I II Iu
C
1-_~I fi
CJ ~ I-- (j) CJ I G I I p +l
ro +-l I I I1~3~V)
0 +) UDf)Pr u I I 11 I I
CJ 1iII
Vi I 0 I I It- I a X
0 ~
w I I I I bull
Lower C 0
middotr-
- 4+)
(1j
J Q_
0 bull1
ij a 3- ti I
D d I I I I
I I I bull
II B I t
i ] i amp I
1 ei ~ amp 3 ii
I I I I
e ft a a I bull G
I I I bull I
I I bull I I t If
UpPrf
I I
I I I-_L~ lI I I f
I I I I I
- I I
j I I II I -
lI ImiddotbullII
l I
l Li
Incremental Concentration gm3)
Figure 114-5 Cumulative Population Exposure to Perchloroethylene and Carbon Tetrachloride From Dow Pittsburg (Solid Lines) and to Carbon Tetrachloride From DuPont Antioch (Xs) Results For Upper Lower Bounds on Emission Estimates Are Shown
1i-1 - middotbullth~middot t =lttF--i-+-middotr+-r+middot~middoti-middot-~-4-4middoti t -~--is -~middot-rmiddotibull--+ u ~~ i--t4~ t-bull~~311 li-i1middoti- C - JJ _ CL
-- I 1 t middot1 c7 Ibull ) ~ I-middot bulli C bull 1-_ 11 11gt1l11 ~bull -a bull-bullbull t a- II bullbullbull bullI lt_I Iii _- I
Incremehtal Concentration (microgm 3)
Figure 114-7 Cumulative Population Exposure to Carbon Tetrachloride From Allied Chemical El Segundo (Curves Corresond to Upper and Lower Bounds on Emission Rate Estimate)
--- I
~__~-unt2wbull--o1WCUatJ41ia1
16 0amp (V)
0 ~
gtlt
14~11l 1l QJ _J
~
1 middot-0
c t1middotmiddotmiddot 0
micro ro
_ c micro ~ 1euro1pound1
0- QJ u-J c 0 u O =-middot ~-= QJ
c_1 micro ro +- V)
0 micro ~ t1 O QJ 1l 0 a 40gtlt w c 0
-
ro --
micro
20middot l ---a I ~ I a 0
0
pound1r-t k
Figure 114-8
8tWI-S~1111111N~-
a I __ I wi~ i awLi--~~ 2 I I
1--+ I ~--if bull =~~--bullJ-+--f-1-----middot+--+middot-middot--r9c- 1 c- J
bull _ bull -~
Incremental Concentration (microgm 3)
Cumulative Population Exposure to Carbon Tetrachloride and Chlorofonn From Allied Chemical El Segundo Assumingsix-months Operation With Each Feedstock (Curves Correspond to Upper and Lower Bounds on Emission Rate Estimate)
i u---J--bull-+ i i--+ --4--+-imf-bullbullQ~-+-+-+~bullbull-4 171 i R = Fi middot1 1bull1 r1__ bullltaa ~a
Incre0ental Concentration (microgm 3)
Figure 114-9 Cumualtive Population Exposure to Carbon Tetrachloride From Allied Chemical El Segundo During Offloading From Midnight to 6 am middot-middotmiddot------middot--middot
--bull ft ampLA 1111-~lliaIN_WPl1A 1JIU_Mt91 MJ1WWlaquoPI-W1111~ --------__----Et1
M gt~middotI
0 -I
-X
~ t~ bullM
V) middot-middot ~ V)
QJ _J
f 0
0 -
C -4 0middot~middotmicro re f micro C a u
10 Cdeg u 0 3 0middotmiddotmiddotbull~ -0 a
+-gt ro
+J V)
+-gt 0
2E1-0 QJ V)
0 a X
LLJ
C 0 1pound1--middot+-gt ro
-- Q_ 0 0
I I
~ -~middot bull I -bull-_ bull I
middot lbullgtbull___I I
frac14 llk I
I bull
-~
I
-1_ I
Ibull II__ JI
I~
(1 f---1-+ bull-r_--f bull middot~-bull +-- bull i -bull 1 -1-+--+bull-+-t-+bull--t1 ~ _ 4middot 6 E 1t1 1middot2
1 3 5 I middot~ _ 11
Incremental Concentration (ngm3)
Figure 11 4-10 Cumu1ative Population Exposure to Ethylene Dichloride From Stauffer Plant Carson
_~ ~ t M
C X E k1~
-
- c- 1-1 ~ C middot-middot -_J
I--- -J 4t1middotbullmiddotmiddota
D
-+-shy --middot liiit1J
r ~
ltL
t-1 _ V
0 -+-1
middotr -bull ~1
X
C
+- bull1 ~1 atr l -=
~~lll(k~IHmiddot~~-Wa~middot~nL1middot~~~ta KF u n P middot~dE-tveRS ~-tl~~~-u-~~~tmiddot_-~~~~liil-~-~~~ tc r
I bull
bull -r~l-yen--~1l~a l fl __ A4o bullbull bullbullbull ~ 111bull--middotbull
8 NJ industrial sites 28 samples in industrial areas with benzene consumption facilities Torrance CA Tuscaloosa AL National Forest 1000 sarip Ies-Torontt) Azusa CA El Monte CA Long Beach CA Upland CA Los Angeles CA Riverside CA
bull Assumes all year operation on either feedstock If plant operates 50 on each feed the population exposed to greater than 50 increment over background goes to 16025 - 19377 and 4044 - 16025 for 100
Ambient concentrations of benzene vdry over one order of magnitude in the literature and therefore make tnis calculation questionable For Kaiser therefore the carcinogenic PAH assessment was used to evaluate incremental population exposure above background
+ The partition of emissions from Oow are 78 carbon tetrachloride and 22 perchloroethylene by weight
173
1142 Comparison of Incremental and l3ackground Concentratiors
Those sites whicn elevate background concentrations greater than SO~~
City Kentucky 6600 The latter two are associated wi t11 EOC pl ants Data
taken in senFice s~ations ano lrctffic areas ir various cities range from -~
300-3640 ngm~ As previous~J mentioned the Stauffer plant is discontinuing
operations These two sources should be examined as part of any possible
start--up permitting activity
11422 Kaiser Steel Corporation
The Kaiser steel plant polycyclic aromatic hydrocarbon (PAH)
emissions cffmiddotise from the coke oven omicroerations Comfdrison of the cumulative
population exposure Figure 114-3 and Table 114-4 with the specified
background levels for urban areas illustrates the breadth of the exposure
distribution The five known PAH carcinogens that were isolated in the coke
oven emissions were quantified in tile ambient air of Los Angeles by Gordon
1Y76
Benzo[a~pyrene 046 ngm J
t3enzo~e]pyrene osio Benzaanthracene U lo
Chrysene U60
lndeno Cl23-cd~pyrene 134
Althouyn these concentrations ctr~ low co1npared with 1 number of other cities
cited in the literature and repres2nt a very I imited data base tt1e predicted
concentrations from the Kaiser plant ~enerally exceed these levels by d
174
significant margin ie greater t11an 30000 are predict2d to be exposed to
4redter than l()-lt t 1tbull 1111tlit~nt background of 3j nqm 3 It ic 1 iklly tlat
lHrllerte exmicroo~urt 11 Ilie drld dre also 11evatPd ovr r11nbient I1owever since
backyround concer1tr1tions of t)enzlmiddotne show large variation no population
calculation was specified
11 4 bull 2 bull 3 Dmv
Carl1on tetrdcnloride releases from Dow constitute approximately 78
of tl1e total CT plus perc emissions in Table 114-6 and were found to elevate
urban background concentrations greater than 50 in five census tracts Emissions are predominately from storage and check tank working and brething remiddot1 eases
1142~4 Allied Chemical
The Al lied plant was modeled several ways since the plant can
operrttt~ wiUl chlorotor111 dr cc1rbon tetrctlt hloridt fo(~d The c1bull-i pr~middot11ntlid in
ldlgtlt~o ll4- drld llil-B represent dflrlUil 01Hrdtion with eitt1er teed Partial
yedr operation with each feed can be scaled from the individual annual modes
and is presented for equal hltllf year operation in Table 114-9 As with the
Du Pont plant carbon tetrachloride emissions arise from feed tank working
loss fable 114-10 illustrates peak exposures predicted to arise during an
off loading cycle As expected concentrations in that six hour period far
~xceed annual average values Insufficient data were available to contrast
levels with backyround transient concentrations
175
l~O
AUHLAlHJ~ CONTROL n CHNil)UES
Alternative control approacnes for the most significant emission
sources dinmig ~he various pl1ants n ees fitie(J n the follo11 ric suJsections
EmpiiasVi nas Deen placed in dedlinq ~riU1 Uiose sources of 91reatest absolute
inaynHud2 (lbyr) and those const1 ut 0ing trie largest increm121middot1t middot] bulli~he
bdc kgrnund con cent ration of Hie em~ tted subs tcince i~o ctt tention Wi 11 be ~i ven
to the s~iondc1tY sources witt1middotn cK1 microant or to trlL case of J011ns-Manvi I le
since -UH imiddotajor source is less tran LOO 1byr and is not microrejicted to raise
bctckgro 1wd exposure levels to the ~~ntr21 population Furt12rmore a11
emission sources dt Kaiser Ste 1 1iC(1 v1 1~n directly dedlt -1ith in this
program r1re elltlteci ro t11e cc1lt1bull ovcmiddotr nperatmiddotions These faci 1ities are to be
closed du1rm and all primary reel 11i~ oH~rdtions discontinmiddot 2d Kaiser
forecasters have predicted further deterioration of the plant economics and
the phased closure has been accelerdted for primary steel making operations
Note that this closure is essentially irreversible cince differential
exmicroansion and contraction of the coke oven structur1 occurs in the cooliny
process and it would be improbable that ovens could be reheated without
extensive reuuildiny at major expense
12l STORAGE TANK EMISSIONS - fAUFFrn IHJW OUPUNf ANU ALLIED
At c1ll four sites emissions from storage tanks constitute either the
primary or near dominant (Stauffer) source of carcinogen release andor
genero1i iJOpulation exposure Curreritly the tanks of interest at each site are
permitted by the local Air Quality IJistricts hJwever they do not require
emission ontrol systems for vdrious reasons Tl1e estimated releases grounds
for exemption and other pertinent information are given in Table 121-1
In order to amicromicroreciate tth~ practicdl dlternatives for emission
controls the provisions of SCAWMU Rile 463 ar~ listed below which specify the
acceptable alternatives for tdnks r4uirin9 controls ie tanks 11aving
capacities greater than 39630 yal with suostances of true vapor pressure
176
Table 121-1
Site
Stduffer Uff-Site Tanks (J) (Sdn Pedro)
Allied (El Seyunrlo) - KdJ
1bulllctLer1 d I reld iturctjc
-J -J
Dow (Pittsbury)-Product Check Tanks (4)
Uow - microroduct sturag~
DuPont (Antioch) - Raw Materidl Feed Stordge
Approx Capacity~
6lUS x 1g (c) 84 X 10 (1)
lt3Y630
18000 each
70000L) 3UUOUO
30000 working 42636 liquid full
STORAGE TANK
Substance
Ethylene Uichloride
Carbon Tetrctci1 I Jr i J1
Perchloreshythylene and Carbon Tetrashychloride
CT perc
Carbon Tetracr1 l ori de
SUMMARIES
Emission Mode
Tank breattli ng
Tank v10rk i ng Tank breatttiny
PERC-~JOrk i ng PERC-b r2a t ~ i ng CT-working CT-breattling
Tank breathing Tank breatt1i ng
Tank workin9 Tank breathing
Vamicroor Pressure
014 at 70 F psi
1 63 psi at 68 F
16d psi at 68 F(CT) UJ9 psi at 68 F(perc)
168 i)Si dt 68 F
Grounds for Exemption
Exempt from control re4uirements to SCAQNJ Rule 463 since vapor microressur-2 dues nut iXC~ec
l~ io at sturdJ~ conditions
txe11pt TlJil fule -+b3 since tank capaL ~J ~ s u~der 39630 gal (lSO 11 )
Exempt from Regulation Rule v~~vu u~~~~~shytank capacity is under 39630 ga 1 andor substances are not classified as organic 1i4uicts
Exempt f ro1n Ru l ~ 85300 because subsLctnce is not classified as dn organic li4uid accordin to Regulation-Rule 8120
exceeding l~ microsi at storage conditions
~ floating roof tanks
~ fixed roof tanks with an internal floating type cover
bull a vapor recovery system with vapor collection and retllrn (or disposal) proce5sinJ lXctbullerlin~ 95 efficiency
ln a cllemicai plant a typical vapor recovery systein (KR Evans
SCAQMU) migm consist of a collection inanifold to d recovery system (suctl as a
vapor sphere) to a compression syst2m dnd subsequently to absorption and
recovery systems The absorpt~on st~n c~nsisting possibly of (rubber
stripper or activated carbon
In dealing with speciti( carcinogenic substances such as vinyl
chloride hiyhly efficient vapor control system perf0rmance has sometimes been
stipulated necessitating incineration systems
Fer th~ cases of corccrn realistic alternltltives are as follows
Stauffer Off-Site TanKs - Tnere are a number of options which can 1~arkedly 3
reduce ei1issicns from these tanks from th~ calculaL~d value of over 20 x 10
lbyr One alternative is to consolidat~ the material in tne three tanks
One of the large tdnks can contain the currently stured material and reduce 3emissions to apmicroroximately 13 0 x 10 lbyr Anotl1er alternative is to
transfer dll E0C to a single floating root tank Various types of such tanks
exist dlld are reviewed in the EPA report Lrganic Cht~mical Manufacturing Volume
J Storage Fugitive dnd Secondary Sources EPA-450J-80-025 eg internal and
external floating roofs and a variety of seal desiyn configurcttions Generalshy
ly such designs would be expected to reduce emissions to the order of
one-fourth to one-fifth the current level Cost of 1nstalliny ct contact
single seal internal floating roof wds estimated by BB Lumquist Pentrex
for EPA as $27770 in 1Y8U dollars for 7U ft diameter tdnk Cost to build an
external floating roof tank was estimated by G Stilt of Pittsburg Oes Moines
for EPA as ~14UUUO for D=67 H=4Uft Anottler com111on approach is to utilize
cdroon ctdsorption This works we11 witn nonpolar hydrocarbons dS VvC is
removed from the vapor phase A basic syte111 consists ot tvo carbon beds and
a regeneration system Regeneration is typically performed with steam or
vacuum Figure 121-1 illustrattS rhe systems (Basd2kis 1~110) Steam raises
tne VOC vctpor microressure The resulting sti~am-VOC mixture is condensed and
173
AC-TJA~D CA-e0-l
-----~---
LOW- ffiESiJ~C ~T~H----
PiltOCESS COOL-J~ At-JO oLOWER DRY t-J~ 2gtLOWE~
COgtJDENSER
J
COOLIN(a - --------
WATi=-A
RECOVeRE SOLVENT
WASTE WATER
Fi l 1 1 middot gure middot - Activated-c arbon Adsorption System
179
rout2d to c ~epordtor decanted and returned to storage In vacuum regenerashy
tion VOC vapor is desorbed by pu1 ling a vact1um on the carbon bed then condens-
ed a1d returned System efficiency ~ ~ est ii1a ted at 96 f(duc middot i m from fixed
roof levels (EPA 4503-80-025)
H~fri 9ctated vent concicr1sers dre une of the most cu1mon emission
reduction prJcesses -for conttol ~ ng hx2-d-- storage tai vUC Figure
12 1-- 2 i l 1ust rates a u n i t ( Er i scmiddot n 19HU ) eff i c i en cy of middot middot~ ~ c middotlt middotmiddot _y i s rate ct
bet1Jee~middoti 60-90 for the vapor pressure range of concern For such a large tank
the capital costs would be high ~igure 121-3 illustrat~s EPA estimates
(EPA l ~~SOb) for the condenser section Condenser sys tern a ea mu l d be in
excess of 1000 ftL
It should be noted t -at Jtc1 uf fer Uiemi cc1 l has arH1ounced the c 1 osu re
of its VCPVC plart PrevfoJsly tt12) had planned to purge tlle off-site tanks
of EUC Since any future po~sible micro1ant stdrt-umicro will necessitate a compreshy
hensive SCAQMD review this ltoument can assist in evaluating proposed control
measures
Allied and DuPont Feed Tanks - In bott1 of these cases the nore significant
quantity of emissions arise from tank workinltJ ie during the off-loading
activity rather than tank breathing Contnl ~easures taken for working
emissions are thus of primary concern Ther~fore no detailed discussion will
be provided on the alternatives for control gtf bredthiny emissions Additionshy
ally both tanks are in the range of 20U00 gallons which is a capacity where
floating roof tanks are almost nonexistent Out of 670 floating roof tanks
surveyd by EPA less then 15 were smaller than 30000 gallons in capacity
Therefore the utilization of any floating roof concept and seal combination
will not be considered
Commmerc i a11 y it is estimated by Ou Pont that carbon tetrachloride
feed costs approximately $017lb (E Taylor personal communication)
Ttierefore less than i2500 is lost to l)uPont annud l ly as a result of working
l eve l em i s s fo ns and 1es s fr om Al l i ed bull Thus i t is u n l i kel y that any
appreciable economic incentive exists to dev1~lop ci vapor recovery system
However a Cdndidate system could be pattern~d after the chloroform
feed-storage unit currently at Allied This is a dedicated vapor balance
system Chloroform is off-lndded from tank cars and stored in a closed
unvented tdnk As the 5tordye tank is filled the c1ir spdce d1spldced is fed
Figure 121-3 Installed Capital Cost vs Condenser Area for Various Materials of Cunstruction for a Complete Condenser Section
182
back to the tank car Also in this case the storage tank is not vented in its
normal breathing mode and is part of a closed feed system to the reactor
Vapor recovery system alternatives which are particularly effective
in loading and handling include r~frigeration adsorption andor absorption
Control efficienci~- 1re est1mateJ in the rlt1nge of 90-95 (EPA 1980b) and of
course depend on the speci tic su1)Stance and equipment used Carbon adsorpshy
tion systems were discussed above The smallest capacity carbon adsorptionmiddot
system priced by EPA (EPA 1980b) has 2 vertical beds of carbon (900 lbs - 4ft
diameter by 3 ft depth) witl1 an ilstalled cctpital cost of $135000 based on
December 1979 dollars
Dow Tanks - Dow has two pair of p~oduct check tanks which alternately are
filled dnd off-loaded Emhsions from th~sl tanks were calculated based upon
operating cycle (3 day fi 11) satubull-dtion vapor pressure and physicdl
chardcteristics Direct hec1d-spa1e testing was planned however it could not
be accomodated because of r 1stri c ed access due to unscheduled mai ntendnce on
the field test day
Dow has indicated that chey are studying the option of installing a
vapor controlrecovery systt~m in 1hese and their largerproduct storage tanks
The extent of their engi neri ng ind assessment work is unknown as are their
current plans It may be possibk to incorporate a vapor balance design into
the system whereby the disp1aced apor is trdnsferred to another process point
within tl1e system Alterna1ively a vapor r1middotcovery system such as carbon
adsmiddotorption is fedsible Hot-1ever since emissions for t11e check tanks are
primarily due to working loss tht~ use of a conservation vent or an adjustment
of its operational differential p~essure would be ineffective for tr1e
reduction of the bulk of such emissions Furthermore since check tank size
is re1at i ve 1y sma 11 no cons i derctt I on was given to conversion to a fl oat i ng
roof configuration for those tanks Conversion would be possible for
the large storage tanks
12~ WASTEWATE~ EMISSIONS - STAUFFER
Emissions of me throug11 ~-a~)te-1dter discharge are the largest single
source identified at the plctnt sites The discharge limit was set at 25 ppm
by the Los Angeles County Sc1nitation iistrict and was estdblished with the
occupational limit in mind of 50 ppm JVer an 8 hour work shift dt the District
18~
tredtrient center~ T11ere hctd been suine di~cuss1on between pdrti~s of possibly
raisiny tne discharge limit since it is felt by Stauffer that dilution of the
stream is sufficient to d 11 ow it Itmiddot shou Id he noted t10-1ever tna NIUSH has
recommended the permissible exposure limit be reduced to 5 ppm averaged over a
work period of 10 nours microer ddy 40 ours per Ieek witt1 a cei l iny level of 15
pmicroin dVeraged over a 15-mi nute microet~ v(J
It is not certain at 1~hrt rdte me 1rill be reledscd from the
wastewater stream At the microlant discharge points wastewater is both hot and
aerated thus favoring release~ No measurements have been ta~en downstream of
the plant after considerable dilut1on has occured The distance to the water
treatment pa11t is approximately 5 kin at 1i1ic~1 point anerobic diyestion is
conducted lt is presumed that a11 ELlC will be released before final ocean
di scnar~Ji
A possible emi ss i D~ cont ro process for nmiddotduct ion of the EDC in the
effluent is by ~recess adjust~Pnts or additio~s For example process
modifications to the EUC stripping stage coula dramatically reduce discharge
lev~ls A control alternative is the use of octivdted carnon or XA0-2 resin
to recover EOC in tne discharge stream Tests of Gulf South Researd1
Institute on XA0-2 and activated carbon (Coco 1980) show high recovery yields
for nonpolar organic carcinogens unaer a range of concentrations Viable
suggested alternatives oy Srnitn included regeneration of Ue trapping
materials dnu even consideration of burning tt1e concentrate carbon media (at
greater than 1000 pmicrom)
Control approaches to reduce emissions from so called secondary
sources in general and waste water emissions in particular fa 11 into four
categories - waste source control resour(e recovery alternative disposal and
add-on controls Alternative control pro(~esses which were considered but
appear to be inappropriate to this case i11clude chtmiddotmica means eg
neutralization precipitation coagulatior1 and chemical oxidation thermal
destruction of tne unconcentrated ~dste s1ream is ii1practical biological
treatment eg aerdtion and biomass-wasteater contcct ~ienerally relates to
the treatment of soluole degrddable organics in the concentration range
bet~~een U01 and l~~ terminal storagt eg lanctfilliny urface impoundment
and deep well injection dre either indppl 1cable or 1mmicroractical Therefore in
summary the poss i b I e contra l approac ties fur this ca e inc 1ude the improvement
r
184
of semicrodration efficiencies in steam stripmicroing tl1e internal recycle of ~aste
stredms ctnd the ddsorption b activdted carbon The design configuration
~fficiency and cost )bviously depend upon numerous microlarit specific factors and
their determination vJOuld rec-Jire detailed analyses
n1e ~~dsrewctter systbull-m of a model cl1emical production plant based
upon the average properties uf a composite of 30 chemicals was evaluated for
EPA by IT Enviroscience (EPA 1980b) Included prominantly among the 30 was
tUC with the hignest uncontrolled secondary emission wastewater release rate
ie ~ percent of tt1e production and 34 perc~nt of the emission Cost and
impact analyses were evaluated for alternative control systems to reduce
secondary voe emissions from wastewater Four systems were considered a
carbon adsorption system (eAS) for recovery of the voe from the wastewater a
cover to reduce secondary VOC emissions from the wastewater clarifier a cover
for the clarifier plus a carbon desorption system and a cover for the
clarifier plus ct eAS system usinu a fume incinerator The scale of the model
system was greatly in excess of the Stauffer plant thus further making
detailed comparison i mpract i cal bull However emission reduction factors ~-1ere 6diven as ~9 and cost effectiveness per 10 g reduction genera1middot1y ranged
between $450 to 1733 These factors would likely grossly underestimate the
system cost if scaled down to the range of the Stauffer plant ie the order
of 34600 lb annual discharge
The alternative approaches of improved steam stripping efficiency
and internal recycling of the stripper iischarge stream with a reduction in
makeup re4uirements could decreas~ net ~DC wastewater content release
123 CONTROLS FUR SECONDARY LEAD S~1tLTER STACK EMISSIONS
Given our findiny of low (16 kgyr) ~missions of arsenic from the
reverberatory furndce at RSR it middotJOuld ippear that the arsenic content of themiddot
lead feedstock is low andor that the microants system for reducing lead
emissions is dlso quite effective for arsenic ~SR it will be recalled from
Section 32 uses a quenchinu cnariber ctnd Dayiwuse filters to remove
pctrticuldte matter and a carbonate scruooer to remove sulfur dioxide
There are no federal new source microertormance standards for lead
eini ssions per se t10wever the Standdrds of Pbullrformance for Secondary Lead
SmeHers (40 CFR 6Ull2) limit total particuLte emission from a blast furnace
185
or reverberatory furnace to SU mgm 3 According to Augenstein et al (1978)
who reviewed the techno1ogy for control ing lead ~missions from ttwse sources
baghouse filters or wet scrubbers are yenerally used to contro1 particulate
emissions When fabric filters are used to control blast furnace emissions
they are nurma11y preceded by an afterburner which inciner~tes hydrocarbons
that bull~GU Id otnerwi se b 1 ind t ne f o1 ur-i c After-burners a re nor necessary for
reverb2middotmiddotatcry furnace emiss-ion smiddotnce the excess air and temperature are
usually sufficient to oxidize tn2 hydrocarbons
According to Augenstein et ali 11 sr1aker--type baghouse filters are
the 1nost effective means of controlling ead fume emissions from secondary
furnace opErations 11 Collection etficiencies can exceed 99 oercent One
advantage to this control approact is that l~ad oxide dust can be recovered
easily cmd recycled in the smeller Flue gases must be coated to below 300 degF
for dacro~ bags and to below ~00 degF for fiberglass bags (Hi~h et al 1977)
Although wet scrubbers are effecti1e under some circumstances in
controlling lead emissions it is more difficult to recover the lead oxide for
recycling In addition sulfur dioxide presi~nt in the flue gases becomes
oxidized to sulfuric acid and can cause corr1sion problems For this reason
sodium carbonate or other basic reagents ar~ added to the scrubber solution
Although it concerns a yold smelter an approach described by
Marchant dnd Meek (1980) provides an exampli~ of an arsenic control
alternative which might be apmicrolicatgtle to secmdary lead smelters At the
Campbe1 1 Red lake Gold Smelter in Balmerton Ontarion Canada Smelter gases
are first passed throuyh a hot electrostatic precipitator (ESP) The ESP is
heated so that the arsenic (which is principally in the form of As ) remains2o3 gaseous and is not yet collected This exclusion of arsenic allo~~s the ESP to
recover particulate gold rnore easily The ESP exhaust is then quenched with
ambient air to condense the arsenic trioxide Baghouse filters then remove
the arsenic along with other particulate matter
124 CUNT~OLS FUR STEEL MILL EMISSIONS
Given the imminent and irreversibli~ cessdtion of coking activities
at Ka 15 er Stee 1 Corpora t i on a rev I e~ o r t e c Imo 1 o gi es for ( on t r o 1 l i ng
~missions from the coke ovens rnd he Clt1ke byproduct recovimiddotry pldnt WdS not
deemed to be necessary Tnis is t1e only primary steel mill in California
1Pi6
KEFUENCE)
Al lf~n Jr Ll~ llJBU1 J 111odel to 1middotstimite hi1drdous (bullmissions from coke oven Juurs pr~pdrelt1 DY 1-kSecircn Tr1dnye7nrffut fur U~ Environmental Protection Agency Uffi ce of Air ~ua 1i ty and Standards fltesearch Tri ang I e Park Nortn Cdrolina KTl17362-UL
Jl len Jr CC 1~8Ub tstination of chdr in(i emissions for by-product coke oven microrepared by fltesedrch Tr1anglt Insc1tute for US nvironmental Protection 1-yency Uf f ice of Air l)ua l i ty and Standards Ke search Triangle Park North Cdrolina RTI17362-0
Allen Jr CC 198Uc 11 Environmeritdl c1ssess1ent of coke by-product recovery plants Proceedings First Symposium on Iron c1nd Steel Pollution Abatement Technology Ctncayo Illrno1s--Tin1ctober - 1 November 1979 EPA-600-9-80-012 pp 7~-dd
Augenstein ur1 T Cornin et al 1978 Con_trol techniques for lead air emissions prepared by PEDCU Envircnmental Inc for US Environmental Protection Agency Research Triangle Park North Carolina EPA-4502-77-012-B ( NT IS P88U-197551)
jarber WC 1977 Interim air pcllution a~sessment for eight t1igh-volume industrial organic chemicals 11 US Environin~ntal Protection Agency memorandum to Air and Hazardous Materials Uivision Kegions I III-X and to Uirector Environmental middotPrograms Uivision Region II
t3drrett KE WL Margard et al 1977 Sampling and analysis of coke-oven door emissions Prepared by ~dttelle-Columbus Laboratories for US Env 1 ronmenta I Protection Agen y Industri a I Environmental Research Laboratory Research Triangle Park North Carolina EPA-b002-77-213
t3 a r ret t flt bull r bull and P bull K bull ~~ebb 1Y78 bull 11 E f f e ct i venes s of a wet el e ct r o stat i c precipitator for controlling PUM emissions from coke oven door leakage Presented at 71st Meetiny of the Air Pollution Control Association Houston Texas 25-29 June 1978
l3asdekis HS 1980 Carbon Adsorption Control Uevice Evaluation IT Enviroscience Inc report in prepdration for US Environmental Protection Agency ESEU
8ee fltW et al 1974 Coke oven charging emission control test program Vol I c1nd II US Environmental Protection Jgency EPA-6502-74-062
~eimer RG 1978 Air pollution emission test Analysis of polynuclear aromatic hydrocarbons from coke oven effluents Prepared by TRW Uefense and Smicrodce Systems Groumicro for US tnvirorunental Protection Agency Office of Air wuality Planning and Standards Kes~arch Triangle Park North Carolina EMB Kemicroort 8-CKU-12
187
Braman RS 1976 Application of the arsine evolution methods to environmental analyses presented at the International Conference on Environmental Arsenic Ft Lauderdale Florida 5-8 October
Bratina J E 1979 11 Fabric filter applications on coke oven pushing operations J Air Poll Cont Assoc 29(9) lll6-920
tiuonicore AJ 1980 Environmental assessment of coke quench towers in proceedings First Symposium on Iron and Steel Pollution Abltlternent Technology Chicago~ Uirnois JO October - f N~vernber 1979 EPA-6009-~1b-62~ pp 112-142
t3usse A D and dR Zimmerrao1 1973 User 1 s guide for tdegh~ Climatological Uisper~ion Model US Environmental Protection Agency Resea~ch Triangle Park Nortl Carolina) EPA-R4--73~0024
California Air Resources 8oard 1976 Coke oven emissions miscellaneous emissions and their control at Kaiser Steel Corporations Fontana Steel making facihty Enforcement l3ranch) S2crcmento California L amp E-76-11
Calvert~ CA 1976 Envir __ Sci_ i( __ Technol 10256
Coco L~i 2t aL~ 1979~ 11 02c10pmert of trec1tment and control technology for 11refractory pet rochemi co was tt=1s r middotl 1d l Report Gu 1f South Research In st it ute
to U S tnv i ronmenta l Protection Agency EPA-ti002- 79-080
Colucci~ LM and CR Begemi 1971 Palynuclear aromatic hydrocarbons and other pollutants in Los Angeles air In Proceedings of the International Clean Jir Congress Vol 2 Academic Press pp 28-35
Cooper JA and JG Watson Jr 1980 Receptor oriented methods of air partkulate source apportionment 11 J Air Pollution Control Assoc 30(10) 1116-1125
Coutant R W JS McNulty and RD Grammar 1975 Final report on determi 11at ion of trace elements in a combustion sys tern Prepared by Batte11 e Columbus Laboratories for the Electric Power Research Institute EPRI 122-1
11 11Dilling W 1977 lnterphase transfer processes Envir Sci TechnoL 11 4 pp 405-409
Oow1irgl) MP JO Jeffry and AH Laube 1978 11 Reduction of quench tower emissions Presented at the 71st Air Pollution Conotrol Association Conference Houston Texas 25-30 June APCA No 78-92
EPA 1980a Review of Criteria for Vapor Phase Hydrocarbons 11 US Environmental Protection Agency ~eport 6008-80-045
EPA 1980b Organic Chemical Manufacturing Volume J Storage Fugitive and Secondary Source EPA report 450J-80-025
EPA 1981 Compilation of Air Pol lutdnt Emission Fdctors (Including Suppleshyments 1-7) Third Edition Suppl1111ent 12 Utfice of ir l)uality Planning dnd Stdndlt1rds AP-42-ED-3-Supp I - lt
188
Erikson Uli 1~80 11 Control Uevice Evaluation Condensdtion 11 IT Envioscience lnc report in prepdration for US Environmental Protection Agency ESEU
lrtel liL 197Y Quench tower particuldte emissions J Air Poll Cont Assoc 2Y(9) 913-916
Ev a n s K bull P bull 1981 Pe r son a l Commun i cat i on v i th R bull Zi s k i n d bull
Goodwin DR 1980 11 Air pollution emissions standards in Proceedings First Symposium on Iron and Steel Pollution Abatement Technology Chicago 11 lrno1s 30 October - 1 November 1979 EPA-6009-80-012 pp 37-45
Gordon GE 10 Receptor Models Environ Sci Tech 14(7)792-800
Gordon KJ 1976 11 lJistribution of airborne polycyclic aromatic hydrocarbons throuyhout Los Angeles Envir Sci Technol 370-373
Great Lakes Cdrbon Corporation 1977 Study of coke side coke-oven emissions~ Vol 1 Source testing of a stationary coke side enclosure EPA-340l-77-014A
lirimsrud EP and HA Rasmussen 1975 Atmos Envir Y 1010
Hartnan MW 1~80 Source test c1t US Steel Clairton Coke Ovens Clairton P~nnsylvania Prepared by rRW Environmental Engrneerrny D1v1s1on for US lnv1ronrnental Protection Agency Emissions Standards and Engineering l)ivision Research Trianyle Pdrk North Carolina EM~ Report 78-CKU-13
Hendriks Ilt V et a I 1Y7Y Urgani c air emissions from coke quench towers 11
Presented at 2nd Arrnual M(etiny ot the ir IJol lution Control Associdtiont Cir1cinndti Utlio June 1~7lJ l-dmicroer NU 9-391
Higt1 MU ME Lukey dnd TA Li Puma llJ77 Inspection manual for secondary lead smelters premicrolt1red b Enyineeriny Science lnc for US lnvironmentdl Protection Ayency Office 0f ~nforc~nent EPA-3401-77-001
Holzer G et al 197 11 Collection amp analyses of trace organic emissions from natura I sources 11 Journal of Cnromatogra flhY 142 pp 755-764
Science Applications lnc Final lemicroort to US Environmental Protection Agency tPA-560lJ-80-UU2
J bull Mi l n e l lJ 81 Person a I Cornm un i cat i on wi t t1 bull Zi ski nd bull
Kemner WF 1979 Cost tffectiven2ss model for pollution control at coking facilities Premicroarect oy PEDCo Environmental lnc for US Environmental Protection Agency Industrial Environmental Research Laboratory Research Tridnyle Park Nortll Carolind EPA-GU02-7l-1U5 (NTS PB 8U-118706)
Kessler T AG Sharkey Jr and kA Friedel 1973 Analysis of trace ele111ents in coal by smicroark-source mass spectrometry US 8ureau of Mines Report ot I n v es t 1 gd t 1on s No bull 77 14 bull
189
Kolak NP J Hyde and R Forrest~r 19 1 9 Particulate source contributions in tt1e Niltlgara Frontierffi Prepared t)_y Nr2bulll York State IJepartrnent of Env i ronm~ntc i Con serv d ti on for US En vi r inment ct l Protect ion Agency Region 11 EPA-9024-79-006
Mackay U and PJ Leinonen llt175 Hate of Evamicroordcion of low - solubility contaminants from vJater bodies to c1tmospmre 11 Envir Sci Tecnnol 9 13 pp 1178-libU
Magee tiv HJ Hctll and GM Vc~rlt_a Jr 1973 Potential Doimiddotiutants in fossi1 f1Jel Prepared by ESSO fbullt~scarcr middotnd Engineeriny Co tor US Enviromihm~ct Protection Agency Ufl-R2-7 --2l1~L (NTIS Pt3 as u~9)
Marcriant middotG0H ard RL Meek 1~gu Eva uation of technology for control of arsenic tmissfons at the CdlmpJel Hd Lakmiddot G~ld Smelter prepared for the US
1Env i ronmentd l Protection Agenc)-l-l(lU-sfrT l lnv i ronmenta l kesedrCiI Laboratory Cincirrn2ti Uh70 lYA-60U2-8U--141 (NTIS 8 oU-21Y36J)
Margler L$ HA Ziskind M8 lcgozen (Ind M Axelrod 1979 11 An inventory of care i nog)n i c substances re i ea sect into ne ambient air of California Vol une I - Scn~enin~ and id2nt~fication o-f r~1rci1ogens of greatest concern Science J- pp 1i ca middott~ L J s In c bull Fi nd l kcmicro~ rt ~ i I - U 6 C) - U- 5 0 4 to U1 e Ca I i ~- o r n i a Ai r kesources 130ard
Milne J 198L Persond1 C~mmunicdtmiddotion middot1itt1 R Ziskind
llurchio JC WrC Cooper and A ue Leo11 1970 Asbestos fibers in ambient air of California 11 Cal Horn i ct Air lesourc2s l3oa rd ARl3 4-0S4- l EHS Report 73-2 (March)
National Academy of Sciences 1Y72 Part1culate polycyclic organic matter Washington~ DC pmicro 4-ll
Nati ona I Kesedrcn Counc i I 197b Arsenic prepi1 red by Subcommittee on Arsenic Co111mittee on 1bull1edical ctnd t)ioloyicai Effects of Environmental Pollutants washin~ton uc
Oliver JF and JT Lane 1979 11 Control of visible emissions at CFampl 1 s coke plant-Pueblo Colorado J Air PolL Cont Assoc 29(9) 920-925
Pe I I i z z a r i E bull 0 bull anct J bull E bull l$ u n c h 1s 19 1 nbi en t ai r ca r c i n o gen i c v a po r s Improved samplin~ and analysis tectrnology EPA Contract 68-02-2764
Pellizzari E U 1977 11 Tle 11easure11ent of carcinogenic vamicroors in ambient middotatmospheres EPA-b007-77-J55
Phelps RG 1Y79 lSI-CPA-oatLlle c-ilte oven door sealing proyram J Air Pol 1 Cont Assoc 29(J) JlJ8--JlL
Porter RA 1976 Uismicroersion ~yuation s lucions Dy calculator I guide for air pollution engineers ctnd sci12nthts S cond edition rexas Air Control tioard ~MT1S P~ 262 lJ~)
Redburn T 1980 Kaiser battles aginst l~gacy of woes 11 Los Angeles Timesmiddot (7 Semicroternber irno) Part VI pp 111
Research Triangle lnstittJte 1977 11 Qudntification of benzene in 15D -1bullnoient air samples 11 for US Environmental 11rotection Agency Office of Air Quality Planning a11d Stitndards
Rinkus SJ and MS Legator 1979 Chemical characterization of 465 known or suspected carcinogens and their correlation with mutagnic activity in the Salmonella typhimurium system Cancer Research 393289-3318
Koberts R M 1980 11 An inventory of arc i nogen i c substances r~ leased nto the ambient air of California Tasks 11 and IV KVB Final Report KVB 26900-836 to Science Applications Inc
Rogozen MB LW Margler P Mankiemiddott1icz cind M Axelrod 1978 Water pollution control for coal slurry pipeiines Prepared by Science Appl1cat10ns Inc for US Department of Energy (NTIS SAI-068-79-516
Rogozen MB DF Hausknecht and RA Ziskind 1976 Methodology for ranking elements in fossil fuels according to their potential health impact Science Applications Inc Final Report 260-77-539 to the Electric Power ~esearch Institute
Russel 1 JW and LA Shadoff The sdmpl ing and determination of halocarbons in ambient air usin~ ~rncentration on porous polymers
Siebert PC CA Craig and Eb Coffey 1978 Prelimary assessment of the sources control and population to airoorne polycyclic organic matter as indicated by benzo(d)pyrene Prepare( by lnergy and Environmental Analyses Inc for US Environmental Protection Agency Office of Air Quality Planning and Standards Resedrch Triangle Park North Carolina
Simmonds PG et amiddot1 1974 Distribution of atmospheric halocarbons in the air over Los Angeles basin Atmos Envir Vol 8 pp 209-216
Singh HB LJ Salas and RE Stiles Distribution of Selected Gaseous Oryanic Mutagens and Suspected Carcinoyens in Ambient Air in Proceedings Annual Meeting Air Pollution Controrl Association 82-651 June 20-25 1982 New Url eans
Smith WM 1979 Evaluation of cok oven emissions in Yearbook of the American Iron and Steel Institute pp 163-179
SRI International 1980 Asessment oi human exposures to atmospheric benzene SJ Mara and SS Lee finai report to US Environmental Protection Agency EPA-4503-78-031
Taback HJ 1978 Control uf Hydrocarbon emissions from stationary sources in the California South Codst Air Basir Prepared by KVB Inc for the Cal1fornid Air Resources Bod-rdmiddotKVB No 5804-714
191
Trenholm J-f dnd LL )eek 1Y78 11 Assessment of t1azardous organic emissions fror1 slot type coke oven Datteries unpublished paper US Environmental l)rotection Ayency Ernssion Standdrds and Engineeriny Jivision
Turner U~ J970 Workbook of atmusphe~ic dispersion estiimiddot1Jt2s IJS Environmental Protection Agency k~sedrch Tri1ngle Park North Carolina
US Bureau of Mines lYl f-inercls yearbook 197U Vol 11 p 4~3
Van Os dE I l U bull iJ bull lL Ma rs l and et ct 1Y7~1 o En i r onmt~ nta I a s s es smen t of co k e Dy-product recovery plants prepareu by ReseMcn Trictngle lnstitute for US Environmental Protection AgerKy 11cJustrial Environ11ental R~search Laboratory Kesearcll Tri2n~e Idrk NorU1 Caroline EPA-6U02-7J-Ul6 (NTfS PB 293-278)
Westbrool( C W 1979a Level l ass~ssment of uncontrolled sinter plant emissions Prepared by fesearch Triangle Institute for US Environmenta1 Protection Agcncy Industria1 tnvirnnmental fltesearc-1 Laboratory kesearch Trianyle Park North Carolind EPA-bOU2-7Y-112
Westbroollt C~~ 197ltJb Level l css~ssment of uncontro1led l)-BOP emissions Prepared by Researcn Triangle institute for US Environmental Protection Agency Industrial Environmental Hesearch Laboratory Research Triangle Park ~Orth Carolina EPA-6002-79-190
Wetherold rLb LP Provost and CD S111ith 198U Assessment of atmospneric emissions from microetroleu111 refinin~ Volume 3 Appendix l3 Radian Corp Fina Report to US tnv i ron111enta I Protection Agency EPA-600 2-80-07 5c
Ziskind RA OF S1ilith JL Hdhn etnd G Smicroivey (1980) Determinants of Cancer and Carctiovasculdr Uisease Mortality in Asbestos Mining counties of California SAi Report 068-dl-514 l May
192
APPENDIX A
Rapid Screening and Identification of Airborne Carcinogens of Greatest Concern in California
Lawrence W Margler Michael B Rogozen Richard A Ziskind and Robert Reynolds Science Applications Inc Los Angeles California
This paper describes a method for establishing a priority list of airborne carcinogens within a
state jurisdiction In this case it was necessary to identify from among hundreds of potential
candidates the live to ten materials of greatest potential concern In California as airborne
carcinogens
Because no previous Inventory of carcinogens In California existed published lists rankings
and assessments of national scope were used to Identify candidates By systematic manipulation
and comparison of these data sources 47 materials of some notoriety were chosen for closer
scrutiny This selection was pared to 22 candidates largely by eliminating those which had
very little production and use In California (Substances primarily used as pesticides were
excluded from the scope of this study) The remc1ining candidates were then ranked by additive
and multiplicative algorithms and by a panel of experts The results of these rankings were
combined to produce a single selection of 11 priority candidates In alphabetical order they
ethylene dichloride N-nllrosoamlncs perchloroethylene and polycyclk aromatic hydroiarbons
In continuing studies a baseline emissions inventory Is being prepared and a source testing
program is being designed
In recent years concern has grown over su~pected arcinug-ens New Jersey seshythe possibility that certain materials lected ten volatile organic compounds released to the atmosphere through inshy and five htbullavy metals to be examined dustrial and commercial activity may be further and is currently measuring responsible for a significant portion of ambient atmospheric concentrations of the incidence of cancer in the general these substances in a variety of areas In population This concern is manifested the second year of the study the state at the federal level in the National h1s incre1sed the volatile organics Emissiltm Standards for Hazardous Air stt1died to 20 and begun measuring Pollutants (NESHAP) which limit heavier orianics associated with parshyemissions of the known carcinogens asshy ticulatcs2 In California a very different bestos beryllium and vinyl chloride 1 approach-was taken
Only a few states including New Jersey and California have begun efshy Overview of Californias Approach forts to identify airborne carcinogens of concern to the general public for the The California Air Resources Board purpose of setting state emission regushy (( ARB) is sponsoring a three-stage lations for these substances After reshy st 11d~bull of airbornC rmissions of carcinoshyviewing national use data for known and g1middot11s from anthropogenic atlivities The
November 1979 Volume 29 No 11
first stage which is the subject of this paper was to identify roughly five to ten materials which of the hundreds of known or suspected airborne carcinoshygens are most likely to be of greatest concern to Californias general populashytion Also of interest were those which in order to satisfy occupational health and safety regulations might be transshyferred from the workplace air to the outside atmosphere The second stage which is now underway includes pinshypointing of emission sources for each of the carcinogens of importance quantishyfiltation of emissions nnd design of source-testing methods A subsequent stage will consist of source testing and measuring public exposures to those substances for which data are unavailshyable The basis for regulatory action if appropriate will include the results of this program and other related reshysearch
Screening of Candidate Carcinogens
The National Institute for Occupashytional Safety and Health (NIOSH) lists 1905 chemicals which have reported neoplastigenic or carcinogenic effects and 510 which have otherwise received attention for their neoplastigenic poshytential1 The need to select five to ten materials from such a large number of potential candidates dictated that we devise a way to rapidly eliminate from further consideration the vast majority of the substances Given the paucity of published data on most of the candidate substances the screening method was designed to make best use of readilymiddot
C1111yrich1 19i9-Air Pullulion Control A~~ociatlon
1153
A- I
----- ---- ----- ---------------------available information The screening process was as follows (1) eight comshyplications of known and suspected carshycinogens were reviewed and those subshystances which were not used in Califorshynia were highly unstable in air or were very doubtfully carcinogenic were eliminated (2) after more detailed inshyformation was obtained for the reshymaining 25 substances candidates were rated by two different analytical methshyods (3) an expert panel was convened to review dossiers on the candidates and to independeitly nmk them (-i) from the eight to deven substances ranked highest by all three approaches eleven were selected for the emission identifishycation and source-testing design stages of the CAHB effort
lnitiI $cre~ng ot Potential Candidate Carcingtgens
65 compnunds middotwe ire selected from 642 industrial ur~nnic air pollutants comshypiled by MrTRE Corporation for the US Envircmnental Protection Agency (USEPA) in U1at study pollutants were scored by multiplying four explicshyitly defined rating factors annual US production fraction of production lost to the environment volaflity and toxshyicity To adapt this york to our pmpose we first selected the 114 substances listed as being carcinogeni 1~ or neoplasshytigenic Then the scores of each of these compounds under the criteria annual US production fraction of producshytion lost volatility and carcinogeshynicity were multiplied together Seshylected for further consideration were those substances which had a product score above 50 Another 15 substances listed as being carcinogenic but lacking information for one of the other rating factors were also selected This list of 65 was then compar~d with seven other lists of corcinogeris1
-middot 11 Materials comshy
mon to the reduced MITRE list and at least one of the other lists were chosen for further consideration Added as candidates were those suhstances which are regulated as occupational carcino-
gens by the Occupational Safety and Health Administration (OSHA) and certain inorganic carcinogens Finally substances were added which in our judgment should be investigated but had been eliminated at this point Exshyamples of these are bis(chloromethyl)shyether epichlorohydrin and hydrazine
Next the refined list which now contained 47 substances or chemical groups was pared further hy another rapid screening process Eliminated
were all candidates (1) whose production andor use in California was very low (under 10~ lbyr) and wns not thought likely to pose a risk to a localized popushylation (2) which are very unstable in 1ir or CH whilh should not on t lw hnsis of CUtrltbullnL cidlnre liebull considtbullnmiddotd r11rci-
1154
Tahllt I Suh tances n-iewed in letail
Candidate Substances
Arsenic Inorganic lead Ashestos Alkyl lead Benzene Maieic anhydride
nogenic The result of the initial the rating under that criterion and the screening was a list of 22 candidate mashy corresponding criterion weight The terials which is presented in Table I overall rating for pollutant i is then the
sum of the scores under all the crishyteriaRanking Candidates by AddiUve ancll
Mulllpllcatlve AigorUhms The additive approach has several virtues the main one being that it forces
Many screening or ranking systems the user to make all assumptions exshyfaH into one oft wo cat~middotgories additive plicit In the process of setting up such and multiplicative Some systems are a a ranking system new insights into the combination of the two while others problem under consideration may be combine nn ol1jective appronc 11 with g1ined Once the system is set up it is suhjlctive cvnluition of the remiddot ults 1l rdatively easy to use Where data for The L2 substances surviving th1middot initial scoring pollutants are unavailable arshyscreening were ranked independtmiddotntly by tificial scales can be constructed to the tvo approaches If the same subshy quantify subjective factors Finally the stance rated high~middot under both systems sensitivity of the results to the systems its importance to California was judged subjective aspects may be measured For to be more likely than if it had scored example one can determine the effect of highly in only one method changing criteria weights upon the final
Additive Approach In the additive pollutant ranking Similarly an appreshyapproach the user identifies one or more ciation may be gained of the significance criteria and rates each alternatiw subshy of the range of uncertainty for a particshystance against each criterion while sishy ular required data element by varying _ multaneously deciding the relative imshy rating factor values portance of the criteria Eq (1) shows its A fundamental problem with the apshymathematical formulation proach is that there is no logical basis for
adding the individual scores assignedRating for pollutant i = f W1 R11 under the criteria other than the asshy
rbullI sumption that this simulates or even
(1) improves upon the users thought proshyEach criterion or rating factor (1 ) is crss A major operational problem is assined a value for each pollutant and that of weighting the criteria A common each rating factor is weighted (b W1 ) practice is to give all critNia equal -icnrding to its importance reht veto wltgtight hut that is in itstbulllf a st1Ubullm1middotnl lhe otlwr rritnia llw sc11rtbull for 1111 aLnut the 1rcl11tive importance of Ow 1n1 1 undmiddotr nitn1 n j is the prod11middott of n1ttria
Journal of the Air Pollution Control Association
A-2
---------
Multiplicative Approach In the multiplicative approach the rating for each alternative is the product of the rat ihgs under each criterion
Hating for pollutant i --- 11Ill
UJ (~) Jbull I
A multiplicative approach can have ~omc altlvrntnges over additive onrs First in some cases the ratings can be physical parameters such as concenshytrations or volatilities there is then no need to weight the criteria and hence less controversy over subjective judgshyments Second multiplication generally provides a wider range of scores than does addition allowing clearer disshycrimination among alternatives Finally this approach provides results which are more intuitively acceptable As an exshyample of this last point suppose that exposure and harmfulness levels for candidate substance are each converted to values on aO to IO scale and that a certain substance is both extremely toxic and extremely rare An additive approach would give the compound a rating ofO + 10 = 10 which is equivalent to that of a moderately prevalent subshystance (rated say at 5) which is modshyerately harmful (rated also at =-) A multiplicative approach on the other hand would rate the first substance at 0 and the second at 25
Criteria The six criteria used in the additive and multiplicative ranking procedures are defined in Table 11 Asshysignment of values to the Rij was based upon data gathered from published litshyerature personal communications and panel discussion and has been fully documented 13
Because the purpose of this exercise was to determine the relative imporshytance of the suspected candidate carshycinogens R 1 was scaled to the most heavily used candidate substance benshyzene whose annual production and use in California is nearly 109 lb Materials with a use under 105 lbyr would be rated zero for R 1 and rejected Howevshyerbefore rejecting a substance by this criterion we considered whether its emissions could in particular circumshystances result in high exposures to a Joshycalized population
R2 takes into account the fact that the chemical industry is in continual change Substances of concern today may be phased out while the use of others may rise dramatically increasing their imshyporta nee asmiddot poll utan ts Information on developments which could likely result in a change in the growth rate was facshytored into the choice of a value for R2- As an example asbestos consumption in California has been stable in recent years However the pending phaseout of asbestos in motor vehicle friction materials will hasten the decline in asshybestos consumption hence we assigned a value of 1 for R2bull
November 1979 Volume 29 No 11
Ideally one cou1 1 use pouu1c11u
sion as a criterion However in this case emissions were to l e estimated in detail only for the five t1 ten carcinogens fishynally ~elected Thmiddotrtfor(bull a 11w1st1n of tt II iim pol cint ial vus 1~1middotd 1atbulld llpon
knowledge of the s11bstances manufacshyture and USP for R The higlwst rating went to substanc middots which are v idely used especially in consumer products A slightly lower rating went to subshystances which an~ routinely emitted from industrial processes during proshyduction and use Some materials are employed in such a way that emissions are quite low even though tight emission control may not be required by law Materials in this cntegory were assigned a value of two for H3bull Substances which under federal or st ate regulations may not be discharged to the exterior envishyronment but which could be dischJrged by accident received the lowest rating
Each candidate was evaluated (n the basis of its 1ropen-ity to decompbullgtSe in ambient air Materials with half-lives greater than eight hours were considered moderately to highly stable and rated five for R4 bull Low to moderate stability was assigned to substances with halfshylives between zero and eight hours Compounds known to exist in air for only a few minutes would be rated zero
Ta hie II Dcfinitinns of the criteria used
R1 Prbullbullsent use in California
100 of max (109 lbyr) 5 10 1)f max (108 lbyr) 4
1 of max (107 lbvr) 3 01 of max (106 lbyr) 2
001 of max (105 lbyr) l lt001 of max (lt105 lbyr) O
R2 Growth in California use
+ 20 5 +Oolti to +20 4 Positive growth to 0 3 Stahle or unknown 2 DPcline 1 B1middotin~ phased out 0
R Emision potential
Widespread use in ltonsumer products 5 R1middotlatively p1or con rot over emissions 4 Rdatively good control over emissions 2 Tihtly c1int rolled 1
R4 8tabilit~- in ambient Air4 tvlnderale t1 high st 1bility (l 1 middot2 gt 8 hr) 5 Lw to modmiddotrate s1 bility (t1 l -0-8 hr) 3 Unstahle(t2~febull minutes) 0
R Dispcision Potential
Emitted lar~ely as apor or iine 5 particulnte
Emit led largely as rnarse particulate
Rlt Evidence f Carcinu~enici ty
Known or suspected human carcinogen 5 Known mammalian carcinogtmiddotn 4 ~usptctcd 111ammalan carcino~en or 3
known m mmalin muta~tmiddotn Ames ltmiddotst p sitive 2 Prpoundgtcursor 01 co-car inogen
a t11 is the half-liftmiddot
A-3
tion state or anion associations may change in the atmosphere metals do not degrade and were considered stable Asbestos is likewise itahlc Many of the clcc-ompo~ilion rtbullnctions oi orinnic molecules are mediated by light Such substances if released at night would have several hours to disperse in the surrounding area
A rapid way of assessing the relative potential of different substances to spread from a release point is to note their physical state under normal amshybient conditions Accordingly we scored materials emitted as vapors or fine particulates the highest for R5 and coarse particulates the lowest Intershymediate values are possible for varying amounts of fine and coarse particulate emissions from the same source or from different sources
There is as yet no widely agreed upon measure of the relative potenciemiddots of carcinogens although some ranking systems have been proposed14 Extrapshyolating data from in vitro techniques such as the Ames bacterial mutagenicity test and from laboratory animal studies to humans is problematic Therefore a less quantitative measure of the carcishynogenic potential of each candidate substance was used The candidates receiving the highest scores for Rs would be those for which there is strong evishydence of carcinogenesis in humans Exshyamples are asbestos which is implicated in mesothelioma vinyl chloride which has been identified as the agent of liver cancer in exposed workers and bisshy(chloroinethyl)ether shown by epideshymiological studies to cause lung cancer in resin workers The next highest rated substances are those for which human carcinogenicity is unknown but which have produced cancer in one or more mammalian species in laboratory tests Next are those which have not been shown to be carcinogens but which have proven to be mutagenic in test animals Substances for which the only knowlshyedge of carcinogenic potential is a posishytive Ames test (producing mutations in histidine-requiring strains of Salmoshynella) are rated 2 Finally substances which are implicated only as precursors or co-carcinogens would be rated lowest
Substances unequivocally associated with carcinogenesis were considered as carcinogens in this study Conditions of emission and exposure including the presence of co-carcinogens were facshytored into the evaluation of each candishydate where possible Carcinogenic subshystances derived from the metabolism of a precursor were considered as carcinoshygens However ubiquitous substances which have been hypothesized to be precursors (eg secondary amines nishytrous acid and nitric oxide which combine under certain circumstances to
1155
form N-nitrosoamines) were not conshysidered because of uncertainties in the importance of their link to the carcinoshygenic compound and the practical conshysiderations demanded by the scope of the study
It was beyond the scope of this study to jud~e the validity and interpietation of the experimental ad epidemiological evidence upon which the carcinogenicity of candidate substances has been esshytablished We accepted the conclusions about carcinogenicity drawn by the Inshyternational Agency for Research on Cancer or the National Cancer Institute and did not consider dosage or route of administratioll of tested substances However considerations of test validity did enter into the subjective evaluation by the panel of experts
Weight lr1 the sdditive ranking scheme each rating criterion is weighted according to its limportance relative to the other criteria Little precedent exists for assigning these weights In our judgment and as generally agreed by the panel of experts (see below) R 1 RJ and R6 are more important than the other criteria Evidence of carcinogeshynicity was considered to be the most important criterion of aB so W 15 was assigned a valtze of 3 W 1 aid W 3 were set at 2 and W W-i and W5 were set at 1 In order to discern the potential senshysitivity of the rankings to the weight assignments the candidates were also ranked using equaily weighted crishyteria
Ranking Candidates by Panel o~ Experts
Anine-member panel of experienced governmental industrial and academic scientists whose disciplines include~ -organic and physical chemistry indusshytrial hygiene toxicology epidemiology and iegulatory control of toxic sub~ stances was convened to provide addishytional data for our ranking algorithms
to discuss our candidate --ubstanc -sand rejections to suggest pos-ibie ne suushystances for consiceration and Lt rank the candidates indepen11mtly ( f our own ranking Tvo v-ee ilts he fore the meeting pa-d uember-s were jven one-to three-page dossiers on eac I canshydidaie substance
At th~ start of the meeting- before any group discussion the pand was asked to rate each candidate suhtance with a score from Oto 5 Next cmiddotach candidate was discussed at length e provided ~n ovrview and summarized critical issues identifi2d up to that Pbullgtnnt Through materials brought to thtmiddot meetin and their persoril experience panel memshyber Veimiddote able to provide much useful information on the candidates and adshyditional insight in to our ~a ting criteria
At the encl of the two-d1y sc~sion the pa-iel ugain -ated the cai 1didates
Flnal SelediDn
Tabe III show- the highest-scoring substance a~ det(rraine1 by the 1ddishytive and mlltiplictive approaches and by th~ panel in ~e additive approach 21 single r1nking as obuined by avershyaging the two ran kings resulting from using equal and uneguai weights The ran kings ltif most candidi tes were unafshyfected hut carbo11 tetralt hloride chloshyroform chromium and inorganic lead changed more tban thmiddotee positions_ Because uncertainties in the data base predude imputing signifiance to small differences in the Lnal ordering the lists in Table III are prtmiddotsented in alphabetishycal order However it is of interest to point out that benzene consistently ranked hihest
Becausimiddot some cindidates had equal rating scores we ltould nut choose exshyactly ten candidat ~s from the additive and multiplicative rankings Instead the
top nine and eleven were selected from the two exercises respectively We also considered the ten substances scored highest hy the panel at the end of the session The final consensus selection cornsistt-d of the 11 candidate substances appearing on at least two of the three lisL- For the substances included in the cun~ensus ranking a baseline emissions inventory is being conducted and a source testing program is being deshysigned
1ejected Substances
Some comments about certain subshySiF1ces not appearing on the final list are in order inasmuch as they include known carcinogens and compounds which hae received considerable atshyllttion in recent years as occupational c11c111ogens
rouisionally Rejected Substances Appended to the consensus ranking (T1ble III) were vinyl chloride gasoline and engine exhausts tobacco smoke and pesticides No further action by the CRB is recommended at this time for vinyl chloride because it is already subject tu the USEPA emissions stanshydard a CARB ambient air quality strndard and an OSHA standard
Gasoline and tobacco smoke were appcndlmiddotd to this list because each ocshycurs very widely and contains several of the candidate substances reviewed in this study some of which are in the final listing For example gasoline contains benzene ethylene dibromide ethylene dilmiddothloride and alkyl lead compounds the last three being in leaded grades only Both gasoline and diesel combusshytion products include PAHs Tobacco smoke contains among other neoplasshytigenic substances nitrosamines PAHs nickel arsenic cadmium and other heavv metals Manv individuals are inshyvolu~tarily and i~ many situations
Table HI Hi~hest ranked candidates from each rnnkin~ methodbull
Highly ranked but no inventory
l I ighesl consensus recommended Additive Multiplicativ(~ Pan I of eqwrts rnnkmg at this time
1156 Journal of the Air Pollution Control Association
A-4
lirtualy unavoidably exposed to t11-k1ltco smoke Because the sources of gasoline its combustion products and tobacco smoke emissions are well known no specific action was recomshymended for these materials during the tbullmissions innnlory and -ource testing design stages of the present study It was considered importint ho-vemiddoter to draw attrntion to the general publics exposhysure tot hrse substances PrsticidltS are listed for the same reason thou~h ad(bullmiddot tailed examination of pesticides w1s beyond the scope of this study Many pesticides are widely used and some of them are known to be carcinogenic
Other Rejected Substances Acryshylonitrile and vinylidene chloride were placed in a provisionally rejected group because of the panels suspicion that imports of these compounds to Cnlifornia from Japan may be appreshyciable yet are hard to substantiate Should such imports be verified in the future these two compounds would take on greater importance Formaldehyde was also provisionally rejected because the preponderance of evidence indicates that it is not carcinogenic and that bisshy(chloromethyl)ether is not formed from formaldehyde in appreciable quintities in industrial environments 15 The ocshycupationally controlled carcinogens ethyleneimine and beta-propiolactone were rejected in part because of their reactivity in air At the time of this study DBCP a pesticide was no longer heing middotproduced in California and was therefore rejected from further considshyeration
Occupational Regulations and Community Exposure
A question of interest to the CARB was whether the regulation of acknowlshyedged occupational carcinogens adshyverselv affects the ambient air outside the v~rkplace Our general findings can be illustrated by the example of asshybestos
Asbestos is a very widely used mateshyrial for which no ambient air standard exists Concentrations in the workplace are limited to an eight-hour timtmiddotshyweighted average concentration of two fiberscm=1 of air and a ceiling concenshytration of ten fihcrscm 1bull In meeting this standard exhausting air containin~ asshybestos to the ambient air is not reshystricted except by the USEPAs reshyquirement that there shall he no visible emissions containing asbestos piut ides from s11eh foci lit i1bulls (bullxd11din~ hrak(bull shops 1 Considning that undc-r certain conditions 101 asbestos fiherscm1 could ht Pmitted without being visihl( 11 this standard mamiddot allow considerable asshybestos emissicns It is unlikely however that emissions would actually approach these levels First since the OSHA standard cannot generally be achieved
November 1979 Volume 29 No 11
by ventilation n ine the generation or asbestos partid in the workplace must be greatly reltlu ed Second the air is usually filtered 11gt prevent recirculation of asbestos to 1 workplace Asbestos waste must bed 1sposed of in sealed imshypermeable bag~ or containers Thus under current onupational regulations the amhient air 1enerally appCars to be afford(bulld greatn protection than it would without s11ch regulations 1
Conclusion
The screenin and ranilting methodshyology presented in this paper proved to be a feasible approach to establishing a priority list of Jirborne carcinc)gens in California We ftel that it is an efficient means of focu~ing further efforts on emissi()ns inventories source testing and ambient measurement for it not only identifie all the carcinogens of potential concern but it also permits the state regulatory agency to direct its reshysearch resourCtmiddots toward those subshystances of partilmiddotular interest within its jurisdiction
References l National Elllission Standards for Hazshy
ardous Air P llutants A Compilation as of April 1 19 3 PEDCO Environmenshytal Inc for US Environmental Proshytection Ag1middotncy EPA3401-78008 (NTIS No I B 288 205) 19i6
S Gray Ne Jersey Department of Enshyvironmental l 1rotection Trenton NJ private com111unication 1979
1 H E Chrislllsen E J Fairchild and R J Lewis Sr middotmiddotSuspected Carcinogens A Suhfile of th1middot NIOSH Regii-try of Toxic Effects of Chemical Substances 2nd Ed National Institute for Occupational Safety and HPalth NIOSH i7-149
4 B 8 Fuller J Hushon 1 Kornreich R Ouellette L Thomas and P Valker Scoring of rganic air pollut1nts for US Environrnental Protection Agency Mitre Corpt ration T1middotchnical Report MTR-6248 I 9i6
S F R Lozan11 Toxicants Selected for Priority En- 1ronmental Assessment corresponde11ce with EPA Region VI Administrat r 19i7
1gt Survey of inyl Chloride Levels in the Vicinity of Keysor-Century Saugus California US Environmental Proshytection AJet middoty National Enforcement lnvestiiatim Center Denn-rand San Francisco E A-n02-77-017a 19i8
0 Bardin l middotstimon~middot in Hearings on Helationshi1 Bet wfen Cancer and the Environmei 1 US llltuse of Hepreshysentatives bull th Congress 2nd Session Committee (In Interstate and Foreicn Commerce ~crial No 94-141 1976 pp 7~i0
middot T H Ma11middoth Carcinogens in the workplace lwrE to start d1middotaning up Sc-imiddotrmmiddot l17 1-1110) lfi8 (197)
I CH ar1111 Classif1r11t1on of the Exshyposure Haza d Prtbullstgtnled Ii~middot Thirt~middot-Six Clwmirals 011 I llixllmmiddots lternorandum from Clllm1middot I Offi(middote of thtgt Assmmiddotiate lgt1rector for 1 middotarcinoltnesis Division of Cancer Cau~ l and Prevention lo SushyJwrmiddoti-or~middot ( 0 Auditor Gtmiddotneral Acshycounting Oft lmiddote19ii1
I0 Chemicals ith Sufficient Evidence of ( arcinogenit ty in Experimental Ani-
A-5
ma1s Jnlernauona1 Agency 1or 1c-curcn on Cancer IAHC Working Group Reshyport Lyon France 1978
11 Threshold Limit Values for Chemical Substances and Physical Agents in the Workroom Environment with Intended Changes for 1977 American Conference of Governmental Industrial Hygienists 1977
12 M H Rogozen D 1 Hausknecht and R A Zisk ind 1976 A Methodology for Ranking Trace Elements in Fossil Fuels According to Their Potential Health Impact by Science Applications Inc for Electric Power Research Institute 1976
1t L Margler R Ziskind M Rogozen and M Axelrod An Inventory of Carcinoshygenic Substances Released into the Amshybient Air of California Vol I Final Reshyport Screening and Identification of Carcinogens of Greatest Concern by Science Applications Inc for California Air Resources Board 1979
14 T H Maugh Estimating potency of carcinogens is an inexact science Science 202 (4363) 38(1978)
15 C C Yao and G C Miller Research Study on ais(Chloromethyl)Ether Forshymation and Detection in Selected Work Environments by the Bendix Corposhyration for National Institute for Occushypational Safety and Health Cincinnati Ohio Contract No 210-75-0056 1979
16 Occupational Cancer Control Unit Calshyifornia Department of Industrial Relashytions Los Angeles California private communications with staff members 19i9
This paper was developed while Dr Margler was a Staff SriPntist in the Energy-Environment Systems Division of Science Applications Inc (SAIi 1801 Avenue of the Stars Suite 1205 Los Angeles CA 90067 and Mr Reynolds was a Project Manager with the California Air Reshysources Board Sacramento CA 95812 Dr ltar~ler is currently emshyployed as DirEgtrtnr of Bnvironmental Scitbulllltcs for WinzlN nnd Kelly Conshysultini EnintNi t1 Third Stretbullt EurekH CA 9gtgt01 and Mr Heynolds is Air Pollution Control Diretbulltor for the Lake County Air Pollution Conshytrol District 55 N Forhes Street Lakeport CA 95-13 Dr Rogozpn is a Staff Scientist and Dr Ziskind is Deputy Manager of the Energy-Enshyvironment Systems Division of SAi
1157
Appendix f~
_l)ispobullition of Miscellaneous Sites
Gould Inc Ver_n)_~ St~condary Lead S~_n_elter
The emissions of arsenic from the four large secondary lead smelt~rs
in Cali forni) J1~1 ~ bull~tiilnted in the program This estimate was based upon a
uniform fugitive emission f1cJr nt WPll supported by measurement inforriation
Therefore source monitoring was r2cq11)nded 1nd Gould Inc Vernon being
~he largest was singled out It was however proposed to monitor both Gouli
and RSR Corp (Quemetco) in the City of Industry since analysis of the plants
revealed significant differences in plant equipment and engineering The
latter being more typical of 1 nodern facility
As a r2sult Jf pr-~-middott -iiscussions and plant inspections we became
aware that Gould was actively in the process 1)f constructing a new facility
which would completely replace the existing one We have monitored progress
on the new site and concluded it would he inappropriate to utilize program
resources to conduct field tests at Gould ~rr1issi 1rns from the new Gould
faci 1ity wi 11 be bas~rl 11p11) test results from RSR
PG ampE - Pittsburg PG ampE - Salinas a~d So Cal Edison - Long Beach Oil Fuel Power Plants
It was app ro p ri ate tJ cons i dP r the (~mission of arsenic from power
plants during the initial study stage inc~ imiddotrace quantities in the fuel oil
are known to be ernitted Because of the population distribution in tile
vicinity of three plants and some 11realistically conservative estimat2s of
erlissiJn c-1ctors tht~ facilities appearc~d among the top seventeen stationary
sources of potential c1Jn-r-n HovJ~ver as a result of a reexamination of
emission data it was concluded that an error ~dj ~een made in the material
balance of arsenic - resulting in tn emission factor equivalent of a 300
release ~io lit~rature was found to justify 1reater than a 30 transfer
function Thus we estimated at the outside th~ emission factor should be
reduced from 013 lb per 1000 lb to approximately 0013 lb and resulting
ars9nic 2missions for the entire state w~r~ cJ11servatively estimated to be
1760 lbs (from 17600) divided amongst 311 the states power plants Clearly
then the four secondary lPad smelters estimat~ of S9410 lbs of arsenic(
B-1
emissions per year makes consideration of poer pl int ernissio1s if ffsenic a
very low priority
In a literdture review by SAI - Methodokgy __f_or Ranking Trace
Elements in Fossil Fuels Accordi~9__t_~_their Potential Health Iripact (Roqozen
1976) - emissions estimates f0r 15 trac~ substances were researched for c11l
and fuel oil pomiddot1er plant c 1J1v---r-sion Source content combustion pr-)~~
transfer functions and contrcl methodology were co1~idered to develop e~ission
factors The output to input ~atio computed and utilized n -~i-i~ study for
arsenic was 002 to 03 (ie tra1sfer function betw~en 2 11d JO~) reflecti~q
the wide variety of fuels processses and controls nationwiJe The upper end
reflecti1g both high arsenic coals and poor e11iss10n contro 1 devices Clearly
neither of those conditions accurtely apply to th2 thr~~ si~~~ r1nrl thus
substantiate the rlecision not to perform emissions neasur-i-~n~s
Calaveras Asbestos Company - Copperopo l middotis Asbestos Mini n(i and Mi 11 i nq
In the past (late sixties and early seventies) ~h2 ~3laveras
Asbestos Company came under heavy criti-is1 1fter imiddot1sp~ction rieasurements
revealed serious problems However significant reduction of e1nissi0ns have
occured prompted by NESHAP regulation and occupcttional standards SAI
inspected the site in f)~c~1b-2r 1979 under an EPA contract An missions
inventory was published under that work (Ztskind 1980) and the bottom line
conclusion is that currently no significant ernissioi1 are gteing released as a
r~sult of hl-isting which would reach the publ le i 1le 0pen pit is some 900
feet deep with blasting at the bottom 0ver 80 percent of emissions in the
CARB Emission Inventory System were attributej to pit blasting In fact at
its current depth blasted 1aterial does not reach the mine surface with tile
explosive implacement used Additionally the site is more remote than might
be realized The ~ituation is vastly different today than in the past and
currently attention should be paid to the issues of occupational exposure and
breakdown of ventilation syste111 controls in t1e inilling operationsmiddot He
recorrrnended that no furtw~ corsideration be Jiven to this site for the
purposes of tl1is w)JJil ie identification )f silt~middot1ificant releases vhich
might he responsible for causi1J middot1 fmiddot middotgtbullgt middot 1= 1(11 1bull11 tinn Pxposure
8-2
Vctri0J5 Refineries
It was est=tblished early in the anrilysis prmiddotJr111 V1at gtenzene
emissions from refinery operations coulrl in th~ 1guregate constitute a
significrnt source Since there ilre a large rrnrnber 0f refineri~s in
California (46 in Los Angeles County itself -it t(~ i 1~ middot)f r-1ltariination) with a
wide variety of types sizes ages etc their eva l uat 11) 1 i =r~sent -3~ i)7
monumental task It was noted immediately t 13t ~~it hi n the total scope of the
study the design and conduct )f c r~f i r1~ y testing program which would develop
a complete benzene emissions in1~ntory for 01e site was impractical let alone
to d1aracteri ze emissions fro111 three ie those listed among the 17 rnost
s i g n if i ca n t potent i a l st a t i on a r y sou r c e em it 1-_ l rs bull
The three refineries were singled out for special attention in the
previol1s phase ~ecause they uriiquely had components which process materials
containing 10 or more per~nt hy middot1-bulli1 11t )-~112~ne (46 FR 1165 Jan 5 1981 pg
1491 ) Est i mates by E P A ( F e d bull r l Rr-~ g i s t ~ r 1q81 ) i n d i cate that 90 p e r cent o r
more of the total benzene fu 1Jitive emissions arise from such components
Therefore attention is appropriately focused on the three henzene production
consumption refineries Chevron (Richmond) Arco (Crson) and Chevron (El
Segundo) SAI staff considered the possibliJ )f -1 testing program at one of
these refineries and cone l url 1middotd it would not 1)e cost-effect i v2 for o number of
reasons
California is a minor producer and consumer of benzene with
approximately 15 of the 114 hill ion pounds produced and
consumed nation1lly in 1977 This can 1)e riY1pared with the fact
that California has approximately 10 of th~ ~ations population
Benzene exposur~ to the generil nJmiddot1lation has been partitioned
among the various sources Aproximately 90 of exposure is
estimated (SRI 1978) to be caused hy gasoline distribution
activities and 1ehiclEmiddot emissions in 1 ir~an areas with nearly all
of the balance )Y benene hrndl ing operations (refineries and
chemical plants) In the case middotlf California this percent1g~ Jill
be even JT1ore di middotpruJor-tionate b~cause of its greater s1are of the
population and ~esser share of the henzene ha~dling Furthermore
since the three refin 1bullry cites ire heavily urbanized no new rural
population cegmmiddotnt is beinq exposed
B-3
Californias most heavily urbanized Air Quality Management
Di st r i cts have a l ready adopted t en zen e f ugit i v e em i s s i on
staridards comparable and vith S( 1ine features VJr stringent than
the proposed national (bullmission lttandarrl (4f-i FR 1165 Jan 5
1981) Thus the concmiddotlusion rleri1ed above (Le refinery emissions
are secondary cause of exposure to the urb~n microopu~ation) is
further reinforced since H is irojected that re1eases from
components in benzene service (gt10 benzene) will be reduced by
73 by the proposed F~deral Standards T~~ ~istrict rules (eg
South Coast Air Quality Management District 456 and 4fi6l) are
not restricted to berz~ne per senor to only components in
benzene service The rjLs 1lso i-lclude flanges in addition to
the components ca i led uut in the propose nat i 01a l standard
The EPA estimated that if the proposed emission standard were adopted the
1~aximum annual benzene concertr0Vioi1 -~01middot 1 plant would be 36 ppb at a
distance of 0 1 kilomeL~~ avay Cor1paring this ground level concentration
with the general urban background i1 California of 19 ppb shows the latt2r t1
dominate In recognition of the secondary im0ortance of these thr~~ i~~s i~
was decided to utilize program resources to d~velop field data at other sites
1111here little emissions informati1)i1 1as cll-1aila1)le
8-4
APPENlJIX C
US tn v 1 ronrnen ta I Protection Agency
middotJc k
METHOD 108 - UETERMINATION OF PA~TICULATE ANlJ GASEOUS ARSENIC EMISSION
FRUM NONFE~RUUS SMELTERS
1 Applicaoil ity and Principle
11 Apmicroli1ability This rnetnod amicropliltS to tt1e determination of
i nor~an i c d rsen i c ( s) e1n is sions trom non f errou Sm(~ I ters dnd other sources as
smicroecified in the reuulations
12 Principle Particulate and gaseous As emissions are withdrawn
isokinetically from the source and collected on a glass mat filter and in
water The collected As is ther1 analyzed by means of atomic aosorption
spectrophotometry
2 Apparatus
21 Sa111pliny Train A schematic of tt1e sampling train is shown in
fiyure 421 it is similar to tt1e Method 5 train of 40 CRF 60 Appendix A
Tne sampliny trdin consists of tt1e followiny components
Note H1i s and a11 subse~uent nferences to ottler methods refer to the Methods in 40 CFflt 60 Appendix I
3 Annual Book of AST11 Stdndarctbull Part 31 Water Atmospheric
Analysis American Society for T~sting and 1-1aterials Philadelphia PA
1974 micromicro 40-4~
C-15
APPENDIX D
~f-~__d_y_ e s f o r Pr o c e s s i ng a n d An a l y s i s of PA H i n
Co~e Oven Emissions from Kaiser Steel
Global Geochemistry Corporation
10 Samples expected
l 1 Connective tubes between section welded to oven port cover and remainder of sampling system - These are to be stainless steel tubing disconnected and capped off for each sample They will be about 10-15 long
1 2 Impingers in series first two water-filled third dry - The center tubes will be removed and wrapped in clean foil and the impingers stoppered as is for shipping One set for each sample Protect from light during shipping
20 Extraction -procedure
Sample fractions are to be covered with foil or in opaque containers with inert gas flush during storage in freezer Solvents used in extraction will be BampJ glass distilled pesticide grade low residue) Working bench area will be screened with amber plastic and light bulbs
2 l Connective tubes will be rinsed with dichloromethane by partial filling capping tilting several times and draining into a beaker This will be repeated until the washings are colorless but at least three times
22 Impinger contents will be transferred to a separatory funnel The impingers and center tubes will be rinsed inlo the funnel using a measured amount of dichloroshymethane in portions The funnel is shaken well and the extract drawn The rinsing and shaking is repeated with two more portions of solvent The extracts and washshyings are combined over Na 2 S0 4 bull
A separate portion of the water used for impingers is extracted in the same way
For each liter of water 100 ml dichloromethane will be used per extraction ie 300 ml total
Internal standard is added to the combined exshytracts at this p0int
D-1
_rmiddot
30
40
Concentration of extracts
Bulk solvent is removed on a rotary evaporator in a water bath at not over 40deg stopp~ng short of dryness The res~dual volume should be just less than the f i n a l v o l ti 11 e o wl1 i ch the s o l uti on i s to be d i l u t e d (accurately) The object is a final solution with roughly five mgml solids concentration in dichloromethane After the vol w1 e ~ s made up to the mar k i n a vol umet r i c flask an aliquot is removed and evaporated on a tared disk in a nitrogen stream and weighed to determine yields The remaining solution is stored in a ial or ampoule sealed with a Teflon-lined septum A split for SAI QA may be made at this point (see Section 43 also)
Gravity column c1eanup chromatogram
4 1 Into a 10 min 1 d column with co2rse sinter and Teflon stopcoc~ 6 g 120 mesh silica gel (activated at 120deg) 1s sluiry packed in pentane cooling the column by evaporation from a tissue wrapped around it and wet with acetone) This minimizes vapor gaps in the column A cap of 3g sodium sulshyfate is added to the top The column must not dry out before the chromatogram is complete
42 A sample of the extract concentrate is taken This is to be based on experience of expected PAH conshytents but initially would be 5-10 mg This is exchanged with pentane by adding 10 ml portions of pentane and evaporating over about 100 mg silica to small volume Three to five portions of pentane are used The final residue is rinsed onto the column with pentane
4 3 The chromatogram is developed with four solvent steps
l 25 ml pentane 2 l 0 ml 2 0 ~~ dichloromethane in pentane
II II ii II3 l 0 ml 50 u II II4 l 0 ml 5 0 ~~ methanol
The column volume is approximately 8 cc This is allowed for while collecting fractions During the first chromatogram the graduated cylinder collector is observed for visible solvent changes and a more precise estimate of the column volume
T h e f o u r f r a c t i o n ~ a r e e v a p o r a t e d i n a n 1 t r o g e n stream at not oveimiddot 40 to l 0 ml then stored under nitrogen in the freezer The principal constituents
0-2
of the fractions will be
l Aliphatics 2 Lower molecular weight PAH
11 113 Higher 11
4 Polar materials
Fractions 2 and 3 are to be analyzed by HPLC the others retained for future interest or in case of an incomplete or defective separation Splits for SAI QA may also be made from these fractions
50 HPLC Analysis
5 l The column is a reverse phase partition column C18 bonded to microparticle silica 4 x 250 mm The solvent system is a gradient from 6040 acetonitrile water to 100 acetonitrile initial slope setting l 5min The flow rate is one mlmin the injecshytion sample volume one to ten microl depending on conshycentration Detectors are UV absorbance and fluorescence in series The absorbance is set routinely at 296 nm but may be varied in certain instances to aid in identification of peaks The fluorescence filtars are narrow pass excitation at 360 nm and cut-off emission above ~10 nm Both d e t e c to rs a re rec o rd e d s i 1i ul ta n e o u s l y
52 Reference calibration is based initially on indishyvidual chromatograms of single compounds and roushytinely interspersed chromatograms of multicomponent reference mixtures The proposed calibration stanshydards are the following at minimum
Phenathrene Pyrene Chrysene Perylene Benzo(a)pyrene Benzo (ghi)pcrylene I n d e n o ( l 2 ~ c d ) p y r e n e Coronene
The internal standards (added in Section 22) are 910-Dimethylanthracene
9-Phenylanthracene 9 10-Diphenylanthracene
60 Peak Identification
Significant peaks not recognizable as present in the reference standard will be investigated by alternate methods for identification
D-3
6 bull I Stopped-flow examination of UV absorbance for maxima
by manual wavelength varmiddotiation allows checking against known reference spectra (published or measured) This is usually sufficient confirm a component already suspected to be present
62 Collection of the fraction contairino the unknown peak allows later determination o~ t~e full absorbance andor the fluorescence spectrum for comparison with reference compounds
63 If not already run on GCMS the sample can be sent to SAI for that purpose possibly allowing the comshyponent to be identified Since the HPLC elution order is not the same as the GC retention sequence it may be necessary in affibiguous cases to collect the HPLC fraction and run it separately by GCMS
The~e may be numerous minor constituents not identified The decision on how many of these to track down and identify will be made in consultation with CARB adv~sors a~d with SAI since at some point the returns ute not worth further expense
_64 IdeAtified peaks are quantified by comparing their areas to the areas of the known concentrations of reference compounds A recovery factor based on the added internal standard nearest in elution order to the sample component is then applied the amount of internal standard originally added to the sample is divided by the amount found in the HPLC analysis This ratio is the multiplier used to correct for losses in concentration and chromatography steps
70 Quality Assurance
71 Transmittal of sample ~plits to SAI provides for external QA This should be done with 15 (or one in seven) of all samples
72 Blank water extracts are carried through the same analytical procedure (10 of total)
73 The cleanup and HPLC analysis are duplicated on 15 (or one in seven) of all extracts
74 The multicomponent reference is run first every day then elution volumes and peak area for each comshyponent are plotted on a control chdrt to pinpoint developing problems At least evetmiddoty tenth chromato-gram will be this standard It wi 11 also be rerun after any service on the instrument~ such as a column change
D-4
APPENDIX E
Gas chromatograph 1mass spcmiddotctromlt~try protocols used by SAI Trace
Organic Chemistry Laborator1
Aliquots of sampl=s received from Global Geochemistry Corporation
were analyzed by combined c11s chromatography mass spectrometry (GCMS)
Liquid extracts were spiked with an internal standard compound (deuteriumshy
labeled phenanthrene) priot to analysis by gas chromatography mass spectroshy
metrymiddot A Finnigan model 4( 21 Quadrupole GCMS system including an Incas data
system was used for all analysis Liquid samples (1 microliter) were injected
into a 30-miter x 025-mm ID SE-54 fused silica open tubular 9as chromatoshy
graphy column (JampW Scientific) A programmed GC oven rate of a 4 minute hold
at 30degc followed by a 40degCmin temerature increase to 160degc and a s0 cmin
temperature increase to 270degc was used for sample analysis The detector
system was a quadrupole mass spectrometer operated in the electron ionization
mode at 70eV The quadrupole mass filter was scanned from 35 to 475 atomic
mass units in 095 seconds A hold time of 005 seconds was used to stabilize
the electronics prior tote next scan Data were continuously acquired using
a Finnigan Incas Data system which writes time intensity mass spectral data
to a magnetic disc The mass spectrometer was operated under computer control
during acquisition Figures 33-7 through 33-9 show the total ionization
chromatograms for each of the three samples that were analyzed
Following analyses the data files were searched for priority
pollutant polynuclear aromatic hydrocarbons (PNAs) In addition to the
quantitative data reduction used for priority pollutant PNAs a survey analysis
was performed on all significant GC peaks (ie signal to noise of greater
than 101) The survey analysis was based upon a computerized library search of
individual background subtracted peak maxima scans for qualifying GC peaks
The library used for qualitative identification AJas a combined NIHEPA
National Bureau of Standards library 1hich contains approximately 26000
entries
Tables 33-11 through 3-13 present the results of this survey analysis As expected a series of al~yl Slbstituted PNAs were observed in addition to heterocyc l i c compounds con ta i ni n~ nitrogen iind sulfur moieties Further
investigation of the mass spectre of the nitrogen substituted identifications
confirmed the presence of benzonitrile isoquinoline andor quinoline acridine
carbazole and alkyl substibted ~yridines Ion specific searches for N-nitroso
substituted amines were neg1tive
E-1
APPENDIX F
Detailed Fiber and Mass Concentration Data - Johns Manville
=-- -i l L - [- - =-- -= f- - =-- -l J middot - -=-
i-- L =-- - ~ ~ L ~ ~ =--shy r---- --
- 7 l -=----7- -- - - - -- =-- - - ~ =-- - 1- -~- l
g~~~~ r_-middot~ ~-i [-- L ~ L Ldeg
middot--~ l - --------=-- r- L r- r-
~~~ =-- -- r- = - -r---
a- - - - - _ - -I~=-- L -- l 7 -~--
gtZlt _
- _(_ -lt lt =- = ZJZ~
-
__ _
~ -- __ middot- -~ ~~~ = L ~f~ Z ~ C
middot -
__ ~ lt lt =- _~~ijE
gt = z lt gt
rbullshy__
F-2
SI I HMllLfc J H 1 I OBr I lll-H ANI) NS~ coHcrrrllliTJ ON CUfllllTI VImiddot SIJMMIIY
TIITIL JnI~ SfiNtlEO TOTtl Ill- or FI ITETI TOTI Illlt ~111-11 11 ltllOT FIIACT I Ori 1111 OF ~middotWI MUllllNJbull lOTI 01IIMImiddot 01- JII TOTI FI lilmiddott~ UgtIJH nD
= t nwo 00 so MI CllONH = t I 1JOO SO (M = II 900 SO CM t bull (0000 l1T~
I I JO SO CM 70000 CUil JC Ml-TEHS
= ~lt O V l lWfl~
Manville D-2 Baghouse
723 347 -633
CIIHYSOTI LE JfllII I nou J11BICUOUS NO PTTFllN NON-SBI-STOS tLL FIIIEIIS
FI 111-11 COllflT ltFIIWll-
ITllClmiddotrir r lllrf
2gt()
) ( I f S bull1 bullbull
I 0
l ~-Irbullmiddot
00
fl JI)(~
00
0 OUU~~
00
0 OflO
~60
1110 000~~
FI IWII ( ()[IClmiddotNTHT I ON ltI- I mn ll11 ~41 GI OF FI LTEH) (FIBlmiddotn~ ll-11 CIJB f-11bullTl-H OF IH)
I (2fi71bull+0G (aG(l-+05
( gt027 F+04 OfmiddotHH
OOOOOE+OO OOOOOEtOO
OOOOOF+00 o uuoul~middotH)O
0 O(OOE+OO o 0000I+oo
J bull f 1)07 E+O( fi r71HE+OG
I
w
TcfLI ~~lllllmiddotCE 1lllmiddot lt~O Ctl lFII ~~O UI OF FIiTEi) t q UI 11-11 Cll II fllmiddotTlbullll OF J I H)
I 590)1+07 6 t 90llbull-HH
( HMOE+O 2 40411bull-1 o
0 OIHHlF+OO o 00001+oo
0 OUO(I 1 00 OOOOOJi+00
() 10110~ Ji)
OOOOOE+OO 1 lHHllJ i 00
0 00001middotgt1 00
M ~~~ middotor-ic Irrrn TI or
( PGRAMS 1bull111 ~c UI OF F J LTEH) (PGRAMS 1111 UJI illI Lil ()j Ill)
IUitTII Ml-Ml ltMI tJIOfl~~) ~Tlgt lgtEV
Ml- H Lot c1middoton nr COlmiddotF I II
I) I MWTEll Ml-N ltMI c11or1~~) STI) tHV
MEtl IOC cImiddotor1 r1N CCWF V11
AIt lbullI rt 1bulli I~ (Sil MIC) STIgt 1gt1-V
Ml- N 101 c1-ori r1u COlbulltmiddot V1ll
I 0 I ~if1+07 bull tJf I (11bull-HH
r 1gta 1 1) bull 2 (10()
0 lt700 l 1Hi4~ 2G44
() ~( 1 027B
- l bull 6 I 1bull1middot 0 J l)J 09J~fgt
1a 1071 1~ 09rn
0 2(jlt)fi l ~HVil 4040fi
~ HltF+O~ U (H~61bull+o I
o rooo 0()()00
-0GJOB 0 (000 0 (000
OOfiOO 00000
- )()57
OOGOO 00000
0 0 1)02 00000
- I J 0 0 0 1W 00000
00000 0 ()000 00000 00000 00000
00000 00000 00000 0 (WHO 00000
0()000 0()()()()
0 ()()()) 00000 00000
00000 00000 0000() 00000 00000
00000 00000 00000 00000 00000
00000 00000 00000 00000 00000
00000 00000 00000 00000 00000
00000 00000 00000 00000 00000
00000 00000 0 (WOO 00000 ()0000
( ~1701 l B bullJl~H 0 lt~( I BIJI ~ ~i Ill
o~-~ 0 ~j
- I ult-fc 0 I Hbull)I o bull)abullgt~
12 (1 41 17bullgt7
0 1 l I H) 4 ltmiddotI 1
VOUJME (ClJB MIC)
MFiN ~Tl) 111-V ru-1N 10C Ct-OM MN COE- Vtll
2lHH7 gt HUB
-2 7m 0 060 1)
( IJl)()C)
0 001 00000
-ltj 74a 1gt 0001~ 00000
00000 0()000 00000 00000 00000
00000 00000 00000 00000 00000
ooono 0 1)000 00000 00000 0 ltWOO
2 7 0 1J lJ I J J
-2J+H () ()~(
B 77gtB
lt ~
lt
C
J c
~ lf
_
ii
(]j (fJ
J I 0
a1 c r--- l)-- Clltr (Y) r- ro C)(V) gt deg CN C1 (0 I ~ C M
N --- r--
w J
s rJ gt-
c
~
LJ
t-
~ - -lt-[-bull
0- ~ l L L l~ -18~~
L a- - ~ I-
-I - ---
-
-- -- -- - - ~l
-1_ - - --
--~--- [-middot-~--- t-- -- [- -_ 3J ~I r-
z g ---
l
t--[-
l l c +-shyc
~ Ci - - I
-1--r-middot~- ~
L l- L ~
L
II V
C
c
t--
-= -- - - ---- - _
middot= - middot-= l J l~ - ~ ~ L - ~ L ~ [--
II II II II II II II
_
---
--
z
- I
==== lt _ --middot --~ 7 _
- ~~- _
z ~ ~ =lt ~~~ti3
f
gt =2 lt_
2 z _ lt=lt i~iS
-- z ~ lt
~~~t3
~lt~ =--~~~sect~ _ -- -middotImiddot_
L -_ -
F-4
(lJ II) I
~ Or--M
ltlJCsr-M ~ O) bullbullbullbull ~ lCMIO bull- ci gt (= N lC I M ~ 0 N
_
J J
---r--co - - ~l r-j = ~i
~~ 2-
88 rJ--_-ltI -
lt ~lt
z
Cl J ~ _ e-l ~~
- ~ middot~ e-- l ~ l~ [- - ~ -~=----~
_ L
L
--o---
-o-~shyc L~ ~ - 0 - Ll-C d o~-
11middot II 11 11 II II II
-_ 0- _
--
-- lt== ~~ ltZ z-== -~~=E~8C~l~-=rI - lt ~ fJ
ltZ lt---=lt===~~~- _ ltltlt=gt ~~=~~~~~ -
~ ~ lt -=z~ z _ -shy middot--z =-
-middotmiddotmiddotmiddot middot-middot -----
r bull bull _
- _ - ~-middot __ _
~ _
middot- - --- middotmiddot
_
- l
~=~r -~= ==
lt =~~~ti
F-5
FIHJ-ll Nil MAS~ CONCFNTllJTlOi SJ I SMllLE I l) Al OJ CUMJIATI JIbull t-llHMHY
TOTL iHf- SCNNFD lOTiL Illmiddot OF FILTFTI TOJI Ill- JSlllmiddot~ll I I OIIOT FlltCT [ ON 1111middot OImiddot ~1-M iIHlllllJNE TO 1 1 I ()lIJMImiddot ()I t I H TOTI V 1111-11~ COUNTED
= 1080000 SQ MI~RlozNS = 11 900 SQ CM = 1 l bull )00 SO CM = 1 00000 liHlS
11 IH) so CM 2G70000 CUBIC METfiHS
1 ) fi FI JllIIS
Man vi l1 e D-2 Baghouse
723 0 347 -633
Cl Ill Y-OJ l LE MIPII I BOLF Jf1BI(UOUS NO PATTEnN NON-ARHFSTOS ALL FI BFllf~
F 11l-ll f()llrlr ( I 1IFH-- l 17 G o 00 00 00 19fi
Tl fL P1 Sl~JNNFD Tl lI ltl- OF 1bullrn~n H 11I HF ~IJlbullI) lll 1 ll(IT 1-ILCTON IL CW ~U-1 HErlllllNE
THl1 (11llrJI~ OF 1IH Tt1T-d 11111middotH COIJrfflI)
= 697~oo so MICRONS l l 1)00 so CM t I JOO t-o CM
= 1 00(00 lHT~ = I l 1)0 ~o GI
ant()Oo cun1r NETEns = 7 0 I l lWllS
Manville (upwind)
CIIIIYsOllY AMPII I nou JMfl[CUOUS NO PA1vrrHN NON-A~~HFSTOS ALL FIBFHS
v1mI1 Cttllffl l 1-- I II I fl~ 1 () () oo l 0 00 40 70
1middot1111unmiddot ToT1L f J BFllS o ooo~ ooow- t OTT~ ( 01rn 57 I 1~m too ooo~
1middot 1111--11 1or111rnnT I ON I Imiddot I 111 II~~ 1111 ~o CM OF f I LTFTl) 1 11111 1middot111 CUil middot11bullnn Illmiddot J l ll)
O OOOlJEmiddotH)O UUUUOL+OO
0 00001-1-00 0 00011 lbull-Imiddot00
G 1 I nn+OG 1 HJ llmiddot~+OG
00001+oo 0001101-1-00
r n~+H+uG l 7llHmiddotlE+Ofl
l bull l 1)1-n+O(i I J ~ 1)(11-middotl()i
T11T I ~Ull FiCImiddot 1 ll f ( ~ l lrl ll11 tn Ii OF F 1ITFll)
f ill I I I i 11 I U l ( i I Ill J
0 OOl)Ol ll() () U()0 I Imiddot()()
0 UOOOlmiddotgt1-00 O OOlgtUlbullgt100
l 0 1) ) I 1 lJ J ~ tl( 1H-HH
0 UUUOlmiddotH)I) o 00001-+oo
~ lt1Hlt1(E+OG 7(H(Ml-middotHM
0 OOOOJbullH)O 0 0000lbullgtHlO
77 I
N ~
rl gt- 10NCJfffll1T I ON
r orR AMS ITll ~q lfI (II- FIT n~ll) - LIi LIIll 111middotTU OI 1lll)(PGRAMS II i1 Ill MlmiddotN
bull I t bull I 1 f I ~ middotr1 Ill lmiddotl 1- [~ IOC ( Hl1I f r~ Ull-F ll
0 0000 Imiddot bull-110 0 0000 Ibull HO
0 ltHgtOO 000110 t) (HJ()()
0()()()()
0()000
llOOOOF+Oo 0 00001bull-100
0 ()000 OOtlOO 00000 0 (WOO 00000
0 (((7
0 ~0-i -0 middotlbullribull)bulll-
0 ltl l lt 0 oHi
00000 00000 00000 0(0HO 0000()
077[0 0 -1117
-0 ~bull)I)()
0 7middot1middot1 (i () L~bull)7
0 7~B( 0 I I
-o alt17n 0 (1)~1 0 17lt1 I
11 I 11ITlt I j i 1l 1 111bull I
f 11middotgt N gt I ii ii I V 111 ii IOC tHlfl flf1 Clll-tmiddotmiddot VAIi
00000 0()000 0000() 00000 O 0000
041000 ()0000 () 0000 o 0(1()0
0000()
0 llfl(i7 OltHJ
-~ lt-11 I 0 07 I 1 () I)lt)
0 (H)ll()
00000 00000 0 ()()()() 0 0(100
0 0 Omiddotlmiddot l
-c07 1)
0 I 1 I( 0 1101
0 l l j--
OOGli -~bullI) ( (J
0 1lt11 I 077G7
I 1 (71 rJI C)
tll N -r11 111-V rmiddotllfl trn CI llf I 111bull1 COIT rll
() ())(l(J
00000 0 ()(1()()
0000() () (1()(10
00000 00000 00000 0 (I()(()
00000
0 ~ I Ifdeggt 0 I Imiddot~ I
- I H 1)7
O I GOO I I i ) l)
0 0001) 00000 00000 00000 00000
0 1 1)17 o~ViH
-1 0 1)gtG 01lbullH 0 7 I till
0 l l lt G 0 ~r1
- J bull J1 1) I 0 17 I bull at-lt
lllllrmiddot lt Clll ii IC)
~I Imiddotmiddot N ~middot1 D 1)1bullI f-1 Imiddotgt f IOC ClltHI rm uw-- i ll
O(lOOO 00000 00000 () ()(J()(J
(l(t(()()
0()()00 00000 00000 (1(1(1(()
00000
000fi7 OOOGO
-G Jll7l 0011G 1 H20
00000 00000 00000 000()() 00000
0 0 I bulllCi 001w
-4 iHd 00101 I ~700
OOIOH 0 0 I Ilt
-(1 j l)~G
000(i 1tHH
--
--
L c~
U c t c
(
J
J -
C
C
c g 0 0
+ + e--c~ =~ - r~ C- l~
--
+ +---
- -C 0 o- -c--1
I
lt t w
lt
C
ta J ~ ltCi
lt ~
a---~ -- c gt C 0 rtl 5 -
VJ C f-rJJ c
[J_
ltI ~ C z
z _ ~
~ lte 0 ~
C
tJ
e
C
~ c C
L N
~~ C 0
C
l~ 1 0
t CI ~ ----~ -
i-
cc cc + + t=~= g~ --
-~ C + + t~ ---[w (~ lC
-
ccc ~t c5 ~ ee 00000
co--oc I
- - - - -_____
------ _ - _ _--__ ---__ ---- _ - _
ooocc
rr ~ c-~ E lt
C
-
~ C g L-
gt o + + ~~ ( L C- lN o -l
t l - -I
~ - L~ [ti e- [
l _ 1
- - ~_ C
gg -- -- ~ -0000-tr--- ____
----C __
- __ - _ --ggggg
- -
C
+ + -- --- -~_shy-----_ _ -----___ ooo~
-----__ ___ _ - - - _ - _ _- - - - ___ 0~000
--__ ---_ __ -- --c5 ooco
C cc
1 II II 11 11 II II
z
-
r ~ c
z
~ t lt i C z __ -- -~tt ~ middot~ ~
-middot middotmiddotmiddot _ - -----=___
_ =~ - _ -=shy ~
-- -~== 2~= ~ () V)
_middot ~ c( c(
0 er ~- ~ ~ lt 0 0___
-
- amp lt~ ~~~t8
F-21
) ) ) )
TOT1L AllE SCiNrmn TIIT1 L ll lbull1 tW FI 1lFll TIIL1 11bulli 1~lllbullll lllIIOT lIIJClION -11- OV TUI Mlbullrllll1Nft TOT1 I VOIIHIIbull 01- I Ill TOTmiddot I Imiddot I I 11-11~~ COi l NTlbulln
_LI__l_ ___l 4 bull P 7 bull P A R l n A 4 R 1 n 4 4 ___ 7 A ___ _RA____ 1 A 1A 8 1 SL__ 1 R 5 1 ~ 3 _1] ________ 111~____JJ__) 1 ~---~-~ 4 9 1 ----1___11
A0 1A4 JR 17 lAR A 41 4C 174 A ll~ JnR ll7 9 5 R12
----- --------- - ---- ---- -- middot-middot --- ---- -- ----middot-- ----middot ----~ 7 c ~ bull n 1 1 2 l 9 -s 1 -i s s ~ c c__ bull A____02 1 1 2 1 r n 1 c 0 r-- l ~ 1 o 1 0 1 t 1 o c- ----~--- -Pl Ol OP lA2 11A 61 c 0 ~ cq03 l 5 4 l 4 4 1 i r 7 -i r) c q 1 deg 1n
qA RO 107 ~_12_7 SA c 7 A04 _l _ s_ _________l 7 l RA____ 7 ______l _o 4 __ l nn_________ 0 _7________ R_l __ o ~ _____ J(__7 11__r ___ _t-__ o ____ ln ___ h_ ______ An____ c-_n __05 ]lA 140 170 7cA __2n1 qc- 7q 1n1
c 2c 7 lR 1 O 406 11 c A 0 4 n 3 4 7 l 4 R ~ 7 7 5
___ ___~ c Z bull A ~R_ ---2_h 2 bull bull 4 e R bull C07 l_ H l_ ri ------- R1 --- c ----------- AQ _middotbull 4_p____ r 7 ___1[_2_ _r--_______ _ A____________ )_3-____ _____ __t 9 ______ ~_____ _3 __ --~___()_______ __I__08 C7 lR9 9A 1n~ ll7 S AS 84 ~Q 33 11 11 1n R 3A 3309 7RJ 1R lA 14-1 17fI 77 4R 7(_
---~- L~ 1 7 -~ 1 7 4 1 l 1 9 S 3 9 C) 4 9 A ~ 9 q10 __ 7 c _ ___ _ 2 RA l A ____J__6_R 1 C 4 7 P C 1 rq I
-----~~a-R___-=4bull2_1 e __ middot-middot middot bulls __ - __ IL 9 1 ~_1_ ___3 7 40__19 1 ~~- gt q6_ (- l Q 144 43 _37 _ A f ---middot ~ bull r bull q ~ 9 IL ~ 1 _0 -~ bull 1 A 3420 l r -- I- I q4 l qo l c c 0
sectJ~ BLVDII EJ SEi~D~ 1 nu tmiddot-middotmiddot-middot J _
tl~ r-middotn ----i J middot--1_ ___ ~1-1
A f gtiff_ ffmiddotc YYt)II ~ r bull Ja------ ijlttQ~QW~
___ ~Q -~imiddotbull1 ~H fl(--middot
-lt) bullbull
1 middot
middot--middot middot-- _
middotmiddot_ u - -_-~~middot_1r
~-- ~ _
~ -middotmiddot bullY
j ~ ~ L ~ ll c_l
~ pf-
111-r ~ (i middot) ~
bI middot_FT )-I Imiddot ~middotmiddot () --
I filfltf) J ~3 ilj-t ~ C _G~ llli_A T1 AN B~ACI
I~
603 middot Alondro Pork
l U (
Goll Coorll
~ htyen 1~J-(_
d n t U Wik c rr 1 i t_y of Indus try
(
H-7
Plant Site - Dow ch em i cal US A Pittsburg
I bullbull middot1 Sfl Wi - bull AUIIUlllJLII+
BELMOr r l -- -- 7iSHOfgt ITHACA lN I
N[VAOA lN TGERS llltI
Q LN
GfORC~
72copy~~~~-0AYTON l
I I+
__ middot
1- - -~ middot
I I
- ----------
Plrint lite - ilu rnnt Antioch
) j
(
H-9
Plant Site - Johns rlanville ctockton
a a
LEGEND
CENSUS T1ACT1
51 )I 51 07
1980 CENSUS TRACTS SAN JOAQUIN COUNTY
37
E storaye tankworkiny loss 27x 0-u 5F fugitiv~ emission bxll -
-JLi ildSteJdter l gtX] U
Em~ssion factors are si~~lJ fer oxychlorination processes
Enlission factor vau-5 (~1rbulle Tdilt2~ frcn tne literature (JR13 198U) and
assume 9b iciency for incineration (A-E) It is also assumed above that
EIJC emissions to water are Lf~ uf ~rihsion to citir ard that in wastewater
treatment 100 of EOC discharged tc1 vater is rel easl~d to a i 1middot ~ These emission
factors 1bull1e1middote used to prioritize r-2ieaies iJut were cldrly uude approximations
to tre planL l)ecause of thlt ncine-xt~on system it was anticipated that
fdctors A-E would be ri2dtKcdft 1-middotactor - 1nii~t1t be reduced since a monitoring
program had been in force al20st o~e y2ar Factor G and the off site storage
tanks~ vni~cn are not tied into lt1n ircinerator system rniyht -iominate emissions
Emissmiddotion limits and concomitant regulations embodied in Rule l00S of
the SC-l)MU have necessitctted tlw incinera-cion system Ninety gas chronatograph
probes are located throughout t11e plant including the stack of the primary
incinerator Con cent rations of vc1v1 are reported es sent id I I y be I ow tt1e r~yu la -
tory limit of 10 ppm and in fact below the limit of detection somewhat less
tl1an 01 micropra
lUl MEASUK~MENT APPROACH
Tne objective of tt1e proyram is to determine a pl ant emissions for
EOC It is not accemicrotabl~ to dtmiddotvelop sucll informatmiddot1on basect upon published
indlllstry Jide esti11dtes of plant control efficiencivs Fortunately it was
possible to desiyn a monitoring and calculational progra1n to determine EUC
emissions for tne site and not ctbsorb a disproportionate share of program
resources
Hie re are tnre~ modes ot re I ease from tt12 micro Iant dnd d fourt11
offsite
bull Post-Process 1nci11er-dtion
Processes A-E of ecc ion lll 1 dre a 11 1ented into the pl ant
incinerdtio11 syste11 rt1e conceritrcttion of VCfmiddotl contir1uously iectsured in till=
stctck cts SJdS output remicroresents d rectsondble UjJper li11it to diJply tor EDC
con cent rat i ons s i nce i ts ef f i c i ency of i nc i nera t i on 1s d t 1 east c l1 u a I to tr1 at
134
ot VCibulll Tlerefore knowledge of the system dirflow rnd VCM concentrcttion are
tt1e necessary and sufficient conditions for determining the bounds of EDC
reledse fhe pldnt expn~sse I wi 11 inyne~)s to provide tnese data in order to
support the calculations
bull Fugitive Emisions - Valves Flanges Seals and Other Sources
Fugitive emission from the valves pumps and flanges can be
determined more precisely tnan was anticipated since Stauffer has completed a
comprehensive leak inventory of all such components in compliance with Rule
466 466 1 and 1005 The Inventory delineates all leaks uncovered by their
three man crew throughout tne year and indicates the screening level in ppmv
and component identity In consultation with Stauffer we will be able to
identify the total number of components associated with EDC handling systems
( 700) the distribution ot substance composition streams the distribution
and incidence of leaks by hardware component type and leak rate We will
utilize this information to provide historical data to compliment our Foxboro
OVA sampling at the site dased upon this monitoring we predict an EDC mass
emission rate for the fugitive releases The plant has agreed to provide the
necessary information We propose to meet witt plant personnel prior to the
start of monitoring and finlize the sdmpliny strategy to accomplish 100
coverage of the lines of hiuhest EDC composition The sampling approach and
calculation of mass emission are described in Section 72
bull ~~as tewater
From examination of the emission factors derived from published
literature (see Section 10~) it is clear that EDC release from wastewater to
air is potentially several 11rders of magnitude higher than any other plant
source l1ased upon microl ant ml~a sured concentrcJ ti ons of 8 ppm EDC in water a more
realistic emission factor wgtuld b 24xlo-4 1-iassunit mass EOC produced
Clearly this could still be the dominant source Furthermore the release
point would be expected to middotgte locctted between the plant and the sanitary
district treatment site whi1 his~ kmfrom the plant at 24501 S Figueroa We
believe it is necessary to ndemicroendently confirm the average 24 hour EDC
concentration in the dischar-qe valer by obt1ininy the refrigerated composite
sample We t1ave identified ttie I 1nes of int~rest and received agrt~ement to
sample and analyze for EDC in the stream ~e propose to draw a duplicate
sample from the compositor nd ani1lyze for its rnc content This will be
135
compared with Stauffers para11eI tnalysis It is not reliable to obtain
direct readings of EDC by survey instrument in the air above the water flow since concentrations at crny single point will be i 1 the near dmbient range
Emissions will be cc1lcJLtecl 1_sing the plants vulumetric daily flm
and ass~m~ng that complete degass40~ of the effluent will eventually occur
This assumption is based umicroon the relative volatility uf EDC and tne distances
involved However it should be noted that no direct experimental or
monitoring data is available with which to confirn this Th2 composition of
the discharge stream is unknomiddotwn since it merges into a larqe multisource flow
Conversation with the LA County Sanitation l)istrict (J f1ilne) reveals the
stream to be both exposed and covered Vents exist where me measurements
could De made to assess gross leaka~e at key points
We will obtain samp1es of several plant process discharge streams in
40 ml bottles witn no head space Transit time fr0m sample collection to
analysis will be minimized and wi l I not exceed 24 nours This time frame is
conservat i v e al though ED C i s a v o l ct ti l e mater i al ar1 ct vJi 11 undergo
concentration degradation and outgdssing lnalysi) will b(~ performed in tt1e
SAI Trace Environmental Chemistry Laborcttory utilizing pur 1Je and trap analysis
FID gas chromatography A trial a11alysmiddotis was conducted and the EOC
characteristic peak was distinctive down to the ppb level Therefore the
analysis should easily confirm concentrations in t1e 8 ppm range
Offsite Storage ranks
Three EOC leased storage tanks are located offsite at the Port
of Los Angeles and are owned by annther firm Annual emissions from the tanks
were very roughly estimated based un AP-42 JRB 1980) as 44 kkgyear
based on preliminary estimates of stored quantites
Considering tne magnitude of this source these storage tanks must
be investigated We will gather information about their configuration and
control i n order to ca I cu l ate their e111 i s s i on s bull
136
lU3 UEfERM I NAmiddot 10~~ OF MISS IONS
1031 Incinerator
Vinyl chl 11ride co11centration is monitored dt approximately 90
locations throughoul the pl trlt (Lanyner 19Bl) including the incinerator
output Conc~ntrations are reported by Stauffer dS less than 0l ppm at a
flow rate between 10000 ant 12000 SCFM Although no direct monitoring of
EDC is conducted it is posible to conservatively bound the concentration of
EDC at U 1 pp111 Tl1at conce 1ltrat ion is a suitable ci10i ce s i nee it represents a
conservative bound on tile VCM levels measured near the stack output
Furthermore the molar volume will be taken as 224 L rather than the higher
value it would nave because of the sl i~thtly elevated temperature at the
detector location
Using 12000 SCFM one has thE annual emission of EDC as 12 x 103
SCFM x 283 LSCF x 526x105 minyr x lmole x 97gmole x 10-7vv 224L
Therefore incinerator emissions of EOG are thought to be bounded by 773 kg or
170 1byear
10J2 Fugitive Emissions
Seven hundred (7UlJ) sources were surveyed comprising nearly 100 of
EDC service All accessibl~ plant areas with streams containing greater than
1 EUC were screened except for a small number of devices located in areas
11lere active maintenance was being condJCted It is believed unnecessary to
p2rf-gtrm any emission factor adjustment since it is estimated that gredter than
95 of the requisite components were screened
Three leaks above an arbitrary OVA reading threshold of 20 ppm were
parameters and mass emission numbers fhe upper bound leak rate was
determined to be close to twice the nominal leak rate which accounted for the
response fltlctor of amicroproximamiddotely 05 by Radirn for TLV detection of EDC
( 13 rown 1980) bull
Stauffer found and reported approximately 10 leaks in the EDC
service during 9 months previous to the pLrnt testing Assuming no111inal leak
137
) )
Table 103-1
STAUFFER CHEMICAL MASS EMISSIONS PREDICTION FROM OVA SCREENING VALUES
OVA SAI Actual Upper Device Response Response Cone Source Leak Rate Bound Location (ppm) Rae tor (ppm) a b Se I Re Function lbyr lbyr
Gate Valve Unit P1559A
Valve Heavy Ends Column Unit 14439
Compressor Seal EDC Storage Unit Tl062
I-- w co
300 11 273 -035 096 0 17 153 D 6
11 I500 114 439 II 241 C 1
r III II II o-is lUI bull2000 vl 1 1818 A 9~
All emissions 100 EDC (12-Dichloroethane)
11
values in the nei~hborhood of 2000 lbs~ total could be emitted annually
Therefore it appears ttiat major reductions in the mass emissions were achieved
by the company run inspection prugram and the fugitive emission source has now
become of seconqary importance as a fraction of total plant emissions
lUJJ Wastewater Discharge
Wastewater samples were collected in duplicate from four sites
within the plant for determination of ethylene dichloride (EDC) concentration
The samples were collected in EPP standard 40 ml VOA vials on July 1 1981
Analyses were performed using standard purge and trap techniques coupled with flame ionization detection gas chromatography The results are given in the
table below
Sample description EDC conc~ntration range (microJml)
EDC concentration average ( 1-19ml)
Approximate fl ow ( gpm)
PVC Interceptor Box 82 - 108 95 200
EDC Stripper number Chlorination area C1404
2
011 - 014 012 25
Final collection site for pH adjustmentP663 349 - 3B2 366 350
Final discharge site sanitary sewer 6 - 254 158 500
The water in the PVC lnterc1~ptor ~ox was warm and represents washings from the
PVC reactor which travels t 1J the Interceptor Box in a concrete drainage ditch
Water from the EDC ctripper was vPry t10t and because of its heat was difficult to collect The ~1eat ot trie wat11 r may in pdrt account for tne
relatively low concentratio11 of ElC here as the EDC would out-gas from the hot
~vater more readily The fir1ctl collection site tor pH adjustment is a large
concrete container middotvith mixers in it located just prior to the final discharge
site Water at the final dischdrg~ site is being constantly aerated due to
the speed that it flows thru~yn the concrete drainage trough This aeration
could account for tne relatively large range in the rnc conotent found here
The average EDC concentraticmiddot1 at tne final discharge site is middotbelow the daily
discharye requirement of 25 19ml EDC
139
Plant oersonnel indicdted monthly iverage readings are typically on
the order of 8 ppm but daily averages can re 1ch greater than 20 microgml In
addition LA County Sanitation Uistrict staff (J Milne 1981) indicated that
an on-site impoundment pond can become laden with EOC during certain abnormal
operating periods and discharge Vdriances arl requested by Stauffer This
mi 9t1t Jccur on the order of O1c0 aCii ecH c1 1d t1erefore is no-c expected to
signifcantly impact averag~ cnnJJI discharg ~ values lt s not expected at
this t1ime tEit evaporative eriss uns from this pond are significant except
infrequently during upset conditions and spi 11 control operations Program
staff ~~ 1er2 una111Jare of any periods ~h 1~n EDC cimtent in the ponds could be
appreciable and therefore no sampling was performed
Utilizing the average uf the two final discharge concentration
1readinJs (158 micrograrnsm~ 1 and 1G 1Ji gpm fl)W one has 34600 lbyear of EDC
released into the sanitary sEwer For the pJrposes of calculating population
exposures from plant releases it will be assJmed that the EDC is locally
emitted from the wastewater streams Tl1ere ire no monitoring data available
with which to develop a more accurate release profile
However emissions of me fron the plant wastewater discharge can be
calculated according to the method of Mcmiddotckay (197S) as modified by Dilling
(1977) It should be noted that this mLst be considered an estimate since
conditions of fl ow and the presence of other substances will influence
emission rates
Using Uilling (1977) for non~erated flow first recalculate
Henry 1 s law constant (dimensionless-my of chmiddot~mical per liter of air divided by
mg of chemical per liter of water) dS
1604 P M 1 Wl
TS l
)
wnere
H Henrys law constant dimensionlessl
Pi tile compounds pur1~ componen- vapor pressure in mm Hg at T
Mwi the 11olecular weiglt
T tile absol ~te temperature of he wastewater in K
Si tt1e compounds ~uluhility in myliter at T
Then for UJC
(l6u4)(7~)(9Y)H = = U0447 with 1
(2lJ4)(8690)
LIii bullul1Jl)i I 1 1y 1 middot UL i11 wt1Ler yiven -1L L0degC is 8u4JU UHI tile vapl)r pressure
1 4 psi ( 7 2mm) at 7 0 degF bull
Tl1e overall liquid mass-transfer coefficient Kil is given by Dilling (1975)
as
(2211)(06) in mhr
-f-0-4~ (M _) 12+ 100 Wl-H-
1
=U108111tH
Then from Mackay the percent desorption is given by
c 1 = exp (-Kil tL) where
c 0
c = the concentrd ti on at time t of EOC 1
co = tile initial concentration
L = the Ii quid depth (Ill)
t = tt1e retention time (in hr) of the liquid in the wastewater
system
Retention time is bc1s~d 11pon a flow velocity range of 3 ftsec as estimated by
the Los Angeles County Sdnitatii)middot1 Dibulltrict (1J Milne) over a middotdistance of
ctmicroprox i 1nate I y S km to the Joi n_t Wdter Po 11 ut ion Contra l Pl ant Sewer fl ow
depth throughout the entire route have been estimated by J Milne as typically
1 foot 030m) to the Davison Pump Plant and 3 feet (09m) to the Joint
Tredtment Plant distances assumed to bt~ 1 mile and 2 miles respectively
Trdnsit tiine l)eco111es between 05 tnd LU hours sequentially Then computing
the net emission reduction as the product of each leg
Therefore 26 of the EOC i erni t ted bdween the pl ant and the Joint Water
Pollution Control Plant Rtgtsfdence ti111e and conditions at the plant account
141
for additional releases and residudl content is forther emitted between the
plant cnd the ocean discharge poiiit ~-Herever the flow is r2tained aerated
or shallow emissions will be accentuated
Offsite Storage Tank Emissions
There are three storage tanks leased by Stauffer for EDC storage
1ocated at 22nd and Gaffey Sts ~ bull San Pedro These tan ks a ~e used for long
term storage rather than providing feed on a routine basis Therefore
breathing loss rather than v1orkir 1J lass is )f concern An tanks are white
two being 67 ft in diameter while t~2 third is 57 ft All are 40 ft 3 inches
high and have capacities of e~thi l0~10000 (2) or 840000 gallons The
tanks are cone roof type and are not tied into vapor recovery systems The
South Coast Air Quality r1dna~J~ment u~ str4 ct has based their estimate of
emissions on the specification of 12 foot vapor level Using the AP-42
formula for breathing loss with the vapor pressure of EDC at 20degc and an
average diurnal temperature variation of 26degF one hds for each of the two
larger tanks
l) 173 H bull51 T bull 5 LlI = (619 x 10-S) M( _P_ tH F CKcp 147-P
68 173 51 5 = (619 X 10-S) (9119~ 116 bull (67) ( I 5) (26) (1) (1)
14 7-1 16)
Thus for the two larger tanks LB= 472 lbday
For the third tank 0=57 and = 178 lbdayL8 The total emissions become 23724 lbyear ~ince he plant is in the process
of shutdown it is not clear what the liquid levels currently are Note that
if the material in the three tanks were combined into one totamiddot1 -~1nissions
would be reduced markedly Exposure to a population from this site was
determined separately since it is lbullgtcated over six miles fron tl1e plant
142
110
ASSESSMENT OF PUPULATlUN EXPUSUR~
111 OVEKVIEW
A major program goal wa~ to compare emissions from the various
sources and identify and rank any 11 tlot spots in California where the general
population was exposed to elevated concentrations of carcinogens A simple
Gaussian dispersion model was therefore used to obtain order-of-magnitude estimates of exposure of the genermiddotal population surrounding each source
Since this was essentially a screening study use of more sophisticated models
was not appropriate
It cannot be emphcsized too stronyly tnat most California residents
are exposed to emissions of hazardous substances from a variety of natural and
rnanmade sources Urban dwellers dre typically exposed to greater concentrashy
tions than rural residents however all are subjected to so called 11 background 11 levels from multiple sources In order to place stationary
source exposures in perspective the typical ambient levels of each substance
were identified from the literature and compdred with the concentrations due
to the emissions from each ~lant Exposures were thus expressed both as
absolute quantities and as increments above background
Comparison of plants presents a further difficulty in that various
substances are being consict~red No attempt was made to evaluate the relative
importance of exposure to two different substances such as chloroform versus carbon tetrachloride other than by ambient concentration
112 DATA SOURCES
1121 Meteorologicdl Llata
The dispersion model to be descritgted in Section 113 required input
of annual average wind speed and frequency of occurrence of wind from each
compass direction These data were obtained for most of the sites from the
South Codst Air wuality Management District (SCAQMD) In other cases (Kaiser
Jotms-Manvi l le l)ow and OuPlt)nt) only clai ly cJVerage wind speed and frequency
data were available In al I cases we used data from the meteorological
station nearest the modeled e11issrnn source Table 112-1 summarizes the
141
Table 112-1 METEOROLOGICAL DATA BASE
Observation Site Data Period of Pldnt (Distdnce Source Observation Remarks
from Plant)
A11 i ed Che11 i Ca 1 Lennox SCA~MD 1Y71-1974 Averaged hourly readings available ~ 1 Segundo (3 111iles)
S ff~r-- Uemical Long l3each SCA()MD 1962-1974 Averaged hourmiddoty leadir19s avai1able offsite Carson (3 miles) storage tanks appruxi1ntL~ly 5 miies from
Figure 114-3 Cumulative Population Exposure to Carcinogenic Polycyclic Aromatic Hydrocarbons From Coke Ovens at Kaiser Steel Corporation Steel Mill Fontana
CV)
0 -I X-V) V) a _J
s 0
C 0 micro co s micro C a u
l ~
14
1 --
J- t_bulli
_ C
uw m 0
1 -0 a bullbull micro
microdeg ()
0 micro tbullmiddot C QJ V)
0
X LLJ
C
middot4middot~middot C 0 micro rt1
-- - 0 ~ 0
0
n __ _______bull~ -______ --al_ - ~_
I bull
I I - bull
middot 1
-~1
l i l ~l
~ 0 I 1
Li bull I t_~I
i bull -1 _1-J
0fmiddotbullmiddot-middot+-middott-t-r+-t-+-+middotr+-t--+-1--t+ t f-t-J-t--r-+tmiddotmiddot~~bull+-t--1~~bull-+-1 Id middot t 4 E1 (
I ncrementa 1 Concentration (~gm3)
Figure 11 4-4 Cumulative Population Exposure to Asbestos From Johns-Manville Plant Stockton
I
)
--~ ~w t _ ---~- ----------~ --____~-=~ M
~
bull I bull I I I I ii i Q t I i E e- I Ii tl I I I 6 I a I I J I f t t bull I G fl I
a I e 1 1 a 1 1 1 1 iii bull bull 1 1 bull 1 1 bull 1 1 bull bull bull bull 1 1i bull bull 1 1 bull a bull 1 1 r ~ middot u ~ ~ I0
+H~ ~ ~~~~-~~~~-rmiddot-~~~~~~~~~-~--~~-~~-~t~~~~~-~J C
~1_ c bull 1 t~ bullbull bullabmiddot 11 l bullmiddot II bullbull C- I u~
0 rl
gtlt I I I- 20 () ()
CJ -1 ~_J
J Cbull s
I If I I I I0
C bull I Ii I I I I I I I I0 1E
bull-+l I t I I ro s Lower+l
I I I G I iii I bull
() 14 I I IC
u I0 I I II Iu
C
1-_~I fi
CJ ~ I-- (j) CJ I G I I p +l
ro +-l I I I1~3~V)
0 +) UDf)Pr u I I 11 I I
CJ 1iII
Vi I 0 I I It- I a X
0 ~
w I I I I bull
Lower C 0
middotr-
- 4+)
(1j
J Q_
0 bull1
ij a 3- ti I
D d I I I I
I I I bull
II B I t
i ] i amp I
1 ei ~ amp 3 ii
I I I I
e ft a a I bull G
I I I bull I
I I bull I I t If
UpPrf
I I
I I I-_L~ lI I I f
I I I I I
- I I
j I I II I -
lI ImiddotbullII
l I
l Li
Incremental Concentration gm3)
Figure 114-5 Cumulative Population Exposure to Perchloroethylene and Carbon Tetrachloride From Dow Pittsburg (Solid Lines) and to Carbon Tetrachloride From DuPont Antioch (Xs) Results For Upper Lower Bounds on Emission Estimates Are Shown
1i-1 - middotbullth~middot t =lttF--i-+-middotr+-r+middot~middoti-middot-~-4-4middoti t -~--is -~middot-rmiddotibull--+ u ~~ i--t4~ t-bull~~311 li-i1middoti- C - JJ _ CL
-- I 1 t middot1 c7 Ibull ) ~ I-middot bulli C bull 1-_ 11 11gt1l11 ~bull -a bull-bullbull t a- II bullbullbull bullI lt_I Iii _- I
Incremehtal Concentration (microgm 3)
Figure 114-7 Cumulative Population Exposure to Carbon Tetrachloride From Allied Chemical El Segundo (Curves Corresond to Upper and Lower Bounds on Emission Rate Estimate)
--- I
~__~-unt2wbull--o1WCUatJ41ia1
16 0amp (V)
0 ~
gtlt
14~11l 1l QJ _J
~
1 middot-0
c t1middotmiddotmiddot 0
micro ro
_ c micro ~ 1euro1pound1
0- QJ u-J c 0 u O =-middot ~-= QJ
c_1 micro ro +- V)
0 micro ~ t1 O QJ 1l 0 a 40gtlt w c 0
-
ro --
micro
20middot l ---a I ~ I a 0
0
pound1r-t k
Figure 114-8
8tWI-S~1111111N~-
a I __ I wi~ i awLi--~~ 2 I I
1--+ I ~--if bull =~~--bullJ-+--f-1-----middot+--+middot-middot--r9c- 1 c- J
bull _ bull -~
Incremental Concentration (microgm 3)
Cumulative Population Exposure to Carbon Tetrachloride and Chlorofonn From Allied Chemical El Segundo Assumingsix-months Operation With Each Feedstock (Curves Correspond to Upper and Lower Bounds on Emission Rate Estimate)
i u---J--bull-+ i i--+ --4--+-imf-bullbullQ~-+-+-+~bullbull-4 171 i R = Fi middot1 1bull1 r1__ bullltaa ~a
Incre0ental Concentration (microgm 3)
Figure 114-9 Cumualtive Population Exposure to Carbon Tetrachloride From Allied Chemical El Segundo During Offloading From Midnight to 6 am middot-middotmiddot------middot--middot
--bull ft ampLA 1111-~lliaIN_WPl1A 1JIU_Mt91 MJ1WWlaquoPI-W1111~ --------__----Et1
M gt~middotI
0 -I
-X
~ t~ bullM
V) middot-middot ~ V)
QJ _J
f 0
0 -
C -4 0middot~middotmicro re f micro C a u
10 Cdeg u 0 3 0middotmiddotmiddotbull~ -0 a
+-gt ro
+J V)
+-gt 0
2E1-0 QJ V)
0 a X
LLJ
C 0 1pound1--middot+-gt ro
-- Q_ 0 0
I I
~ -~middot bull I -bull-_ bull I
middot lbullgtbull___I I
frac14 llk I
I bull
-~
I
-1_ I
Ibull II__ JI
I~
(1 f---1-+ bull-r_--f bull middot~-bull +-- bull i -bull 1 -1-+--+bull-+-t-+bull--t1 ~ _ 4middot 6 E 1t1 1middot2
1 3 5 I middot~ _ 11
Incremental Concentration (ngm3)
Figure 11 4-10 Cumu1ative Population Exposure to Ethylene Dichloride From Stauffer Plant Carson
_~ ~ t M
C X E k1~
-
- c- 1-1 ~ C middot-middot -_J
I--- -J 4t1middotbullmiddotmiddota
D
-+-shy --middot liiit1J
r ~
ltL
t-1 _ V
0 -+-1
middotr -bull ~1
X
C
+- bull1 ~1 atr l -=
~~lll(k~IHmiddot~~-Wa~middot~nL1middot~~~ta KF u n P middot~dE-tveRS ~-tl~~~-u-~~~tmiddot_-~~~~liil-~-~~~ tc r
I bull
bull -r~l-yen--~1l~a l fl __ A4o bullbull bullbullbull ~ 111bull--middotbull
8 NJ industrial sites 28 samples in industrial areas with benzene consumption facilities Torrance CA Tuscaloosa AL National Forest 1000 sarip Ies-Torontt) Azusa CA El Monte CA Long Beach CA Upland CA Los Angeles CA Riverside CA
bull Assumes all year operation on either feedstock If plant operates 50 on each feed the population exposed to greater than 50 increment over background goes to 16025 - 19377 and 4044 - 16025 for 100
Ambient concentrations of benzene vdry over one order of magnitude in the literature and therefore make tnis calculation questionable For Kaiser therefore the carcinogenic PAH assessment was used to evaluate incremental population exposure above background
+ The partition of emissions from Oow are 78 carbon tetrachloride and 22 perchloroethylene by weight
173
1142 Comparison of Incremental and l3ackground Concentratiors
Those sites whicn elevate background concentrations greater than SO~~
City Kentucky 6600 The latter two are associated wi t11 EOC pl ants Data
taken in senFice s~ations ano lrctffic areas ir various cities range from -~
300-3640 ngm~ As previous~J mentioned the Stauffer plant is discontinuing
operations These two sources should be examined as part of any possible
start--up permitting activity
11422 Kaiser Steel Corporation
The Kaiser steel plant polycyclic aromatic hydrocarbon (PAH)
emissions cffmiddotise from the coke oven omicroerations Comfdrison of the cumulative
population exposure Figure 114-3 and Table 114-4 with the specified
background levels for urban areas illustrates the breadth of the exposure
distribution The five known PAH carcinogens that were isolated in the coke
oven emissions were quantified in tile ambient air of Los Angeles by Gordon
1Y76
Benzo[a~pyrene 046 ngm J
t3enzo~e]pyrene osio Benzaanthracene U lo
Chrysene U60
lndeno Cl23-cd~pyrene 134
Althouyn these concentrations ctr~ low co1npared with 1 number of other cities
cited in the literature and repres2nt a very I imited data base tt1e predicted
concentrations from the Kaiser plant ~enerally exceed these levels by d
174
significant margin ie greater t11an 30000 are predict2d to be exposed to
4redter than l()-lt t 1tbull 1111tlit~nt background of 3j nqm 3 It ic 1 iklly tlat
lHrllerte exmicroo~urt 11 Ilie drld dre also 11evatPd ovr r11nbient I1owever since
backyround concer1tr1tions of t)enzlmiddotne show large variation no population
calculation was specified
11 4 bull 2 bull 3 Dmv
Carl1on tetrdcnloride releases from Dow constitute approximately 78
of tl1e total CT plus perc emissions in Table 114-6 and were found to elevate
urban background concentrations greater than 50 in five census tracts Emissions are predominately from storage and check tank working and brething remiddot1 eases
1142~4 Allied Chemical
The Al lied plant was modeled several ways since the plant can
operrttt~ wiUl chlorotor111 dr cc1rbon tetrctlt hloridt fo(~d The c1bull-i pr~middot11ntlid in
ldlgtlt~o ll4- drld llil-B represent dflrlUil 01Hrdtion with eitt1er teed Partial
yedr operation with each feed can be scaled from the individual annual modes
and is presented for equal hltllf year operation in Table 114-9 As with the
Du Pont plant carbon tetrachloride emissions arise from feed tank working
loss fable 114-10 illustrates peak exposures predicted to arise during an
off loading cycle As expected concentrations in that six hour period far
~xceed annual average values Insufficient data were available to contrast
levels with backyround transient concentrations
175
l~O
AUHLAlHJ~ CONTROL n CHNil)UES
Alternative control approacnes for the most significant emission
sources dinmig ~he various pl1ants n ees fitie(J n the follo11 ric suJsections
EmpiiasVi nas Deen placed in dedlinq ~riU1 Uiose sources of 91reatest absolute
inaynHud2 (lbyr) and those const1 ut 0ing trie largest increm121middot1t middot] bulli~he
bdc kgrnund con cent ration of Hie em~ tted subs tcince i~o ctt tention Wi 11 be ~i ven
to the s~iondc1tY sources witt1middotn cK1 microant or to trlL case of J011ns-Manvi I le
since -UH imiddotajor source is less tran LOO 1byr and is not microrejicted to raise
bctckgro 1wd exposure levels to the ~~ntr21 population Furt12rmore a11
emission sources dt Kaiser Ste 1 1iC(1 v1 1~n directly dedlt -1ith in this
program r1re elltlteci ro t11e cc1lt1bull ovcmiddotr nperatmiddotions These faci 1ities are to be
closed du1rm and all primary reel 11i~ oH~rdtions discontinmiddot 2d Kaiser
forecasters have predicted further deterioration of the plant economics and
the phased closure has been accelerdted for primary steel making operations
Note that this closure is essentially irreversible cince differential
exmicroansion and contraction of the coke oven structur1 occurs in the cooliny
process and it would be improbable that ovens could be reheated without
extensive reuuildiny at major expense
12l STORAGE TANK EMISSIONS - fAUFFrn IHJW OUPUNf ANU ALLIED
At c1ll four sites emissions from storage tanks constitute either the
primary or near dominant (Stauffer) source of carcinogen release andor
genero1i iJOpulation exposure Curreritly the tanks of interest at each site are
permitted by the local Air Quality IJistricts hJwever they do not require
emission ontrol systems for vdrious reasons Tl1e estimated releases grounds
for exemption and other pertinent information are given in Table 121-1
In order to amicromicroreciate tth~ practicdl dlternatives for emission
controls the provisions of SCAWMU Rile 463 ar~ listed below which specify the
acceptable alternatives for tdnks r4uirin9 controls ie tanks 11aving
capacities greater than 39630 yal with suostances of true vapor pressure
176
Table 121-1
Site
Stduffer Uff-Site Tanks (J) (Sdn Pedro)
Allied (El Seyunrlo) - KdJ
1bulllctLer1 d I reld iturctjc
-J -J
Dow (Pittsbury)-Product Check Tanks (4)
Uow - microroduct sturag~
DuPont (Antioch) - Raw Materidl Feed Stordge
Approx Capacity~
6lUS x 1g (c) 84 X 10 (1)
lt3Y630
18000 each
70000L) 3UUOUO
30000 working 42636 liquid full
STORAGE TANK
Substance
Ethylene Uichloride
Carbon Tetrctci1 I Jr i J1
Perchloreshythylene and Carbon Tetrashychloride
CT perc
Carbon Tetracr1 l ori de
SUMMARIES
Emission Mode
Tank breattli ng
Tank v10rk i ng Tank breatttiny
PERC-~JOrk i ng PERC-b r2a t ~ i ng CT-working CT-breattling
Tank breathing Tank breatt1i ng
Tank workin9 Tank breathing
Vamicroor Pressure
014 at 70 F psi
1 63 psi at 68 F
16d psi at 68 F(CT) UJ9 psi at 68 F(perc)
168 i)Si dt 68 F
Grounds for Exemption
Exempt from control re4uirements to SCAQNJ Rule 463 since vapor microressur-2 dues nut iXC~ec
l~ io at sturdJ~ conditions
txe11pt TlJil fule -+b3 since tank capaL ~J ~ s u~der 39630 gal (lSO 11 )
Exempt from Regulation Rule v~~vu u~~~~~shytank capacity is under 39630 ga 1 andor substances are not classified as organic 1i4uicts
Exempt f ro1n Ru l ~ 85300 because subsLctnce is not classified as dn organic li4uid accordin to Regulation-Rule 8120
exceeding l~ microsi at storage conditions
~ floating roof tanks
~ fixed roof tanks with an internal floating type cover
bull a vapor recovery system with vapor collection and retllrn (or disposal) proce5sinJ lXctbullerlin~ 95 efficiency
ln a cllemicai plant a typical vapor recovery systein (KR Evans
SCAQMU) migm consist of a collection inanifold to d recovery system (suctl as a
vapor sphere) to a compression syst2m dnd subsequently to absorption and
recovery systems The absorpt~on st~n c~nsisting possibly of (rubber
stripper or activated carbon
In dealing with speciti( carcinogenic substances such as vinyl
chloride hiyhly efficient vapor control system perf0rmance has sometimes been
stipulated necessitating incineration systems
Fer th~ cases of corccrn realistic alternltltives are as follows
Stauffer Off-Site TanKs - Tnere are a number of options which can 1~arkedly 3
reduce ei1issicns from these tanks from th~ calculaL~d value of over 20 x 10
lbyr One alternative is to consolidat~ the material in tne three tanks
One of the large tdnks can contain the currently stured material and reduce 3emissions to apmicroroximately 13 0 x 10 lbyr Anotl1er alternative is to
transfer dll E0C to a single floating root tank Various types of such tanks
exist dlld are reviewed in the EPA report Lrganic Cht~mical Manufacturing Volume
J Storage Fugitive dnd Secondary Sources EPA-450J-80-025 eg internal and
external floating roofs and a variety of seal desiyn configurcttions Generalshy
ly such designs would be expected to reduce emissions to the order of
one-fourth to one-fifth the current level Cost of 1nstalliny ct contact
single seal internal floating roof wds estimated by BB Lumquist Pentrex
for EPA as $27770 in 1Y8U dollars for 7U ft diameter tdnk Cost to build an
external floating roof tank was estimated by G Stilt of Pittsburg Oes Moines
for EPA as ~14UUUO for D=67 H=4Uft Anottler com111on approach is to utilize
cdroon ctdsorption This works we11 witn nonpolar hydrocarbons dS VvC is
removed from the vapor phase A basic syte111 consists ot tvo carbon beds and
a regeneration system Regeneration is typically performed with steam or
vacuum Figure 121-1 illustrattS rhe systems (Basd2kis 1~110) Steam raises
tne VOC vctpor microressure The resulting sti~am-VOC mixture is condensed and
173
AC-TJA~D CA-e0-l
-----~---
LOW- ffiESiJ~C ~T~H----
PiltOCESS COOL-J~ At-JO oLOWER DRY t-J~ 2gtLOWE~
COgtJDENSER
J
COOLIN(a - --------
WATi=-A
RECOVeRE SOLVENT
WASTE WATER
Fi l 1 1 middot gure middot - Activated-c arbon Adsorption System
179
rout2d to c ~epordtor decanted and returned to storage In vacuum regenerashy
tion VOC vapor is desorbed by pu1 ling a vact1um on the carbon bed then condens-
ed a1d returned System efficiency ~ ~ est ii1a ted at 96 f(duc middot i m from fixed
roof levels (EPA 4503-80-025)
H~fri 9ctated vent concicr1sers dre une of the most cu1mon emission
reduction prJcesses -for conttol ~ ng hx2-d-- storage tai vUC Figure
12 1-- 2 i l 1ust rates a u n i t ( Er i scmiddot n 19HU ) eff i c i en cy of middot middot~ ~ c middotlt middotmiddot _y i s rate ct
bet1Jee~middoti 60-90 for the vapor pressure range of concern For such a large tank
the capital costs would be high ~igure 121-3 illustrat~s EPA estimates
(EPA l ~~SOb) for the condenser section Condenser sys tern a ea mu l d be in
excess of 1000 ftL
It should be noted t -at Jtc1 uf fer Uiemi cc1 l has arH1ounced the c 1 osu re
of its VCPVC plart PrevfoJsly tt12) had planned to purge tlle off-site tanks
of EUC Since any future po~sible micro1ant stdrt-umicro will necessitate a compreshy
hensive SCAQMD review this ltoument can assist in evaluating proposed control
measures
Allied and DuPont Feed Tanks - In bott1 of these cases the nore significant
quantity of emissions arise from tank workinltJ ie during the off-loading
activity rather than tank breathing Contnl ~easures taken for working
emissions are thus of primary concern Ther~fore no detailed discussion will
be provided on the alternatives for control gtf bredthiny emissions Additionshy
ally both tanks are in the range of 20U00 gallons which is a capacity where
floating roof tanks are almost nonexistent Out of 670 floating roof tanks
surveyd by EPA less then 15 were smaller than 30000 gallons in capacity
Therefore the utilization of any floating roof concept and seal combination
will not be considered
Commmerc i a11 y it is estimated by Ou Pont that carbon tetrachloride
feed costs approximately $017lb (E Taylor personal communication)
Ttierefore less than i2500 is lost to l)uPont annud l ly as a result of working
l eve l em i s s fo ns and 1es s fr om Al l i ed bull Thus i t is u n l i kel y that any
appreciable economic incentive exists to dev1~lop ci vapor recovery system
However a Cdndidate system could be pattern~d after the chloroform
feed-storage unit currently at Allied This is a dedicated vapor balance
system Chloroform is off-lndded from tank cars and stored in a closed
unvented tdnk As the 5tordye tank is filled the c1ir spdce d1spldced is fed
Figure 121-3 Installed Capital Cost vs Condenser Area for Various Materials of Cunstruction for a Complete Condenser Section
182
back to the tank car Also in this case the storage tank is not vented in its
normal breathing mode and is part of a closed feed system to the reactor
Vapor recovery system alternatives which are particularly effective
in loading and handling include r~frigeration adsorption andor absorption
Control efficienci~- 1re est1mateJ in the rlt1nge of 90-95 (EPA 1980b) and of
course depend on the speci tic su1)Stance and equipment used Carbon adsorpshy
tion systems were discussed above The smallest capacity carbon adsorptionmiddot
system priced by EPA (EPA 1980b) has 2 vertical beds of carbon (900 lbs - 4ft
diameter by 3 ft depth) witl1 an ilstalled cctpital cost of $135000 based on
December 1979 dollars
Dow Tanks - Dow has two pair of p~oduct check tanks which alternately are
filled dnd off-loaded Emhsions from th~sl tanks were calculated based upon
operating cycle (3 day fi 11) satubull-dtion vapor pressure and physicdl
chardcteristics Direct hec1d-spa1e testing was planned however it could not
be accomodated because of r 1stri c ed access due to unscheduled mai ntendnce on
the field test day
Dow has indicated that chey are studying the option of installing a
vapor controlrecovery systt~m in 1hese and their largerproduct storage tanks
The extent of their engi neri ng ind assessment work is unknown as are their
current plans It may be possibk to incorporate a vapor balance design into
the system whereby the disp1aced apor is trdnsferred to another process point
within tl1e system Alterna1ively a vapor r1middotcovery system such as carbon
adsmiddotorption is fedsible Hot-1ever since emissions for t11e check tanks are
primarily due to working loss tht~ use of a conservation vent or an adjustment
of its operational differential p~essure would be ineffective for tr1e
reduction of the bulk of such emissions Furthermore since check tank size
is re1at i ve 1y sma 11 no cons i derctt I on was given to conversion to a fl oat i ng
roof configuration for those tanks Conversion would be possible for
the large storage tanks
12~ WASTEWATE~ EMISSIONS - STAUFFER
Emissions of me throug11 ~-a~)te-1dter discharge are the largest single
source identified at the plctnt sites The discharge limit was set at 25 ppm
by the Los Angeles County Sc1nitation iistrict and was estdblished with the
occupational limit in mind of 50 ppm JVer an 8 hour work shift dt the District
18~
tredtrient center~ T11ere hctd been suine di~cuss1on between pdrti~s of possibly
raisiny tne discharge limit since it is felt by Stauffer that dilution of the
stream is sufficient to d 11 ow it Itmiddot shou Id he noted t10-1ever tna NIUSH has
recommended the permissible exposure limit be reduced to 5 ppm averaged over a
work period of 10 nours microer ddy 40 ours per Ieek witt1 a cei l iny level of 15
pmicroin dVeraged over a 15-mi nute microet~ v(J
It is not certain at 1~hrt rdte me 1rill be reledscd from the
wastewater stream At the microlant discharge points wastewater is both hot and
aerated thus favoring release~ No measurements have been ta~en downstream of
the plant after considerable dilut1on has occured The distance to the water
treatment pa11t is approximately 5 kin at 1i1ic~1 point anerobic diyestion is
conducted lt is presumed that a11 ELlC will be released before final ocean
di scnar~Ji
A possible emi ss i D~ cont ro process for nmiddotduct ion of the EDC in the
effluent is by ~recess adjust~Pnts or additio~s For example process
modifications to the EUC stripping stage coula dramatically reduce discharge
lev~ls A control alternative is the use of octivdted carnon or XA0-2 resin
to recover EOC in tne discharge stream Tests of Gulf South Researd1
Institute on XA0-2 and activated carbon (Coco 1980) show high recovery yields
for nonpolar organic carcinogens unaer a range of concentrations Viable
suggested alternatives oy Srnitn included regeneration of Ue trapping
materials dnu even consideration of burning tt1e concentrate carbon media (at
greater than 1000 pmicrom)
Control approaches to reduce emissions from so called secondary
sources in general and waste water emissions in particular fa 11 into four
categories - waste source control resour(e recovery alternative disposal and
add-on controls Alternative control pro(~esses which were considered but
appear to be inappropriate to this case i11clude chtmiddotmica means eg
neutralization precipitation coagulatior1 and chemical oxidation thermal
destruction of tne unconcentrated ~dste s1ream is ii1practical biological
treatment eg aerdtion and biomass-wasteater contcct ~ienerally relates to
the treatment of soluole degrddable organics in the concentration range
bet~~een U01 and l~~ terminal storagt eg lanctfilliny urface impoundment
and deep well injection dre either indppl 1cable or 1mmicroractical Therefore in
summary the poss i b I e contra l approac ties fur this ca e inc 1ude the improvement
r
184
of semicrodration efficiencies in steam stripmicroing tl1e internal recycle of ~aste
stredms ctnd the ddsorption b activdted carbon The design configuration
~fficiency and cost )bviously depend upon numerous microlarit specific factors and
their determination vJOuld rec-Jire detailed analyses
n1e ~~dsrewctter systbull-m of a model cl1emical production plant based
upon the average properties uf a composite of 30 chemicals was evaluated for
EPA by IT Enviroscience (EPA 1980b) Included prominantly among the 30 was
tUC with the hignest uncontrolled secondary emission wastewater release rate
ie ~ percent of tt1e production and 34 perc~nt of the emission Cost and
impact analyses were evaluated for alternative control systems to reduce
secondary voe emissions from wastewater Four systems were considered a
carbon adsorption system (eAS) for recovery of the voe from the wastewater a
cover to reduce secondary VOC emissions from the wastewater clarifier a cover
for the clarifier plus a carbon desorption system and a cover for the
clarifier plus ct eAS system usinu a fume incinerator The scale of the model
system was greatly in excess of the Stauffer plant thus further making
detailed comparison i mpract i cal bull However emission reduction factors ~-1ere 6diven as ~9 and cost effectiveness per 10 g reduction genera1middot1y ranged
between $450 to 1733 These factors would likely grossly underestimate the
system cost if scaled down to the range of the Stauffer plant ie the order
of 34600 lb annual discharge
The alternative approaches of improved steam stripping efficiency
and internal recycling of the stripper iischarge stream with a reduction in
makeup re4uirements could decreas~ net ~DC wastewater content release
123 CONTROLS FUR SECONDARY LEAD S~1tLTER STACK EMISSIONS
Given our findiny of low (16 kgyr) ~missions of arsenic from the
reverberatory furndce at RSR it middotJOuld ippear that the arsenic content of themiddot
lead feedstock is low andor that the microants system for reducing lead
emissions is dlso quite effective for arsenic ~SR it will be recalled from
Section 32 uses a quenchinu cnariber ctnd Dayiwuse filters to remove
pctrticuldte matter and a carbonate scruooer to remove sulfur dioxide
There are no federal new source microertormance standards for lead
eini ssions per se t10wever the Standdrds of Pbullrformance for Secondary Lead
SmeHers (40 CFR 6Ull2) limit total particuLte emission from a blast furnace
185
or reverberatory furnace to SU mgm 3 According to Augenstein et al (1978)
who reviewed the techno1ogy for control ing lead ~missions from ttwse sources
baghouse filters or wet scrubbers are yenerally used to contro1 particulate
emissions When fabric filters are used to control blast furnace emissions
they are nurma11y preceded by an afterburner which inciner~tes hydrocarbons
that bull~GU Id otnerwi se b 1 ind t ne f o1 ur-i c After-burners a re nor necessary for
reverb2middotmiddotatcry furnace emiss-ion smiddotnce the excess air and temperature are
usually sufficient to oxidize tn2 hydrocarbons
According to Augenstein et ali 11 sr1aker--type baghouse filters are
the 1nost effective means of controlling ead fume emissions from secondary
furnace opErations 11 Collection etficiencies can exceed 99 oercent One
advantage to this control approact is that l~ad oxide dust can be recovered
easily cmd recycled in the smeller Flue gases must be coated to below 300 degF
for dacro~ bags and to below ~00 degF for fiberglass bags (Hi~h et al 1977)
Although wet scrubbers are effecti1e under some circumstances in
controlling lead emissions it is more difficult to recover the lead oxide for
recycling In addition sulfur dioxide presi~nt in the flue gases becomes
oxidized to sulfuric acid and can cause corr1sion problems For this reason
sodium carbonate or other basic reagents ar~ added to the scrubber solution
Although it concerns a yold smelter an approach described by
Marchant dnd Meek (1980) provides an exampli~ of an arsenic control
alternative which might be apmicrolicatgtle to secmdary lead smelters At the
Campbe1 1 Red lake Gold Smelter in Balmerton Ontarion Canada Smelter gases
are first passed throuyh a hot electrostatic precipitator (ESP) The ESP is
heated so that the arsenic (which is principally in the form of As ) remains2o3 gaseous and is not yet collected This exclusion of arsenic allo~~s the ESP to
recover particulate gold rnore easily The ESP exhaust is then quenched with
ambient air to condense the arsenic trioxide Baghouse filters then remove
the arsenic along with other particulate matter
124 CUNT~OLS FUR STEEL MILL EMISSIONS
Given the imminent and irreversibli~ cessdtion of coking activities
at Ka 15 er Stee 1 Corpora t i on a rev I e~ o r t e c Imo 1 o gi es for ( on t r o 1 l i ng
~missions from the coke ovens rnd he Clt1ke byproduct recovimiddotry pldnt WdS not
deemed to be necessary Tnis is t1e only primary steel mill in California
1Pi6
KEFUENCE)
Al lf~n Jr Ll~ llJBU1 J 111odel to 1middotstimite hi1drdous (bullmissions from coke oven Juurs pr~pdrelt1 DY 1-kSecircn Tr1dnye7nrffut fur U~ Environmental Protection Agency Uffi ce of Air ~ua 1i ty and Standards fltesearch Tri ang I e Park Nortn Cdrolina KTl17362-UL
Jl len Jr CC 1~8Ub tstination of chdr in(i emissions for by-product coke oven microrepared by fltesedrch Tr1anglt Insc1tute for US nvironmental Protection 1-yency Uf f ice of Air l)ua l i ty and Standards Ke search Triangle Park North Cdrolina RTI17362-0
Allen Jr CC 198Uc 11 Environmeritdl c1ssess1ent of coke by-product recovery plants Proceedings First Symposium on Iron c1nd Steel Pollution Abatement Technology Ctncayo Illrno1s--Tin1ctober - 1 November 1979 EPA-600-9-80-012 pp 7~-dd
Augenstein ur1 T Cornin et al 1978 Con_trol techniques for lead air emissions prepared by PEDCU Envircnmental Inc for US Environmental Protection Agency Research Triangle Park North Carolina EPA-4502-77-012-B ( NT IS P88U-197551)
jarber WC 1977 Interim air pcllution a~sessment for eight t1igh-volume industrial organic chemicals 11 US Environin~ntal Protection Agency memorandum to Air and Hazardous Materials Uivision Kegions I III-X and to Uirector Environmental middotPrograms Uivision Region II
t3drrett KE WL Margard et al 1977 Sampling and analysis of coke-oven door emissions Prepared by ~dttelle-Columbus Laboratories for US Env 1 ronmenta I Protection Agen y Industri a I Environmental Research Laboratory Research Triangle Park North Carolina EPA-b002-77-213
t3 a r ret t flt bull r bull and P bull K bull ~~ebb 1Y78 bull 11 E f f e ct i venes s of a wet el e ct r o stat i c precipitator for controlling PUM emissions from coke oven door leakage Presented at 71st Meetiny of the Air Pollution Control Association Houston Texas 25-29 June 1978
l3asdekis HS 1980 Carbon Adsorption Control Uevice Evaluation IT Enviroscience Inc report in prepdration for US Environmental Protection Agency ESEU
8ee fltW et al 1974 Coke oven charging emission control test program Vol I c1nd II US Environmental Protection Jgency EPA-6502-74-062
~eimer RG 1978 Air pollution emission test Analysis of polynuclear aromatic hydrocarbons from coke oven effluents Prepared by TRW Uefense and Smicrodce Systems Groumicro for US tnvirorunental Protection Agency Office of Air wuality Planning and Standards Kes~arch Triangle Park North Carolina EMB Kemicroort 8-CKU-12
187
Braman RS 1976 Application of the arsine evolution methods to environmental analyses presented at the International Conference on Environmental Arsenic Ft Lauderdale Florida 5-8 October
Bratina J E 1979 11 Fabric filter applications on coke oven pushing operations J Air Poll Cont Assoc 29(9) lll6-920
tiuonicore AJ 1980 Environmental assessment of coke quench towers in proceedings First Symposium on Iron and Steel Pollution Abltlternent Technology Chicago~ Uirnois JO October - f N~vernber 1979 EPA-6009-~1b-62~ pp 112-142
t3usse A D and dR Zimmerrao1 1973 User 1 s guide for tdegh~ Climatological Uisper~ion Model US Environmental Protection Agency Resea~ch Triangle Park Nortl Carolina) EPA-R4--73~0024
California Air Resources 8oard 1976 Coke oven emissions miscellaneous emissions and their control at Kaiser Steel Corporations Fontana Steel making facihty Enforcement l3ranch) S2crcmento California L amp E-76-11
Calvert~ CA 1976 Envir __ Sci_ i( __ Technol 10256
Coco L~i 2t aL~ 1979~ 11 02c10pmert of trec1tment and control technology for 11refractory pet rochemi co was tt=1s r middotl 1d l Report Gu 1f South Research In st it ute
to U S tnv i ronmenta l Protection Agency EPA-ti002- 79-080
Colucci~ LM and CR Begemi 1971 Palynuclear aromatic hydrocarbons and other pollutants in Los Angeles air In Proceedings of the International Clean Jir Congress Vol 2 Academic Press pp 28-35
Cooper JA and JG Watson Jr 1980 Receptor oriented methods of air partkulate source apportionment 11 J Air Pollution Control Assoc 30(10) 1116-1125
Coutant R W JS McNulty and RD Grammar 1975 Final report on determi 11at ion of trace elements in a combustion sys tern Prepared by Batte11 e Columbus Laboratories for the Electric Power Research Institute EPRI 122-1
11 11Dilling W 1977 lnterphase transfer processes Envir Sci TechnoL 11 4 pp 405-409
Oow1irgl) MP JO Jeffry and AH Laube 1978 11 Reduction of quench tower emissions Presented at the 71st Air Pollution Conotrol Association Conference Houston Texas 25-30 June APCA No 78-92
EPA 1980a Review of Criteria for Vapor Phase Hydrocarbons 11 US Environmental Protection Agency ~eport 6008-80-045
EPA 1980b Organic Chemical Manufacturing Volume J Storage Fugitive and Secondary Source EPA report 450J-80-025
EPA 1981 Compilation of Air Pol lutdnt Emission Fdctors (Including Suppleshyments 1-7) Third Edition Suppl1111ent 12 Utfice of ir l)uality Planning dnd Stdndlt1rds AP-42-ED-3-Supp I - lt
188
Erikson Uli 1~80 11 Control Uevice Evaluation Condensdtion 11 IT Envioscience lnc report in prepdration for US Environmental Protection Agency ESEU
lrtel liL 197Y Quench tower particuldte emissions J Air Poll Cont Assoc 2Y(9) 913-916
Ev a n s K bull P bull 1981 Pe r son a l Commun i cat i on v i th R bull Zi s k i n d bull
Goodwin DR 1980 11 Air pollution emissions standards in Proceedings First Symposium on Iron and Steel Pollution Abatement Technology Chicago 11 lrno1s 30 October - 1 November 1979 EPA-6009-80-012 pp 37-45
Gordon GE 10 Receptor Models Environ Sci Tech 14(7)792-800
Gordon KJ 1976 11 lJistribution of airborne polycyclic aromatic hydrocarbons throuyhout Los Angeles Envir Sci Technol 370-373
Great Lakes Cdrbon Corporation 1977 Study of coke side coke-oven emissions~ Vol 1 Source testing of a stationary coke side enclosure EPA-340l-77-014A
lirimsrud EP and HA Rasmussen 1975 Atmos Envir Y 1010
Hartnan MW 1~80 Source test c1t US Steel Clairton Coke Ovens Clairton P~nnsylvania Prepared by rRW Environmental Engrneerrny D1v1s1on for US lnv1ronrnental Protection Agency Emissions Standards and Engineering l)ivision Research Trianyle Pdrk North Carolina EM~ Report 78-CKU-13
Hendriks Ilt V et a I 1Y7Y Urgani c air emissions from coke quench towers 11
Presented at 2nd Arrnual M(etiny ot the ir IJol lution Control Associdtiont Cir1cinndti Utlio June 1~7lJ l-dmicroer NU 9-391
Higt1 MU ME Lukey dnd TA Li Puma llJ77 Inspection manual for secondary lead smelters premicrolt1red b Enyineeriny Science lnc for US lnvironmentdl Protection Ayency Office 0f ~nforc~nent EPA-3401-77-001
Holzer G et al 197 11 Collection amp analyses of trace organic emissions from natura I sources 11 Journal of Cnromatogra flhY 142 pp 755-764
Science Applications lnc Final lemicroort to US Environmental Protection Agency tPA-560lJ-80-UU2
J bull Mi l n e l lJ 81 Person a I Cornm un i cat i on wi t t1 bull Zi ski nd bull
Kemner WF 1979 Cost tffectiven2ss model for pollution control at coking facilities Premicroarect oy PEDCo Environmental lnc for US Environmental Protection Agency Industrial Environmental Research Laboratory Research Tridnyle Park Nortll Carolind EPA-GU02-7l-1U5 (NTS PB 8U-118706)
Kessler T AG Sharkey Jr and kA Friedel 1973 Analysis of trace ele111ents in coal by smicroark-source mass spectrometry US 8ureau of Mines Report ot I n v es t 1 gd t 1on s No bull 77 14 bull
189
Kolak NP J Hyde and R Forrest~r 19 1 9 Particulate source contributions in tt1e Niltlgara Frontierffi Prepared t)_y Nr2bulll York State IJepartrnent of Env i ronm~ntc i Con serv d ti on for US En vi r inment ct l Protect ion Agency Region 11 EPA-9024-79-006
Mackay U and PJ Leinonen llt175 Hate of Evamicroordcion of low - solubility contaminants from vJater bodies to c1tmospmre 11 Envir Sci Tecnnol 9 13 pp 1178-libU
Magee tiv HJ Hctll and GM Vc~rlt_a Jr 1973 Potential Doimiddotiutants in fossi1 f1Jel Prepared by ESSO fbullt~scarcr middotnd Engineeriny Co tor US Enviromihm~ct Protection Agency Ufl-R2-7 --2l1~L (NTIS Pt3 as u~9)
Marcriant middotG0H ard RL Meek 1~gu Eva uation of technology for control of arsenic tmissfons at the CdlmpJel Hd Lakmiddot G~ld Smelter prepared for the US
1Env i ronmentd l Protection Agenc)-l-l(lU-sfrT l lnv i ronmenta l kesedrCiI Laboratory Cincirrn2ti Uh70 lYA-60U2-8U--141 (NTIS 8 oU-21Y36J)
Margler L$ HA Ziskind M8 lcgozen (Ind M Axelrod 1979 11 An inventory of care i nog)n i c substances re i ea sect into ne ambient air of California Vol une I - Scn~enin~ and id2nt~fication o-f r~1rci1ogens of greatest concern Science J- pp 1i ca middott~ L J s In c bull Fi nd l kcmicro~ rt ~ i I - U 6 C) - U- 5 0 4 to U1 e Ca I i ~- o r n i a Ai r kesources 130ard
Milne J 198L Persond1 C~mmunicdtmiddotion middot1itt1 R Ziskind
llurchio JC WrC Cooper and A ue Leo11 1970 Asbestos fibers in ambient air of California 11 Cal Horn i ct Air lesourc2s l3oa rd ARl3 4-0S4- l EHS Report 73-2 (March)
National Academy of Sciences 1Y72 Part1culate polycyclic organic matter Washington~ DC pmicro 4-ll
Nati ona I Kesedrcn Counc i I 197b Arsenic prepi1 red by Subcommittee on Arsenic Co111mittee on 1bull1edical ctnd t)ioloyicai Effects of Environmental Pollutants washin~ton uc
Oliver JF and JT Lane 1979 11 Control of visible emissions at CFampl 1 s coke plant-Pueblo Colorado J Air PolL Cont Assoc 29(9) 920-925
Pe I I i z z a r i E bull 0 bull anct J bull E bull l$ u n c h 1s 19 1 nbi en t ai r ca r c i n o gen i c v a po r s Improved samplin~ and analysis tectrnology EPA Contract 68-02-2764
Pellizzari E U 1977 11 Tle 11easure11ent of carcinogenic vamicroors in ambient middotatmospheres EPA-b007-77-J55
Phelps RG 1Y79 lSI-CPA-oatLlle c-ilte oven door sealing proyram J Air Pol 1 Cont Assoc 29(J) JlJ8--JlL
Porter RA 1976 Uismicroersion ~yuation s lucions Dy calculator I guide for air pollution engineers ctnd sci12nthts S cond edition rexas Air Control tioard ~MT1S P~ 262 lJ~)
Redburn T 1980 Kaiser battles aginst l~gacy of woes 11 Los Angeles Timesmiddot (7 Semicroternber irno) Part VI pp 111
Research Triangle lnstittJte 1977 11 Qudntification of benzene in 15D -1bullnoient air samples 11 for US Environmental 11rotection Agency Office of Air Quality Planning a11d Stitndards
Rinkus SJ and MS Legator 1979 Chemical characterization of 465 known or suspected carcinogens and their correlation with mutagnic activity in the Salmonella typhimurium system Cancer Research 393289-3318
Koberts R M 1980 11 An inventory of arc i nogen i c substances r~ leased nto the ambient air of California Tasks 11 and IV KVB Final Report KVB 26900-836 to Science Applications Inc
Rogozen MB LW Margler P Mankiemiddott1icz cind M Axelrod 1978 Water pollution control for coal slurry pipeiines Prepared by Science Appl1cat10ns Inc for US Department of Energy (NTIS SAI-068-79-516
Rogozen MB DF Hausknecht and RA Ziskind 1976 Methodology for ranking elements in fossil fuels according to their potential health impact Science Applications Inc Final Report 260-77-539 to the Electric Power ~esearch Institute
Russel 1 JW and LA Shadoff The sdmpl ing and determination of halocarbons in ambient air usin~ ~rncentration on porous polymers
Siebert PC CA Craig and Eb Coffey 1978 Prelimary assessment of the sources control and population to airoorne polycyclic organic matter as indicated by benzo(d)pyrene Prepare( by lnergy and Environmental Analyses Inc for US Environmental Protection Agency Office of Air Quality Planning and Standards Resedrch Triangle Park North Carolina
Simmonds PG et amiddot1 1974 Distribution of atmospheric halocarbons in the air over Los Angeles basin Atmos Envir Vol 8 pp 209-216
Singh HB LJ Salas and RE Stiles Distribution of Selected Gaseous Oryanic Mutagens and Suspected Carcinoyens in Ambient Air in Proceedings Annual Meeting Air Pollution Controrl Association 82-651 June 20-25 1982 New Url eans
Smith WM 1979 Evaluation of cok oven emissions in Yearbook of the American Iron and Steel Institute pp 163-179
SRI International 1980 Asessment oi human exposures to atmospheric benzene SJ Mara and SS Lee finai report to US Environmental Protection Agency EPA-4503-78-031
Taback HJ 1978 Control uf Hydrocarbon emissions from stationary sources in the California South Codst Air Basir Prepared by KVB Inc for the Cal1fornid Air Resources Bod-rdmiddotKVB No 5804-714
191
Trenholm J-f dnd LL )eek 1Y78 11 Assessment of t1azardous organic emissions fror1 slot type coke oven Datteries unpublished paper US Environmental l)rotection Ayency Ernssion Standdrds and Engineeriny Jivision
Turner U~ J970 Workbook of atmusphe~ic dispersion estiimiddot1Jt2s IJS Environmental Protection Agency k~sedrch Tri1ngle Park North Carolina
US Bureau of Mines lYl f-inercls yearbook 197U Vol 11 p 4~3
Van Os dE I l U bull iJ bull lL Ma rs l and et ct 1Y7~1 o En i r onmt~ nta I a s s es smen t of co k e Dy-product recovery plants prepareu by ReseMcn Trictngle lnstitute for US Environmental Protection AgerKy 11cJustrial Environ11ental R~search Laboratory Kesearcll Tri2n~e Idrk NorU1 Caroline EPA-6U02-7J-Ul6 (NTfS PB 293-278)
Westbrool( C W 1979a Level l ass~ssment of uncontrolled sinter plant emissions Prepared by fesearch Triangle Institute for US Environmenta1 Protection Agcncy Industria1 tnvirnnmental fltesearc-1 Laboratory kesearch Trianyle Park North Carolind EPA-bOU2-7Y-112
Westbroollt C~~ 197ltJb Level l css~ssment of uncontro1led l)-BOP emissions Prepared by Researcn Triangle institute for US Environmental Protection Agency Industrial Environmental Hesearch Laboratory Research Triangle Park ~Orth Carolina EPA-6002-79-190
Wetherold rLb LP Provost and CD S111ith 198U Assessment of atmospneric emissions from microetroleu111 refinin~ Volume 3 Appendix l3 Radian Corp Fina Report to US tnv i ron111enta I Protection Agency EPA-600 2-80-07 5c
Ziskind RA OF S1ilith JL Hdhn etnd G Smicroivey (1980) Determinants of Cancer and Carctiovasculdr Uisease Mortality in Asbestos Mining counties of California SAi Report 068-dl-514 l May
192
APPENDIX A
Rapid Screening and Identification of Airborne Carcinogens of Greatest Concern in California
Lawrence W Margler Michael B Rogozen Richard A Ziskind and Robert Reynolds Science Applications Inc Los Angeles California
This paper describes a method for establishing a priority list of airborne carcinogens within a
state jurisdiction In this case it was necessary to identify from among hundreds of potential
candidates the live to ten materials of greatest potential concern In California as airborne
carcinogens
Because no previous Inventory of carcinogens In California existed published lists rankings
and assessments of national scope were used to Identify candidates By systematic manipulation
and comparison of these data sources 47 materials of some notoriety were chosen for closer
scrutiny This selection was pared to 22 candidates largely by eliminating those which had
very little production and use In California (Substances primarily used as pesticides were
excluded from the scope of this study) The remc1ining candidates were then ranked by additive
and multiplicative algorithms and by a panel of experts The results of these rankings were
combined to produce a single selection of 11 priority candidates In alphabetical order they
ethylene dichloride N-nllrosoamlncs perchloroethylene and polycyclk aromatic hydroiarbons
In continuing studies a baseline emissions inventory Is being prepared and a source testing
program is being designed
In recent years concern has grown over su~pected arcinug-ens New Jersey seshythe possibility that certain materials lected ten volatile organic compounds released to the atmosphere through inshy and five htbullavy metals to be examined dustrial and commercial activity may be further and is currently measuring responsible for a significant portion of ambient atmospheric concentrations of the incidence of cancer in the general these substances in a variety of areas In population This concern is manifested the second year of the study the state at the federal level in the National h1s incre1sed the volatile organics Emissiltm Standards for Hazardous Air stt1died to 20 and begun measuring Pollutants (NESHAP) which limit heavier orianics associated with parshyemissions of the known carcinogens asshy ticulatcs2 In California a very different bestos beryllium and vinyl chloride 1 approach-was taken
Only a few states including New Jersey and California have begun efshy Overview of Californias Approach forts to identify airborne carcinogens of concern to the general public for the The California Air Resources Board purpose of setting state emission regushy (( ARB) is sponsoring a three-stage lations for these substances After reshy st 11d~bull of airbornC rmissions of carcinoshyviewing national use data for known and g1middot11s from anthropogenic atlivities The
November 1979 Volume 29 No 11
first stage which is the subject of this paper was to identify roughly five to ten materials which of the hundreds of known or suspected airborne carcinoshygens are most likely to be of greatest concern to Californias general populashytion Also of interest were those which in order to satisfy occupational health and safety regulations might be transshyferred from the workplace air to the outside atmosphere The second stage which is now underway includes pinshypointing of emission sources for each of the carcinogens of importance quantishyfiltation of emissions nnd design of source-testing methods A subsequent stage will consist of source testing and measuring public exposures to those substances for which data are unavailshyable The basis for regulatory action if appropriate will include the results of this program and other related reshysearch
Screening of Candidate Carcinogens
The National Institute for Occupashytional Safety and Health (NIOSH) lists 1905 chemicals which have reported neoplastigenic or carcinogenic effects and 510 which have otherwise received attention for their neoplastigenic poshytential1 The need to select five to ten materials from such a large number of potential candidates dictated that we devise a way to rapidly eliminate from further consideration the vast majority of the substances Given the paucity of published data on most of the candidate substances the screening method was designed to make best use of readilymiddot
C1111yrich1 19i9-Air Pullulion Control A~~ociatlon
1153
A- I
----- ---- ----- ---------------------available information The screening process was as follows (1) eight comshyplications of known and suspected carshycinogens were reviewed and those subshystances which were not used in Califorshynia were highly unstable in air or were very doubtfully carcinogenic were eliminated (2) after more detailed inshyformation was obtained for the reshymaining 25 substances candidates were rated by two different analytical methshyods (3) an expert panel was convened to review dossiers on the candidates and to independeitly nmk them (-i) from the eight to deven substances ranked highest by all three approaches eleven were selected for the emission identifishycation and source-testing design stages of the CAHB effort
lnitiI $cre~ng ot Potential Candidate Carcingtgens
65 compnunds middotwe ire selected from 642 industrial ur~nnic air pollutants comshypiled by MrTRE Corporation for the US Envircmnental Protection Agency (USEPA) in U1at study pollutants were scored by multiplying four explicshyitly defined rating factors annual US production fraction of production lost to the environment volaflity and toxshyicity To adapt this york to our pmpose we first selected the 114 substances listed as being carcinogeni 1~ or neoplasshytigenic Then the scores of each of these compounds under the criteria annual US production fraction of producshytion lost volatility and carcinogeshynicity were multiplied together Seshylected for further consideration were those substances which had a product score above 50 Another 15 substances listed as being carcinogenic but lacking information for one of the other rating factors were also selected This list of 65 was then compar~d with seven other lists of corcinogeris1
-middot 11 Materials comshy
mon to the reduced MITRE list and at least one of the other lists were chosen for further consideration Added as candidates were those suhstances which are regulated as occupational carcino-
gens by the Occupational Safety and Health Administration (OSHA) and certain inorganic carcinogens Finally substances were added which in our judgment should be investigated but had been eliminated at this point Exshyamples of these are bis(chloromethyl)shyether epichlorohydrin and hydrazine
Next the refined list which now contained 47 substances or chemical groups was pared further hy another rapid screening process Eliminated
were all candidates (1) whose production andor use in California was very low (under 10~ lbyr) and wns not thought likely to pose a risk to a localized popushylation (2) which are very unstable in 1ir or CH whilh should not on t lw hnsis of CUtrltbullnL cidlnre liebull considtbullnmiddotd r11rci-
1154
Tahllt I Suh tances n-iewed in letail
Candidate Substances
Arsenic Inorganic lead Ashestos Alkyl lead Benzene Maieic anhydride
nogenic The result of the initial the rating under that criterion and the screening was a list of 22 candidate mashy corresponding criterion weight The terials which is presented in Table I overall rating for pollutant i is then the
sum of the scores under all the crishyteriaRanking Candidates by AddiUve ancll
Mulllpllcatlve AigorUhms The additive approach has several virtues the main one being that it forces
Many screening or ranking systems the user to make all assumptions exshyfaH into one oft wo cat~middotgories additive plicit In the process of setting up such and multiplicative Some systems are a a ranking system new insights into the combination of the two while others problem under consideration may be combine nn ol1jective appronc 11 with g1ined Once the system is set up it is suhjlctive cvnluition of the remiddot ults 1l rdatively easy to use Where data for The L2 substances surviving th1middot initial scoring pollutants are unavailable arshyscreening were ranked independtmiddotntly by tificial scales can be constructed to the tvo approaches If the same subshy quantify subjective factors Finally the stance rated high~middot under both systems sensitivity of the results to the systems its importance to California was judged subjective aspects may be measured For to be more likely than if it had scored example one can determine the effect of highly in only one method changing criteria weights upon the final
Additive Approach In the additive pollutant ranking Similarly an appreshyapproach the user identifies one or more ciation may be gained of the significance criteria and rates each alternatiw subshy of the range of uncertainty for a particshystance against each criterion while sishy ular required data element by varying _ multaneously deciding the relative imshy rating factor values portance of the criteria Eq (1) shows its A fundamental problem with the apshymathematical formulation proach is that there is no logical basis for
adding the individual scores assignedRating for pollutant i = f W1 R11 under the criteria other than the asshy
rbullI sumption that this simulates or even
(1) improves upon the users thought proshyEach criterion or rating factor (1 ) is crss A major operational problem is assined a value for each pollutant and that of weighting the criteria A common each rating factor is weighted (b W1 ) practice is to give all critNia equal -icnrding to its importance reht veto wltgtight hut that is in itstbulllf a st1Ubullm1middotnl lhe otlwr rritnia llw sc11rtbull for 1111 aLnut the 1rcl11tive importance of Ow 1n1 1 undmiddotr nitn1 n j is the prod11middott of n1ttria
Journal of the Air Pollution Control Association
A-2
---------
Multiplicative Approach In the multiplicative approach the rating for each alternative is the product of the rat ihgs under each criterion
Hating for pollutant i --- 11Ill
UJ (~) Jbull I
A multiplicative approach can have ~omc altlvrntnges over additive onrs First in some cases the ratings can be physical parameters such as concenshytrations or volatilities there is then no need to weight the criteria and hence less controversy over subjective judgshyments Second multiplication generally provides a wider range of scores than does addition allowing clearer disshycrimination among alternatives Finally this approach provides results which are more intuitively acceptable As an exshyample of this last point suppose that exposure and harmfulness levels for candidate substance are each converted to values on aO to IO scale and that a certain substance is both extremely toxic and extremely rare An additive approach would give the compound a rating ofO + 10 = 10 which is equivalent to that of a moderately prevalent subshystance (rated say at 5) which is modshyerately harmful (rated also at =-) A multiplicative approach on the other hand would rate the first substance at 0 and the second at 25
Criteria The six criteria used in the additive and multiplicative ranking procedures are defined in Table 11 Asshysignment of values to the Rij was based upon data gathered from published litshyerature personal communications and panel discussion and has been fully documented 13
Because the purpose of this exercise was to determine the relative imporshytance of the suspected candidate carshycinogens R 1 was scaled to the most heavily used candidate substance benshyzene whose annual production and use in California is nearly 109 lb Materials with a use under 105 lbyr would be rated zero for R 1 and rejected Howevshyerbefore rejecting a substance by this criterion we considered whether its emissions could in particular circumshystances result in high exposures to a Joshycalized population
R2 takes into account the fact that the chemical industry is in continual change Substances of concern today may be phased out while the use of others may rise dramatically increasing their imshyporta nee asmiddot poll utan ts Information on developments which could likely result in a change in the growth rate was facshytored into the choice of a value for R2- As an example asbestos consumption in California has been stable in recent years However the pending phaseout of asbestos in motor vehicle friction materials will hasten the decline in asshybestos consumption hence we assigned a value of 1 for R2bull
November 1979 Volume 29 No 11
Ideally one cou1 1 use pouu1c11u
sion as a criterion However in this case emissions were to l e estimated in detail only for the five t1 ten carcinogens fishynally ~elected Thmiddotrtfor(bull a 11w1st1n of tt II iim pol cint ial vus 1~1middotd 1atbulld llpon
knowledge of the s11bstances manufacshyture and USP for R The higlwst rating went to substanc middots which are v idely used especially in consumer products A slightly lower rating went to subshystances which an~ routinely emitted from industrial processes during proshyduction and use Some materials are employed in such a way that emissions are quite low even though tight emission control may not be required by law Materials in this cntegory were assigned a value of two for H3bull Substances which under federal or st ate regulations may not be discharged to the exterior envishyronment but which could be dischJrged by accident received the lowest rating
Each candidate was evaluated (n the basis of its 1ropen-ity to decompbullgtSe in ambient air Materials with half-lives greater than eight hours were considered moderately to highly stable and rated five for R4 bull Low to moderate stability was assigned to substances with halfshylives between zero and eight hours Compounds known to exist in air for only a few minutes would be rated zero
Ta hie II Dcfinitinns of the criteria used
R1 Prbullbullsent use in California
100 of max (109 lbyr) 5 10 1)f max (108 lbyr) 4
1 of max (107 lbvr) 3 01 of max (106 lbyr) 2
001 of max (105 lbyr) l lt001 of max (lt105 lbyr) O
R2 Growth in California use
+ 20 5 +Oolti to +20 4 Positive growth to 0 3 Stahle or unknown 2 DPcline 1 B1middotin~ phased out 0
R Emision potential
Widespread use in ltonsumer products 5 R1middotlatively p1or con rot over emissions 4 Rdatively good control over emissions 2 Tihtly c1int rolled 1
R4 8tabilit~- in ambient Air4 tvlnderale t1 high st 1bility (l 1 middot2 gt 8 hr) 5 Lw to modmiddotrate s1 bility (t1 l -0-8 hr) 3 Unstahle(t2~febull minutes) 0
R Dispcision Potential
Emitted lar~ely as apor or iine 5 particulnte
Emit led largely as rnarse particulate
Rlt Evidence f Carcinu~enici ty
Known or suspected human carcinogen 5 Known mammalian carcinogtmiddotn 4 ~usptctcd 111ammalan carcino~en or 3
known m mmalin muta~tmiddotn Ames ltmiddotst p sitive 2 Prpoundgtcursor 01 co-car inogen
a t11 is the half-liftmiddot
A-3
tion state or anion associations may change in the atmosphere metals do not degrade and were considered stable Asbestos is likewise itahlc Many of the clcc-ompo~ilion rtbullnctions oi orinnic molecules are mediated by light Such substances if released at night would have several hours to disperse in the surrounding area
A rapid way of assessing the relative potential of different substances to spread from a release point is to note their physical state under normal amshybient conditions Accordingly we scored materials emitted as vapors or fine particulates the highest for R5 and coarse particulates the lowest Intershymediate values are possible for varying amounts of fine and coarse particulate emissions from the same source or from different sources
There is as yet no widely agreed upon measure of the relative potenciemiddots of carcinogens although some ranking systems have been proposed14 Extrapshyolating data from in vitro techniques such as the Ames bacterial mutagenicity test and from laboratory animal studies to humans is problematic Therefore a less quantitative measure of the carcishynogenic potential of each candidate substance was used The candidates receiving the highest scores for Rs would be those for which there is strong evishydence of carcinogenesis in humans Exshyamples are asbestos which is implicated in mesothelioma vinyl chloride which has been identified as the agent of liver cancer in exposed workers and bisshy(chloroinethyl)ether shown by epideshymiological studies to cause lung cancer in resin workers The next highest rated substances are those for which human carcinogenicity is unknown but which have produced cancer in one or more mammalian species in laboratory tests Next are those which have not been shown to be carcinogens but which have proven to be mutagenic in test animals Substances for which the only knowlshyedge of carcinogenic potential is a posishytive Ames test (producing mutations in histidine-requiring strains of Salmoshynella) are rated 2 Finally substances which are implicated only as precursors or co-carcinogens would be rated lowest
Substances unequivocally associated with carcinogenesis were considered as carcinogens in this study Conditions of emission and exposure including the presence of co-carcinogens were facshytored into the evaluation of each candishydate where possible Carcinogenic subshystances derived from the metabolism of a precursor were considered as carcinoshygens However ubiquitous substances which have been hypothesized to be precursors (eg secondary amines nishytrous acid and nitric oxide which combine under certain circumstances to
1155
form N-nitrosoamines) were not conshysidered because of uncertainties in the importance of their link to the carcinoshygenic compound and the practical conshysiderations demanded by the scope of the study
It was beyond the scope of this study to jud~e the validity and interpietation of the experimental ad epidemiological evidence upon which the carcinogenicity of candidate substances has been esshytablished We accepted the conclusions about carcinogenicity drawn by the Inshyternational Agency for Research on Cancer or the National Cancer Institute and did not consider dosage or route of administratioll of tested substances However considerations of test validity did enter into the subjective evaluation by the panel of experts
Weight lr1 the sdditive ranking scheme each rating criterion is weighted according to its limportance relative to the other criteria Little precedent exists for assigning these weights In our judgment and as generally agreed by the panel of experts (see below) R 1 RJ and R6 are more important than the other criteria Evidence of carcinogeshynicity was considered to be the most important criterion of aB so W 15 was assigned a valtze of 3 W 1 aid W 3 were set at 2 and W W-i and W5 were set at 1 In order to discern the potential senshysitivity of the rankings to the weight assignments the candidates were also ranked using equaily weighted crishyteria
Ranking Candidates by Panel o~ Experts
Anine-member panel of experienced governmental industrial and academic scientists whose disciplines include~ -organic and physical chemistry indusshytrial hygiene toxicology epidemiology and iegulatory control of toxic sub~ stances was convened to provide addishytional data for our ranking algorithms
to discuss our candidate --ubstanc -sand rejections to suggest pos-ibie ne suushystances for consiceration and Lt rank the candidates indepen11mtly ( f our own ranking Tvo v-ee ilts he fore the meeting pa-d uember-s were jven one-to three-page dossiers on eac I canshydidaie substance
At th~ start of the meeting- before any group discussion the pand was asked to rate each candidate suhtance with a score from Oto 5 Next cmiddotach candidate was discussed at length e provided ~n ovrview and summarized critical issues identifi2d up to that Pbullgtnnt Through materials brought to thtmiddot meetin and their persoril experience panel memshyber Veimiddote able to provide much useful information on the candidates and adshyditional insight in to our ~a ting criteria
At the encl of the two-d1y sc~sion the pa-iel ugain -ated the cai 1didates
Flnal SelediDn
Tabe III show- the highest-scoring substance a~ det(rraine1 by the 1ddishytive and mlltiplictive approaches and by th~ panel in ~e additive approach 21 single r1nking as obuined by avershyaging the two ran kings resulting from using equal and uneguai weights The ran kings ltif most candidi tes were unafshyfected hut carbo11 tetralt hloride chloshyroform chromium and inorganic lead changed more tban thmiddotee positions_ Because uncertainties in the data base predude imputing signifiance to small differences in the Lnal ordering the lists in Table III are prtmiddotsented in alphabetishycal order However it is of interest to point out that benzene consistently ranked hihest
Becausimiddot some cindidates had equal rating scores we ltould nut choose exshyactly ten candidat ~s from the additive and multiplicative rankings Instead the
top nine and eleven were selected from the two exercises respectively We also considered the ten substances scored highest hy the panel at the end of the session The final consensus selection cornsistt-d of the 11 candidate substances appearing on at least two of the three lisL- For the substances included in the cun~ensus ranking a baseline emissions inventory is being conducted and a source testing program is being deshysigned
1ejected Substances
Some comments about certain subshySiF1ces not appearing on the final list are in order inasmuch as they include known carcinogens and compounds which hae received considerable atshyllttion in recent years as occupational c11c111ogens
rouisionally Rejected Substances Appended to the consensus ranking (T1ble III) were vinyl chloride gasoline and engine exhausts tobacco smoke and pesticides No further action by the CRB is recommended at this time for vinyl chloride because it is already subject tu the USEPA emissions stanshydard a CARB ambient air quality strndard and an OSHA standard
Gasoline and tobacco smoke were appcndlmiddotd to this list because each ocshycurs very widely and contains several of the candidate substances reviewed in this study some of which are in the final listing For example gasoline contains benzene ethylene dibromide ethylene dilmiddothloride and alkyl lead compounds the last three being in leaded grades only Both gasoline and diesel combusshytion products include PAHs Tobacco smoke contains among other neoplasshytigenic substances nitrosamines PAHs nickel arsenic cadmium and other heavv metals Manv individuals are inshyvolu~tarily and i~ many situations
Table HI Hi~hest ranked candidates from each rnnkin~ methodbull
Highly ranked but no inventory
l I ighesl consensus recommended Additive Multiplicativ(~ Pan I of eqwrts rnnkmg at this time
1156 Journal of the Air Pollution Control Association
A-4
lirtualy unavoidably exposed to t11-k1ltco smoke Because the sources of gasoline its combustion products and tobacco smoke emissions are well known no specific action was recomshymended for these materials during the tbullmissions innnlory and -ource testing design stages of the present study It was considered importint ho-vemiddoter to draw attrntion to the general publics exposhysure tot hrse substances PrsticidltS are listed for the same reason thou~h ad(bullmiddot tailed examination of pesticides w1s beyond the scope of this study Many pesticides are widely used and some of them are known to be carcinogenic
Other Rejected Substances Acryshylonitrile and vinylidene chloride were placed in a provisionally rejected group because of the panels suspicion that imports of these compounds to Cnlifornia from Japan may be appreshyciable yet are hard to substantiate Should such imports be verified in the future these two compounds would take on greater importance Formaldehyde was also provisionally rejected because the preponderance of evidence indicates that it is not carcinogenic and that bisshy(chloromethyl)ether is not formed from formaldehyde in appreciable quintities in industrial environments 15 The ocshycupationally controlled carcinogens ethyleneimine and beta-propiolactone were rejected in part because of their reactivity in air At the time of this study DBCP a pesticide was no longer heing middotproduced in California and was therefore rejected from further considshyeration
Occupational Regulations and Community Exposure
A question of interest to the CARB was whether the regulation of acknowlshyedged occupational carcinogens adshyverselv affects the ambient air outside the v~rkplace Our general findings can be illustrated by the example of asshybestos
Asbestos is a very widely used mateshyrial for which no ambient air standard exists Concentrations in the workplace are limited to an eight-hour timtmiddotshyweighted average concentration of two fiberscm=1 of air and a ceiling concenshytration of ten fihcrscm 1bull In meeting this standard exhausting air containin~ asshybestos to the ambient air is not reshystricted except by the USEPAs reshyquirement that there shall he no visible emissions containing asbestos piut ides from s11eh foci lit i1bulls (bullxd11din~ hrak(bull shops 1 Considning that undc-r certain conditions 101 asbestos fiherscm1 could ht Pmitted without being visihl( 11 this standard mamiddot allow considerable asshybestos emissicns It is unlikely however that emissions would actually approach these levels First since the OSHA standard cannot generally be achieved
November 1979 Volume 29 No 11
by ventilation n ine the generation or asbestos partid in the workplace must be greatly reltlu ed Second the air is usually filtered 11gt prevent recirculation of asbestos to 1 workplace Asbestos waste must bed 1sposed of in sealed imshypermeable bag~ or containers Thus under current onupational regulations the amhient air 1enerally appCars to be afford(bulld greatn protection than it would without s11ch regulations 1
Conclusion
The screenin and ranilting methodshyology presented in this paper proved to be a feasible approach to establishing a priority list of Jirborne carcinc)gens in California We ftel that it is an efficient means of focu~ing further efforts on emissi()ns inventories source testing and ambient measurement for it not only identifie all the carcinogens of potential concern but it also permits the state regulatory agency to direct its reshysearch resourCtmiddots toward those subshystances of partilmiddotular interest within its jurisdiction
References l National Elllission Standards for Hazshy
ardous Air P llutants A Compilation as of April 1 19 3 PEDCO Environmenshytal Inc for US Environmental Proshytection Ag1middotncy EPA3401-78008 (NTIS No I B 288 205) 19i6
S Gray Ne Jersey Department of Enshyvironmental l 1rotection Trenton NJ private com111unication 1979
1 H E Chrislllsen E J Fairchild and R J Lewis Sr middotmiddotSuspected Carcinogens A Suhfile of th1middot NIOSH Regii-try of Toxic Effects of Chemical Substances 2nd Ed National Institute for Occupational Safety and HPalth NIOSH i7-149
4 B 8 Fuller J Hushon 1 Kornreich R Ouellette L Thomas and P Valker Scoring of rganic air pollut1nts for US Environrnental Protection Agency Mitre Corpt ration T1middotchnical Report MTR-6248 I 9i6
S F R Lozan11 Toxicants Selected for Priority En- 1ronmental Assessment corresponde11ce with EPA Region VI Administrat r 19i7
1gt Survey of inyl Chloride Levels in the Vicinity of Keysor-Century Saugus California US Environmental Proshytection AJet middoty National Enforcement lnvestiiatim Center Denn-rand San Francisco E A-n02-77-017a 19i8
0 Bardin l middotstimon~middot in Hearings on Helationshi1 Bet wfen Cancer and the Environmei 1 US llltuse of Hepreshysentatives bull th Congress 2nd Session Committee (In Interstate and Foreicn Commerce ~crial No 94-141 1976 pp 7~i0
middot T H Ma11middoth Carcinogens in the workplace lwrE to start d1middotaning up Sc-imiddotrmmiddot l17 1-1110) lfi8 (197)
I CH ar1111 Classif1r11t1on of the Exshyposure Haza d Prtbullstgtnled Ii~middot Thirt~middot-Six Clwmirals 011 I llixllmmiddots lternorandum from Clllm1middot I Offi(middote of thtgt Assmmiddotiate lgt1rector for 1 middotarcinoltnesis Division of Cancer Cau~ l and Prevention lo SushyJwrmiddoti-or~middot ( 0 Auditor Gtmiddotneral Acshycounting Oft lmiddote19ii1
I0 Chemicals ith Sufficient Evidence of ( arcinogenit ty in Experimental Ani-
A-5
ma1s Jnlernauona1 Agency 1or 1c-curcn on Cancer IAHC Working Group Reshyport Lyon France 1978
11 Threshold Limit Values for Chemical Substances and Physical Agents in the Workroom Environment with Intended Changes for 1977 American Conference of Governmental Industrial Hygienists 1977
12 M H Rogozen D 1 Hausknecht and R A Zisk ind 1976 A Methodology for Ranking Trace Elements in Fossil Fuels According to Their Potential Health Impact by Science Applications Inc for Electric Power Research Institute 1976
1t L Margler R Ziskind M Rogozen and M Axelrod An Inventory of Carcinoshygenic Substances Released into the Amshybient Air of California Vol I Final Reshyport Screening and Identification of Carcinogens of Greatest Concern by Science Applications Inc for California Air Resources Board 1979
14 T H Maugh Estimating potency of carcinogens is an inexact science Science 202 (4363) 38(1978)
15 C C Yao and G C Miller Research Study on ais(Chloromethyl)Ether Forshymation and Detection in Selected Work Environments by the Bendix Corposhyration for National Institute for Occushypational Safety and Health Cincinnati Ohio Contract No 210-75-0056 1979
16 Occupational Cancer Control Unit Calshyifornia Department of Industrial Relashytions Los Angeles California private communications with staff members 19i9
This paper was developed while Dr Margler was a Staff SriPntist in the Energy-Environment Systems Division of Science Applications Inc (SAIi 1801 Avenue of the Stars Suite 1205 Los Angeles CA 90067 and Mr Reynolds was a Project Manager with the California Air Reshysources Board Sacramento CA 95812 Dr ltar~ler is currently emshyployed as DirEgtrtnr of Bnvironmental Scitbulllltcs for WinzlN nnd Kelly Conshysultini EnintNi t1 Third Stretbullt EurekH CA 9gtgt01 and Mr Heynolds is Air Pollution Control Diretbulltor for the Lake County Air Pollution Conshytrol District 55 N Forhes Street Lakeport CA 95-13 Dr Rogozpn is a Staff Scientist and Dr Ziskind is Deputy Manager of the Energy-Enshyvironment Systems Division of SAi
1157
Appendix f~
_l)ispobullition of Miscellaneous Sites
Gould Inc Ver_n)_~ St~condary Lead S~_n_elter
The emissions of arsenic from the four large secondary lead smelt~rs
in Cali forni) J1~1 ~ bull~tiilnted in the program This estimate was based upon a
uniform fugitive emission f1cJr nt WPll supported by measurement inforriation
Therefore source monitoring was r2cq11)nded 1nd Gould Inc Vernon being
~he largest was singled out It was however proposed to monitor both Gouli
and RSR Corp (Quemetco) in the City of Industry since analysis of the plants
revealed significant differences in plant equipment and engineering The
latter being more typical of 1 nodern facility
As a r2sult Jf pr-~-middott -iiscussions and plant inspections we became
aware that Gould was actively in the process 1)f constructing a new facility
which would completely replace the existing one We have monitored progress
on the new site and concluded it would he inappropriate to utilize program
resources to conduct field tests at Gould ~rr1issi 1rns from the new Gould
faci 1ity wi 11 be bas~rl 11p11) test results from RSR
PG ampE - Pittsburg PG ampE - Salinas a~d So Cal Edison - Long Beach Oil Fuel Power Plants
It was app ro p ri ate tJ cons i dP r the (~mission of arsenic from power
plants during the initial study stage inc~ imiddotrace quantities in the fuel oil
are known to be ernitted Because of the population distribution in tile
vicinity of three plants and some 11realistically conservative estimat2s of
erlissiJn c-1ctors tht~ facilities appearc~d among the top seventeen stationary
sources of potential c1Jn-r-n HovJ~ver as a result of a reexamination of
emission data it was concluded that an error ~dj ~een made in the material
balance of arsenic - resulting in tn emission factor equivalent of a 300
release ~io lit~rature was found to justify 1reater than a 30 transfer
function Thus we estimated at the outside th~ emission factor should be
reduced from 013 lb per 1000 lb to approximately 0013 lb and resulting
ars9nic 2missions for the entire state w~r~ cJ11servatively estimated to be
1760 lbs (from 17600) divided amongst 311 the states power plants Clearly
then the four secondary lPad smelters estimat~ of S9410 lbs of arsenic(
B-1
emissions per year makes consideration of poer pl int ernissio1s if ffsenic a
very low priority
In a literdture review by SAI - Methodokgy __f_or Ranking Trace
Elements in Fossil Fuels Accordi~9__t_~_their Potential Health Iripact (Roqozen
1976) - emissions estimates f0r 15 trac~ substances were researched for c11l
and fuel oil pomiddot1er plant c 1J1v---r-sion Source content combustion pr-)~~
transfer functions and contrcl methodology were co1~idered to develop e~ission
factors The output to input ~atio computed and utilized n -~i-i~ study for
arsenic was 002 to 03 (ie tra1sfer function betw~en 2 11d JO~) reflecti~q
the wide variety of fuels processses and controls nationwiJe The upper end
reflecti1g both high arsenic coals and poor e11iss10n contro 1 devices Clearly
neither of those conditions accurtely apply to th2 thr~~ si~~~ r1nrl thus
substantiate the rlecision not to perform emissions neasur-i-~n~s
Calaveras Asbestos Company - Copperopo l middotis Asbestos Mini n(i and Mi 11 i nq
In the past (late sixties and early seventies) ~h2 ~3laveras
Asbestos Company came under heavy criti-is1 1fter imiddot1sp~ction rieasurements
revealed serious problems However significant reduction of e1nissi0ns have
occured prompted by NESHAP regulation and occupcttional standards SAI
inspected the site in f)~c~1b-2r 1979 under an EPA contract An missions
inventory was published under that work (Ztskind 1980) and the bottom line
conclusion is that currently no significant ernissioi1 are gteing released as a
r~sult of hl-isting which would reach the publ le i 1le 0pen pit is some 900
feet deep with blasting at the bottom 0ver 80 percent of emissions in the
CARB Emission Inventory System were attributej to pit blasting In fact at
its current depth blasted 1aterial does not reach the mine surface with tile
explosive implacement used Additionally the site is more remote than might
be realized The ~ituation is vastly different today than in the past and
currently attention should be paid to the issues of occupational exposure and
breakdown of ventilation syste111 controls in t1e inilling operationsmiddot He
recorrrnended that no furtw~ corsideration be Jiven to this site for the
purposes of tl1is w)JJil ie identification )f silt~middot1ificant releases vhich
might he responsible for causi1J middot1 fmiddot middotgtbullgt middot 1= 1(11 1bull11 tinn Pxposure
8-2
Vctri0J5 Refineries
It was est=tblished early in the anrilysis prmiddotJr111 V1at gtenzene
emissions from refinery operations coulrl in th~ 1guregate constitute a
significrnt source Since there ilre a large rrnrnber 0f refineri~s in
California (46 in Los Angeles County itself -it t(~ i 1~ middot)f r-1ltariination) with a
wide variety of types sizes ages etc their eva l uat 11) 1 i =r~sent -3~ i)7
monumental task It was noted immediately t 13t ~~it hi n the total scope of the
study the design and conduct )f c r~f i r1~ y testing program which would develop
a complete benzene emissions in1~ntory for 01e site was impractical let alone
to d1aracteri ze emissions fro111 three ie those listed among the 17 rnost
s i g n if i ca n t potent i a l st a t i on a r y sou r c e em it 1-_ l rs bull
The three refineries were singled out for special attention in the
previol1s phase ~ecause they uriiquely had components which process materials
containing 10 or more per~nt hy middot1-bulli1 11t )-~112~ne (46 FR 1165 Jan 5 1981 pg
1491 ) Est i mates by E P A ( F e d bull r l Rr-~ g i s t ~ r 1q81 ) i n d i cate that 90 p e r cent o r
more of the total benzene fu 1Jitive emissions arise from such components
Therefore attention is appropriately focused on the three henzene production
consumption refineries Chevron (Richmond) Arco (Crson) and Chevron (El
Segundo) SAI staff considered the possibliJ )f -1 testing program at one of
these refineries and cone l url 1middotd it would not 1)e cost-effect i v2 for o number of
reasons
California is a minor producer and consumer of benzene with
approximately 15 of the 114 hill ion pounds produced and
consumed nation1lly in 1977 This can 1)e riY1pared with the fact
that California has approximately 10 of th~ ~ations population
Benzene exposur~ to the generil nJmiddot1lation has been partitioned
among the various sources Aproximately 90 of exposure is
estimated (SRI 1978) to be caused hy gasoline distribution
activities and 1ehiclEmiddot emissions in 1 ir~an areas with nearly all
of the balance )Y benene hrndl ing operations (refineries and
chemical plants) In the case middotlf California this percent1g~ Jill
be even JT1ore di middotpruJor-tionate b~cause of its greater s1are of the
population and ~esser share of the henzene ha~dling Furthermore
since the three refin 1bullry cites ire heavily urbanized no new rural
population cegmmiddotnt is beinq exposed
B-3
Californias most heavily urbanized Air Quality Management
Di st r i cts have a l ready adopted t en zen e f ugit i v e em i s s i on
staridards comparable and vith S( 1ine features VJr stringent than
the proposed national (bullmission lttandarrl (4f-i FR 1165 Jan 5
1981) Thus the concmiddotlusion rleri1ed above (Le refinery emissions
are secondary cause of exposure to the urb~n microopu~ation) is
further reinforced since H is irojected that re1eases from
components in benzene service (gt10 benzene) will be reduced by
73 by the proposed F~deral Standards T~~ ~istrict rules (eg
South Coast Air Quality Management District 456 and 4fi6l) are
not restricted to berz~ne per senor to only components in
benzene service The rjLs 1lso i-lclude flanges in addition to
the components ca i led uut in the propose nat i 01a l standard
The EPA estimated that if the proposed emission standard were adopted the
1~aximum annual benzene concertr0Vioi1 -~01middot 1 plant would be 36 ppb at a
distance of 0 1 kilomeL~~ avay Cor1paring this ground level concentration
with the general urban background i1 California of 19 ppb shows the latt2r t1
dominate In recognition of the secondary im0ortance of these thr~~ i~~s i~
was decided to utilize program resources to d~velop field data at other sites
1111here little emissions informati1)i1 1as cll-1aila1)le
8-4
APPENlJIX C
US tn v 1 ronrnen ta I Protection Agency
middotJc k
METHOD 108 - UETERMINATION OF PA~TICULATE ANlJ GASEOUS ARSENIC EMISSION
FRUM NONFE~RUUS SMELTERS
1 Applicaoil ity and Principle
11 Apmicroli1ability This rnetnod amicropliltS to tt1e determination of
i nor~an i c d rsen i c ( s) e1n is sions trom non f errou Sm(~ I ters dnd other sources as
smicroecified in the reuulations
12 Principle Particulate and gaseous As emissions are withdrawn
isokinetically from the source and collected on a glass mat filter and in
water The collected As is ther1 analyzed by means of atomic aosorption
spectrophotometry
2 Apparatus
21 Sa111pliny Train A schematic of tt1e sampling train is shown in
fiyure 421 it is similar to tt1e Method 5 train of 40 CRF 60 Appendix A
Tne sampliny trdin consists of tt1e followiny components
Note H1i s and a11 subse~uent nferences to ottler methods refer to the Methods in 40 CFflt 60 Appendix I
3 Annual Book of AST11 Stdndarctbull Part 31 Water Atmospheric
Analysis American Society for T~sting and 1-1aterials Philadelphia PA
1974 micromicro 40-4~
C-15
APPENDIX D
~f-~__d_y_ e s f o r Pr o c e s s i ng a n d An a l y s i s of PA H i n
Co~e Oven Emissions from Kaiser Steel
Global Geochemistry Corporation
10 Samples expected
l 1 Connective tubes between section welded to oven port cover and remainder of sampling system - These are to be stainless steel tubing disconnected and capped off for each sample They will be about 10-15 long
1 2 Impingers in series first two water-filled third dry - The center tubes will be removed and wrapped in clean foil and the impingers stoppered as is for shipping One set for each sample Protect from light during shipping
20 Extraction -procedure
Sample fractions are to be covered with foil or in opaque containers with inert gas flush during storage in freezer Solvents used in extraction will be BampJ glass distilled pesticide grade low residue) Working bench area will be screened with amber plastic and light bulbs
2 l Connective tubes will be rinsed with dichloromethane by partial filling capping tilting several times and draining into a beaker This will be repeated until the washings are colorless but at least three times
22 Impinger contents will be transferred to a separatory funnel The impingers and center tubes will be rinsed inlo the funnel using a measured amount of dichloroshymethane in portions The funnel is shaken well and the extract drawn The rinsing and shaking is repeated with two more portions of solvent The extracts and washshyings are combined over Na 2 S0 4 bull
A separate portion of the water used for impingers is extracted in the same way
For each liter of water 100 ml dichloromethane will be used per extraction ie 300 ml total
Internal standard is added to the combined exshytracts at this p0int
D-1
_rmiddot
30
40
Concentration of extracts
Bulk solvent is removed on a rotary evaporator in a water bath at not over 40deg stopp~ng short of dryness The res~dual volume should be just less than the f i n a l v o l ti 11 e o wl1 i ch the s o l uti on i s to be d i l u t e d (accurately) The object is a final solution with roughly five mgml solids concentration in dichloromethane After the vol w1 e ~ s made up to the mar k i n a vol umet r i c flask an aliquot is removed and evaporated on a tared disk in a nitrogen stream and weighed to determine yields The remaining solution is stored in a ial or ampoule sealed with a Teflon-lined septum A split for SAI QA may be made at this point (see Section 43 also)
Gravity column c1eanup chromatogram
4 1 Into a 10 min 1 d column with co2rse sinter and Teflon stopcoc~ 6 g 120 mesh silica gel (activated at 120deg) 1s sluiry packed in pentane cooling the column by evaporation from a tissue wrapped around it and wet with acetone) This minimizes vapor gaps in the column A cap of 3g sodium sulshyfate is added to the top The column must not dry out before the chromatogram is complete
42 A sample of the extract concentrate is taken This is to be based on experience of expected PAH conshytents but initially would be 5-10 mg This is exchanged with pentane by adding 10 ml portions of pentane and evaporating over about 100 mg silica to small volume Three to five portions of pentane are used The final residue is rinsed onto the column with pentane
4 3 The chromatogram is developed with four solvent steps
l 25 ml pentane 2 l 0 ml 2 0 ~~ dichloromethane in pentane
II II ii II3 l 0 ml 50 u II II4 l 0 ml 5 0 ~~ methanol
The column volume is approximately 8 cc This is allowed for while collecting fractions During the first chromatogram the graduated cylinder collector is observed for visible solvent changes and a more precise estimate of the column volume
T h e f o u r f r a c t i o n ~ a r e e v a p o r a t e d i n a n 1 t r o g e n stream at not oveimiddot 40 to l 0 ml then stored under nitrogen in the freezer The principal constituents
0-2
of the fractions will be
l Aliphatics 2 Lower molecular weight PAH
11 113 Higher 11
4 Polar materials
Fractions 2 and 3 are to be analyzed by HPLC the others retained for future interest or in case of an incomplete or defective separation Splits for SAI QA may also be made from these fractions
50 HPLC Analysis
5 l The column is a reverse phase partition column C18 bonded to microparticle silica 4 x 250 mm The solvent system is a gradient from 6040 acetonitrile water to 100 acetonitrile initial slope setting l 5min The flow rate is one mlmin the injecshytion sample volume one to ten microl depending on conshycentration Detectors are UV absorbance and fluorescence in series The absorbance is set routinely at 296 nm but may be varied in certain instances to aid in identification of peaks The fluorescence filtars are narrow pass excitation at 360 nm and cut-off emission above ~10 nm Both d e t e c to rs a re rec o rd e d s i 1i ul ta n e o u s l y
52 Reference calibration is based initially on indishyvidual chromatograms of single compounds and roushytinely interspersed chromatograms of multicomponent reference mixtures The proposed calibration stanshydards are the following at minimum
Phenathrene Pyrene Chrysene Perylene Benzo(a)pyrene Benzo (ghi)pcrylene I n d e n o ( l 2 ~ c d ) p y r e n e Coronene
The internal standards (added in Section 22) are 910-Dimethylanthracene
9-Phenylanthracene 9 10-Diphenylanthracene
60 Peak Identification
Significant peaks not recognizable as present in the reference standard will be investigated by alternate methods for identification
D-3
6 bull I Stopped-flow examination of UV absorbance for maxima
by manual wavelength varmiddotiation allows checking against known reference spectra (published or measured) This is usually sufficient confirm a component already suspected to be present
62 Collection of the fraction contairino the unknown peak allows later determination o~ t~e full absorbance andor the fluorescence spectrum for comparison with reference compounds
63 If not already run on GCMS the sample can be sent to SAI for that purpose possibly allowing the comshyponent to be identified Since the HPLC elution order is not the same as the GC retention sequence it may be necessary in affibiguous cases to collect the HPLC fraction and run it separately by GCMS
The~e may be numerous minor constituents not identified The decision on how many of these to track down and identify will be made in consultation with CARB adv~sors a~d with SAI since at some point the returns ute not worth further expense
_64 IdeAtified peaks are quantified by comparing their areas to the areas of the known concentrations of reference compounds A recovery factor based on the added internal standard nearest in elution order to the sample component is then applied the amount of internal standard originally added to the sample is divided by the amount found in the HPLC analysis This ratio is the multiplier used to correct for losses in concentration and chromatography steps
70 Quality Assurance
71 Transmittal of sample ~plits to SAI provides for external QA This should be done with 15 (or one in seven) of all samples
72 Blank water extracts are carried through the same analytical procedure (10 of total)
73 The cleanup and HPLC analysis are duplicated on 15 (or one in seven) of all extracts
74 The multicomponent reference is run first every day then elution volumes and peak area for each comshyponent are plotted on a control chdrt to pinpoint developing problems At least evetmiddoty tenth chromato-gram will be this standard It wi 11 also be rerun after any service on the instrument~ such as a column change
D-4
APPENDIX E
Gas chromatograph 1mass spcmiddotctromlt~try protocols used by SAI Trace
Organic Chemistry Laborator1
Aliquots of sampl=s received from Global Geochemistry Corporation
were analyzed by combined c11s chromatography mass spectrometry (GCMS)
Liquid extracts were spiked with an internal standard compound (deuteriumshy
labeled phenanthrene) priot to analysis by gas chromatography mass spectroshy
metrymiddot A Finnigan model 4( 21 Quadrupole GCMS system including an Incas data
system was used for all analysis Liquid samples (1 microliter) were injected
into a 30-miter x 025-mm ID SE-54 fused silica open tubular 9as chromatoshy
graphy column (JampW Scientific) A programmed GC oven rate of a 4 minute hold
at 30degc followed by a 40degCmin temerature increase to 160degc and a s0 cmin
temperature increase to 270degc was used for sample analysis The detector
system was a quadrupole mass spectrometer operated in the electron ionization
mode at 70eV The quadrupole mass filter was scanned from 35 to 475 atomic
mass units in 095 seconds A hold time of 005 seconds was used to stabilize
the electronics prior tote next scan Data were continuously acquired using
a Finnigan Incas Data system which writes time intensity mass spectral data
to a magnetic disc The mass spectrometer was operated under computer control
during acquisition Figures 33-7 through 33-9 show the total ionization
chromatograms for each of the three samples that were analyzed
Following analyses the data files were searched for priority
pollutant polynuclear aromatic hydrocarbons (PNAs) In addition to the
quantitative data reduction used for priority pollutant PNAs a survey analysis
was performed on all significant GC peaks (ie signal to noise of greater
than 101) The survey analysis was based upon a computerized library search of
individual background subtracted peak maxima scans for qualifying GC peaks
The library used for qualitative identification AJas a combined NIHEPA
National Bureau of Standards library 1hich contains approximately 26000
entries
Tables 33-11 through 3-13 present the results of this survey analysis As expected a series of al~yl Slbstituted PNAs were observed in addition to heterocyc l i c compounds con ta i ni n~ nitrogen iind sulfur moieties Further
investigation of the mass spectre of the nitrogen substituted identifications
confirmed the presence of benzonitrile isoquinoline andor quinoline acridine
carbazole and alkyl substibted ~yridines Ion specific searches for N-nitroso
substituted amines were neg1tive
E-1
APPENDIX F
Detailed Fiber and Mass Concentration Data - Johns Manville
=-- -i l L - [- - =-- -= f- - =-- -l J middot - -=-
i-- L =-- - ~ ~ L ~ ~ =--shy r---- --
- 7 l -=----7- -- - - - -- =-- - - ~ =-- - 1- -~- l
g~~~~ r_-middot~ ~-i [-- L ~ L Ldeg
middot--~ l - --------=-- r- L r- r-
~~~ =-- -- r- = - -r---
a- - - - - _ - -I~=-- L -- l 7 -~--
gtZlt _
- _(_ -lt lt =- = ZJZ~
-
__ _
~ -- __ middot- -~ ~~~ = L ~f~ Z ~ C
middot -
__ ~ lt lt =- _~~ijE
gt = z lt gt
rbullshy__
F-2
SI I HMllLfc J H 1 I OBr I lll-H ANI) NS~ coHcrrrllliTJ ON CUfllllTI VImiddot SIJMMIIY
TIITIL JnI~ SfiNtlEO TOTtl Ill- or FI ITETI TOTI Illlt ~111-11 11 ltllOT FIIACT I Ori 1111 OF ~middotWI MUllllNJbull lOTI 01IIMImiddot 01- JII TOTI FI lilmiddott~ UgtIJH nD
= t nwo 00 so MI CllONH = t I 1JOO SO (M = II 900 SO CM t bull (0000 l1T~
I I JO SO CM 70000 CUil JC Ml-TEHS
= ~lt O V l lWfl~
Manville D-2 Baghouse
723 347 -633
CIIHYSOTI LE JfllII I nou J11BICUOUS NO PTTFllN NON-SBI-STOS tLL FIIIEIIS
FI 111-11 COllflT ltFIIWll-
ITllClmiddotrir r lllrf
2gt()
) ( I f S bull1 bullbull
I 0
l ~-Irbullmiddot
00
fl JI)(~
00
0 OUU~~
00
0 OflO
~60
1110 000~~
FI IWII ( ()[IClmiddotNTHT I ON ltI- I mn ll11 ~41 GI OF FI LTEH) (FIBlmiddotn~ ll-11 CIJB f-11bullTl-H OF IH)
I (2fi71bull+0G (aG(l-+05
( gt027 F+04 OfmiddotHH
OOOOOE+OO OOOOOEtOO
OOOOOF+00 o uuoul~middotH)O
0 O(OOE+OO o 0000I+oo
J bull f 1)07 E+O( fi r71HE+OG
I
w
TcfLI ~~lllllmiddotCE 1lllmiddot lt~O Ctl lFII ~~O UI OF FIiTEi) t q UI 11-11 Cll II fllmiddotTlbullll OF J I H)
I 590)1+07 6 t 90llbull-HH
( HMOE+O 2 40411bull-1 o
0 OIHHlF+OO o 00001+oo
0 OUO(I 1 00 OOOOOJi+00
() 10110~ Ji)
OOOOOE+OO 1 lHHllJ i 00
0 00001middotgt1 00
M ~~~ middotor-ic Irrrn TI or
( PGRAMS 1bull111 ~c UI OF F J LTEH) (PGRAMS 1111 UJI illI Lil ()j Ill)
IUitTII Ml-Ml ltMI tJIOfl~~) ~Tlgt lgtEV
Ml- H Lot c1middoton nr COlmiddotF I II
I) I MWTEll Ml-N ltMI c11or1~~) STI) tHV
MEtl IOC cImiddotor1 r1N CCWF V11
AIt lbullI rt 1bulli I~ (Sil MIC) STIgt 1gt1-V
Ml- N 101 c1-ori r1u COlbulltmiddot V1ll
I 0 I ~if1+07 bull tJf I (11bull-HH
r 1gta 1 1) bull 2 (10()
0 lt700 l 1Hi4~ 2G44
() ~( 1 027B
- l bull 6 I 1bull1middot 0 J l)J 09J~fgt
1a 1071 1~ 09rn
0 2(jlt)fi l ~HVil 4040fi
~ HltF+O~ U (H~61bull+o I
o rooo 0()()00
-0GJOB 0 (000 0 (000
OOfiOO 00000
- )()57
OOGOO 00000
0 0 1)02 00000
- I J 0 0 0 1W 00000
00000 0 ()000 00000 00000 00000
00000 00000 00000 0 (WHO 00000
0()000 0()()()()
0 ()()()) 00000 00000
00000 00000 0000() 00000 00000
00000 00000 00000 00000 00000
00000 00000 00000 00000 00000
00000 00000 00000 00000 00000
00000 00000 00000 00000 00000
00000 00000 0 (WOO 00000 ()0000
( ~1701 l B bullJl~H 0 lt~( I BIJI ~ ~i Ill
o~-~ 0 ~j
- I ult-fc 0 I Hbull)I o bull)abullgt~
12 (1 41 17bullgt7
0 1 l I H) 4 ltmiddotI 1
VOUJME (ClJB MIC)
MFiN ~Tl) 111-V ru-1N 10C Ct-OM MN COE- Vtll
2lHH7 gt HUB
-2 7m 0 060 1)
( IJl)()C)
0 001 00000
-ltj 74a 1gt 0001~ 00000
00000 0()000 00000 00000 00000
00000 00000 00000 00000 00000
ooono 0 1)000 00000 00000 0 ltWOO
2 7 0 1J lJ I J J
-2J+H () ()~(
B 77gtB
lt ~
lt
C
J c
~ lf
_
ii
(]j (fJ
J I 0
a1 c r--- l)-- Clltr (Y) r- ro C)(V) gt deg CN C1 (0 I ~ C M
N --- r--
w J
s rJ gt-
c
~
LJ
t-
~ - -lt-[-bull
0- ~ l L L l~ -18~~
L a- - ~ I-
-I - ---
-
-- -- -- - - ~l
-1_ - - --
--~--- [-middot-~--- t-- -- [- -_ 3J ~I r-
z g ---
l
t--[-
l l c +-shyc
~ Ci - - I
-1--r-middot~- ~
L l- L ~
L
II V
C
c
t--
-= -- - - ---- - _
middot= - middot-= l J l~ - ~ ~ L - ~ L ~ [--
II II II II II II II
_
---
--
z
- I
==== lt _ --middot --~ 7 _
- ~~- _
z ~ ~ =lt ~~~ti3
f
gt =2 lt_
2 z _ lt=lt i~iS
-- z ~ lt
~~~t3
~lt~ =--~~~sect~ _ -- -middotImiddot_
L -_ -
F-4
(lJ II) I
~ Or--M
ltlJCsr-M ~ O) bullbullbullbull ~ lCMIO bull- ci gt (= N lC I M ~ 0 N
_
J J
---r--co - - ~l r-j = ~i
~~ 2-
88 rJ--_-ltI -
lt ~lt
z
Cl J ~ _ e-l ~~
- ~ middot~ e-- l ~ l~ [- - ~ -~=----~
_ L
L
--o---
-o-~shyc L~ ~ - 0 - Ll-C d o~-
11middot II 11 11 II II II
-_ 0- _
--
-- lt== ~~ ltZ z-== -~~=E~8C~l~-=rI - lt ~ fJ
ltZ lt---=lt===~~~- _ ltltlt=gt ~~=~~~~~ -
~ ~ lt -=z~ z _ -shy middot--z =-
-middotmiddotmiddotmiddot middot-middot -----
r bull bull _
- _ - ~-middot __ _
~ _
middot- - --- middotmiddot
_
- l
~=~r -~= ==
lt =~~~ti
F-5
FIHJ-ll Nil MAS~ CONCFNTllJTlOi SJ I SMllLE I l) Al OJ CUMJIATI JIbull t-llHMHY
TOTL iHf- SCNNFD lOTiL Illmiddot OF FILTFTI TOJI Ill- JSlllmiddot~ll I I OIIOT FlltCT [ ON 1111middot OImiddot ~1-M iIHlllllJNE TO 1 1 I ()lIJMImiddot ()I t I H TOTI V 1111-11~ COUNTED
= 1080000 SQ MI~RlozNS = 11 900 SQ CM = 1 l bull )00 SO CM = 1 00000 liHlS
11 IH) so CM 2G70000 CUBIC METfiHS
1 ) fi FI JllIIS
Man vi l1 e D-2 Baghouse
723 0 347 -633
Cl Ill Y-OJ l LE MIPII I BOLF Jf1BI(UOUS NO PATTEnN NON-ARHFSTOS ALL FI BFllf~
F 11l-ll f()llrlr ( I 1IFH-- l 17 G o 00 00 00 19fi
Tl fL P1 Sl~JNNFD Tl lI ltl- OF 1bullrn~n H 11I HF ~IJlbullI) lll 1 ll(IT 1-ILCTON IL CW ~U-1 HErlllllNE
THl1 (11llrJI~ OF 1IH Tt1T-d 11111middotH COIJrfflI)
= 697~oo so MICRONS l l 1)00 so CM t I JOO t-o CM
= 1 00(00 lHT~ = I l 1)0 ~o GI
ant()Oo cun1r NETEns = 7 0 I l lWllS
Manville (upwind)
CIIIIYsOllY AMPII I nou JMfl[CUOUS NO PA1vrrHN NON-A~~HFSTOS ALL FIBFHS
v1mI1 Cttllffl l 1-- I II I fl~ 1 () () oo l 0 00 40 70
1middot1111unmiddot ToT1L f J BFllS o ooo~ ooow- t OTT~ ( 01rn 57 I 1~m too ooo~
1middot 1111--11 1or111rnnT I ON I Imiddot I 111 II~~ 1111 ~o CM OF f I LTFTl) 1 11111 1middot111 CUil middot11bullnn Illmiddot J l ll)
O OOOlJEmiddotH)O UUUUOL+OO
0 00001-1-00 0 00011 lbull-Imiddot00
G 1 I nn+OG 1 HJ llmiddot~+OG
00001+oo 0001101-1-00
r n~+H+uG l 7llHmiddotlE+Ofl
l bull l 1)1-n+O(i I J ~ 1)(11-middotl()i
T11T I ~Ull FiCImiddot 1 ll f ( ~ l lrl ll11 tn Ii OF F 1ITFll)
f ill I I I i 11 I U l ( i I Ill J
0 OOl)Ol ll() () U()0 I Imiddot()()
0 UOOOlmiddotgt1-00 O OOlgtUlbullgt100
l 0 1) ) I 1 lJ J ~ tl( 1H-HH
0 UUUOlmiddotH)I) o 00001-+oo
~ lt1Hlt1(E+OG 7(H(Ml-middotHM
0 OOOOJbullH)O 0 0000lbullgtHlO
77 I
N ~
rl gt- 10NCJfffll1T I ON
r orR AMS ITll ~q lfI (II- FIT n~ll) - LIi LIIll 111middotTU OI 1lll)(PGRAMS II i1 Ill MlmiddotN
bull I t bull I 1 f I ~ middotr1 Ill lmiddotl 1- [~ IOC ( Hl1I f r~ Ull-F ll
0 0000 Imiddot bull-110 0 0000 Ibull HO
0 ltHgtOO 000110 t) (HJ()()
0()()()()
0()000
llOOOOF+Oo 0 00001bull-100
0 ()000 OOtlOO 00000 0 (WOO 00000
0 (((7
0 ~0-i -0 middotlbullribull)bulll-
0 ltl l lt 0 oHi
00000 00000 00000 0(0HO 0000()
077[0 0 -1117
-0 ~bull)I)()
0 7middot1middot1 (i () L~bull)7
0 7~B( 0 I I
-o alt17n 0 (1)~1 0 17lt1 I
11 I 11ITlt I j i 1l 1 111bull I
f 11middotgt N gt I ii ii I V 111 ii IOC tHlfl flf1 Clll-tmiddotmiddot VAIi
00000 0()000 0000() 00000 O 0000
041000 ()0000 () 0000 o 0(1()0
0000()
0 llfl(i7 OltHJ
-~ lt-11 I 0 07 I 1 () I)lt)
0 (H)ll()
00000 00000 0 ()()()() 0 0(100
0 0 Omiddotlmiddot l
-c07 1)
0 I 1 I( 0 1101
0 l l j--
OOGli -~bullI) ( (J
0 1lt11 I 077G7
I 1 (71 rJI C)
tll N -r11 111-V rmiddotllfl trn CI llf I 111bull1 COIT rll
() ())(l(J
00000 0 ()(1()()
0000() () (1()(10
00000 00000 00000 0 (I()(()
00000
0 ~ I Ifdeggt 0 I Imiddot~ I
- I H 1)7
O I GOO I I i ) l)
0 0001) 00000 00000 00000 00000
0 1 1)17 o~ViH
-1 0 1)gtG 01lbullH 0 7 I till
0 l l lt G 0 ~r1
- J bull J1 1) I 0 17 I bull at-lt
lllllrmiddot lt Clll ii IC)
~I Imiddotmiddot N ~middot1 D 1)1bullI f-1 Imiddotgt f IOC ClltHI rm uw-- i ll
O(lOOO 00000 00000 () ()(J()(J
(l(t(()()
0()()00 00000 00000 (1(1(1(()
00000
000fi7 OOOGO
-G Jll7l 0011G 1 H20
00000 00000 00000 000()() 00000
0 0 I bulllCi 001w
-4 iHd 00101 I ~700
OOIOH 0 0 I Ilt
-(1 j l)~G
000(i 1tHH
--
--
L c~
U c t c
(
J
J -
C
C
c g 0 0
+ + e--c~ =~ - r~ C- l~
--
+ +---
- -C 0 o- -c--1
I
lt t w
lt
C
ta J ~ ltCi
lt ~
a---~ -- c gt C 0 rtl 5 -
VJ C f-rJJ c
[J_
ltI ~ C z
z _ ~
~ lte 0 ~
C
tJ
e
C
~ c C
L N
~~ C 0
C
l~ 1 0
t CI ~ ----~ -
i-
cc cc + + t=~= g~ --
-~ C + + t~ ---[w (~ lC
-
ccc ~t c5 ~ ee 00000
co--oc I
- - - - -_____
------ _ - _ _--__ ---__ ---- _ - _
ooocc
rr ~ c-~ E lt
C
-
~ C g L-
gt o + + ~~ ( L C- lN o -l
t l - -I
~ - L~ [ti e- [
l _ 1
- - ~_ C
gg -- -- ~ -0000-tr--- ____
----C __
- __ - _ --ggggg
- -
C
+ + -- --- -~_shy-----_ _ -----___ ooo~
-----__ ___ _ - - - _ - _ _- - - - ___ 0~000
--__ ---_ __ -- --c5 ooco
C cc
1 II II 11 11 II II
z
-
r ~ c
z
~ t lt i C z __ -- -~tt ~ middot~ ~
-middot middotmiddotmiddot _ - -----=___
_ =~ - _ -=shy ~
-- -~== 2~= ~ () V)
_middot ~ c( c(
0 er ~- ~ ~ lt 0 0___
-
- amp lt~ ~~~t8
F-21
) ) ) )
TOT1L AllE SCiNrmn TIIT1 L ll lbull1 tW FI 1lFll TIIL1 11bulli 1~lllbullll lllIIOT lIIJClION -11- OV TUI Mlbullrllll1Nft TOT1 I VOIIHIIbull 01- I Ill TOTmiddot I Imiddot I I 11-11~~ COi l NTlbulln
_LI__l_ ___l 4 bull P 7 bull P A R l n A 4 R 1 n 4 4 ___ 7 A ___ _RA____ 1 A 1A 8 1 SL__ 1 R 5 1 ~ 3 _1] ________ 111~____JJ__) 1 ~---~-~ 4 9 1 ----1___11
A0 1A4 JR 17 lAR A 41 4C 174 A ll~ JnR ll7 9 5 R12
----- --------- - ---- ---- -- middot-middot --- ---- -- ----middot-- ----middot ----~ 7 c ~ bull n 1 1 2 l 9 -s 1 -i s s ~ c c__ bull A____02 1 1 2 1 r n 1 c 0 r-- l ~ 1 o 1 0 1 t 1 o c- ----~--- -Pl Ol OP lA2 11A 61 c 0 ~ cq03 l 5 4 l 4 4 1 i r 7 -i r) c q 1 deg 1n
qA RO 107 ~_12_7 SA c 7 A04 _l _ s_ _________l 7 l RA____ 7 ______l _o 4 __ l nn_________ 0 _7________ R_l __ o ~ _____ J(__7 11__r ___ _t-__ o ____ ln ___ h_ ______ An____ c-_n __05 ]lA 140 170 7cA __2n1 qc- 7q 1n1
c 2c 7 lR 1 O 406 11 c A 0 4 n 3 4 7 l 4 R ~ 7 7 5
___ ___~ c Z bull A ~R_ ---2_h 2 bull bull 4 e R bull C07 l_ H l_ ri ------- R1 --- c ----------- AQ _middotbull 4_p____ r 7 ___1[_2_ _r--_______ _ A____________ )_3-____ _____ __t 9 ______ ~_____ _3 __ --~___()_______ __I__08 C7 lR9 9A 1n~ ll7 S AS 84 ~Q 33 11 11 1n R 3A 3309 7RJ 1R lA 14-1 17fI 77 4R 7(_
---~- L~ 1 7 -~ 1 7 4 1 l 1 9 S 3 9 C) 4 9 A ~ 9 q10 __ 7 c _ ___ _ 2 RA l A ____J__6_R 1 C 4 7 P C 1 rq I
-----~~a-R___-=4bull2_1 e __ middot-middot middot bulls __ - __ IL 9 1 ~_1_ ___3 7 40__19 1 ~~- gt q6_ (- l Q 144 43 _37 _ A f ---middot ~ bull r bull q ~ 9 IL ~ 1 _0 -~ bull 1 A 3420 l r -- I- I q4 l qo l c c 0