DOE/AL/99567--1 (DE97000259 ) DIStribution Category UC-t414 DEVELOPMENT OF ASPHALTS AND PA VEMENTS USING RECYCLED TIRE RUBBER Phase I: Technical Feasibility Technical Progress Report By Jerry A. Bullin Richard R. Davison Charles J. Glover Cindy Estakhri Raymond W. F1umerfelt Travis Billiter Jay Chum HeamoKoo Vikas Sheth Gerald Elphingstone Clint Eckhardt June 1996 Work Performed Under Contract No. DE-FC04-94AL99567 Prepared for U.S. Department of Energy Office ofIndustrial Technologies Washington, D.C. Prepared by Texas Transportation Institute College Station, Texas
145
Embed
DEVELOPMENT OF ASPHALTS AND PA VEMENTS … · development of asphalts and pa vements using recycled tire rubber ... development of asphalts and pa vements ... mixing speed as a variable
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
DOE/AL/99567--1 (DE97000259 )
DIStribution Category UC-t414
DEVELOPMENT OF ASPHALTS AND PA VEMENTS USING RECYCLED TIRE RUBBER
Phase I: Technical Feasibility
Technical Progress Report
By Jerry A. Bullin
Richard R. Davison Charles J. Glover
Cindy Estakhri Raymond W. F1umerfelt
Travis Billiter Jay Chum
HeamoKoo Vikas Sheth
Gerald Elphingstone Clint Eckhardt
June 1996
Work Performed Under Contract No. DE-FC04-94AL99567
Prepared for U.S. Department of Energy
Office ofIndustrial Technologies Washington, D.C.
Prepared by Texas Transportation Institute
College Station, Texas
DISCLAIMER
This report was prepared as an account of work IIpODSOred by an agency of the United Slates Government. Neither the United States Govcmmcnt nor any agency thereof, nor any of their employees, makes any warranty I express or implied, or assumes any legal liability or Jeoponsibility fur the ~, comp1ercness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference ben:in to any opocific commercial product, proc:cas, or service: by trade name, trademark, lIWIufacturer, or otherwise cines not necessarily constitute or imply its cndonement, recoii1lilCDdalion, or favoring by the United States Clov<nun<m or any agency thereof. The views and opinions of aulhon """","ed herein do not necessarily stale or rellec:t those of the United _ Govemmcm or any agency thereof.
This report has been reproduced directly from the best available copy.
Available to DOE and DOE conttactors from the Office of Scientific and Technical Information, P.O. Box 62, Oak Ridge, TN 37831; prices available from (615)576-8401.
Available to the public from the U.S. Department of Commerce, Technology Administration, National Technical Information Service, Springfield, VA 22161, (703)487-4650.
DOE! ALl99567-1
DEVELOPMENT OF ASPHALTS AND PA VEMENTS USING RECYCLED TIRE RUBBER
Phase I: Technical Feasibility
Technical Progress Report
By Jerry A. Bullin Richard R. Davison Charles J. Glover Cindy Estakhri Raymond W. Flumerfelt Travis Billiter Jay Chum HeamoKoo Vikas Sheth Gerald Elphingstone Clint Eckhardt
June 1996
Work Performed Under Contract No. DE-FC04-94AL99567
For U.S. Department of Energy Office of Industrial Technologies Washington, D.C.
In Cooperation with Texas A&M University Research Foundation Texas Transportation Institute Department of Chemical Engineering
PREFACE
This report documents the technical progress made on the DOE funded project
"Development of Asphalts and Pavements Using Recycled Tire Rubber" for the time period
covering September I, 1994 through August 31, 1995. Cost sharing for this study is being
supplied by the Texas Department of Transportation and industry. Bruce Cranford is the Program
Man<\ger for the DOE Office of Industrial Technologies. Ken Lucien is the Project Officer and
M. Laurene Dubuque is the Contracting Officer, both for the DOE Albuquerque Operations
Office. Frank Childs, the Project Technical Monitor, is on the staff of Scientech, Inc., Idaho
Falls, Idaho. Professors Jerry A. Bullin, Charles J. Glover, Richard R. Davison, and Raymond
W. Flumerfelt, together with Cindy K. Estakhri of the Texas Transportation Institute are the Co
Principal Investigators. Other co-authors of this report are current PhD candidates Travis Billiter,
Vikas Sheth, and Gerald Elphingstone and masters students Jay Chun and Hearno Koo, and
technician Clint Eckhardt.
Work supported by the U.S. Department of Energy, Assistant Secretary for Energy Efficiency and
Renewable Energy, Office of Industrial Technologies, under DOE Albuquerque Operations Office
- . Y = 4.00568015 • 13644x R= 0.99987 - .... Y = 1.96200·17 • 15057x R= 0.99997
10" 0.0020 0.0025 0.0030
1rr (I{"') 0.0035 0.0040
Figure 2·4. Intermediate and High Temperature Data 10% TG·I0 and 90% Exxon AC·5
11
lI(Temperature) correlation, decreases with an increase in curing time. In addition, the viscosity
of the binder decreases as curing time increases at high temperatures. According to Figure 2-5,
extended curing lowers the creep stiffness of the binder. Again, these improvements in the
binder's physical properties can be attributed to the additional amount of rubber that dissolves into
the asphalt as a result of the extended curing time (see Figure 2-6).
Rubber Amount as a Variable
To study the effects of the amount of rubber in a given binder, various samples were
prepared using two different concentrations of rubber, 5 and 10%. All other curing parameters
were held constant except for the amount of rubber in the binders.
Increasing the amount of rubber in a binder was expected to benefit some physical
properties but hinder other properties. The low and intermediate temperature properties benefit
with the additional amount of rubber. Figure 2-7 shows that the binder cured with 10% rubber
has a lower temperature susceptibility than the binder cured with less rubber. However, at high
temperatures, the binder with 10% rubber has a higher viscosity. Figure 2-8 reveals that the creep
stiffness is lower for the binder containing the larger amount of rubber;
Rubber Particle Size as a Variable
The effect of mesh size on the curing process was evaluated by preparing binders using the
various available mesh sizes. Theoretically, smaller rubber particles are much more easily
dissolved in asphalt than larger rubber particles. The smaller graded rubber reacts faster when
cured with asphalt because of the increased surface area. Therefore, if better dissolution improves
asphalt-rubber properties, then using smaller rubber particles benefits all physical properties of
a binder: a lower creep stiffness at low temperatures, a lower temperature susceptibility in the
intermediate temperature region, reduced viscosities at high temperatures (compared to the less
cured state) and reduced curing time.
Figure 2-9 shows that the temperature susceptibility is lower for the binder cured with the
smaller graded rubber (-80 mesh rubber). This figure also reveals that the binder containing the
-80 mesh rubber has lower viscosities at the higher temperatures. From Figure 2-10, it is evident
12
150 r---~~~~~~T7~~~~~--~~~--~ --.10'% TG·10 I Exxon A -5 with 10% G-10 mesh rubber
~ I ~ ::. '",e " m """,00 ."
~ 100 ~ ., ., o
'" @
=:! .. ~ tID
.5 .. C II .. C o " .. 1: ... 2 ... ~ WI WI
is
Curing Time (hrml
Figure 2·5. Low Temperature Data 10% TG·I0 and 90% Exxon AC·S
J
10r-------~~~~~~~_r~~--~~~~ ---10% TG-10 I Exxon "COS with 10% TG-10 me.mh rubber
Cured at 375 F and 500 RPM
B t- -
6 I- -
4 I- .
'CL , , o 10 20 30 40 50
Curing Time (hrml
Figure 2·6. Solubility of Rubber in Asphalt 10% TG·IO and 90% Exxon AC·S
13
'0'
'0'
'0'
'0'
10'
10·'
"'-0 •.. '~:/o TG-40, IT i ........ Y = 2.26'e.' 6 • ""!' .48ge+04Xl R= , -II . 5/Q TG-40, IT :- . y = 7.473e-19 • eA 1 655e+84x R= 1 ---+- FIN A AC·'O TANK, IT ,-Y = 3.3230·20 • eA :748e+ 4x R=
o 10% TG-40. HT 3 • 5% TG-40. HT 1 o FINA AC·' 0 TANK, HT ~"
Fina AC·l0 wllh 5 and 10% TG-40 mesh rubber J Cured for 24 hours @ 375 F and 500 RPM ,
o 0 o • • • 0 o 0
o 0
e e
,.<6'
0.0020 0.0024 0.0028 0.0032 0.0036 1fT (IC"')
Figure 2·7. Effect of Rubber Amount on Temperature Susceptibility 5 and 10% TG·40 with Fina AC·I0
350 r-----~--~-r~~~~~~~~~,_~~-r, --5% Ta-10 Fin .. ACo1!l. lind TI ... Gator Blends • · .. <>···5% TG-40 Cured III 375 F lind 500 RPM
~ 300 -0 ... '0% TG·,O - ·10%TG-40 .
II) -• 4 TANK
... fa 250 '" Tank
~ ... 200
• i 150 -GO
;.
:: 100 -~ iii
50 -
/ o
Curing Tlma (hrs)
Figure 2·8. Low Temperature Data Fina AC·I0 lind TG Blends
14
-
-
-
u o
eli.
!! .. .. ~ iii ... .. I!! u
, -e---- As·ao IT , ........ AS •• W: IT
j j 1 iExxon AC~5 and Rouse Blends
107 ~ ·AS-10,1T :Cured for 24 hours at 375 F and 500 RPM
10'
---'9"- . EXXON AC-5 TANK. IT Q AS·BO, HT II AS-40, HT o AS-10,HT ... EXXON AC-5 TANK HT
00 .0 8 o •
II e ,," ~" "
--- y = 3.36240,14 • e"l12866x) R= 0.99995 .. ....... Y = 1.5650·14 • e"(13167x) R= 1.00000 - . y = 1.0895.·14' e"f13410x) R= 0.99997 - - . y = 1.962 ... 17' e"(15057x) R= 0.99997
c c - (>- . 10/90. EXX AC-l0. TG-40 t. ·5/95, EXX AC-l0 RS -40 Q-. ·10/90. EXX AC-l0, AS -40
POV-aged a12,0 OF with Atmospheric Air
Blends cured at 375 OF and 500 rpm for 12 hours under N2
--Y = 1492' O"{O.'844x) A. 0.94'2
- . y = 1.B736+04 • &",{O.126x) R= 0.9647
- - - y • 1.4651Hll4 • 0"{0.1443x) A. 0.9876
.... -y. 1.1 171Hll4 • 0"{0.158x) A. 0.9759
......... Y • 1.21l41Hll4 • 0"{0.1273x) A. 0.9876
10 15 Time (Dsys)
20
Figure 2·33. Hardening Rates of Exxon AC·I0 lind Blends at 210 OF
ElOIon Ac-5 and Blanda
1 Ha. !ening R~te Activation Energies -e- EXXON Ac-5
-e . 5195. EXX AC-5. TG -10. POV DATA
POf/·AGED at leo. 200. and 210"1' - .. ·10190. EXXAC-S. TG-IO. POV DATA ...... ·5195. EXX AC-S. AS -10. POV DATA .. • ..... ·10190. EXXAC-S. AS-IO. POV DATA and E VROOM·AGED at 14O'F under Air
0.1 l:-
0.01 I:-
~---. ..... ~.-.. ~
• EXX Ac-5. 14O'F DATA
• 5195. EXX AC-5. TG -lOB. 140 OF DATA • 10190. EXX AC-S. TG -lOB. 140 OF DATA
• 5195. EXX AC-S. AS -10. 140 OF DATA
• 10190. EXX AtrS. RS -40. 140 OF DATA
Blends cured at 375 OF and 50 rpm
__ y. 1.2048+10. 0''{.9301.) Ro 0.9792 for 12 hours under N,
- :y _3.1970+06' 0"{-/l249x) Ro 0.9821
- - • y _ 5.0120+06 • 0"{-6479.) R. 0.9911
- •• y. 3.3720+07' 0"{·7163.) Ro 0.957
......... y. 3.550+07 • 0"{·7247.) R.O.9339
/-ENV·Room
140"F Data
0.001 , , , ,
0.0027 0.0028 0.0029 0.003 llTemp (11K)
Figure 2·34. Hardening Rate Kinetics Plot of Exxon AC·5 and Blends
32
corresponding base asphalt. This prediction from the POY elevated-temperature data is
contradicted by the 140°F ENV-Room data in Figure 2-35,2-36, and 2-37, which represent the
3 base asphalts; Fina AC-IO, Exxon AC-5, and Exxon AC-1O aged at 140°F. Figures 2-35,2-36,
and 2-37 show that the hardening rate (slope) at 140°F is not a function of rubber content. The
data that was in the initial jump region is labeled with U in the figures. The initial jump region
is defined as the time before In Tj" is linear with time. An in-depth analysis of numerous asphalt
and asphalt-rubber blends is required to verify this finding.
Although an asphalt-rubber binder hardens as fast or faster than its base asphalt at 140"F,
this hardening may not be as detrimental to the asphalt-rubber binder as it is to the base asphalt.
The hardening is not as detrimental because the rate of change of elasticity of the binder, as
measured by delta, 0, (A material with 0=90" is perfectly viscous, whereas a material with 0=0°
is perfectly elastic.) with aging time is more negative for the asphalt-rubber binder than the base
asphalt. This is shown in Figures 2-38, 2-39, and 2-40, representing the 3 base asphalts; Fina
AC-IO, Exxon AC-5, and Exxon AC-IO aged at 140"F, and implies that for the same amount of
aging time, the elasticity of an asphalt-rubber binder increases more than the elasticity of its base
asphalt.
Additionally, several asphalts and cured asphalt-rubber blends are currently being aged in
the ENV room and will be POV aged as well. These samples will be analyzed to determine the
effects of rubber content (10 and 20%) and high shear rate of curing on aging properties. These
samples include:
Asphalts: (Total of 4)
Fina AC-I0
Exxon AC-5
Exxon AC-IO
Fina Demex Resin
Asphalt-Rubber Blends: (Total of 8, 4 asphalts x 2 blends)
2 blends of each asphalt:
with 10% TG-40 Buff
with 20% TG-40 Buff
33
." c: .. (.) • lil iii '.,.
i
10'
10'
la'
10' o
10'
¥ 10' I-
j C! -." c: .. ~ 10' ~
lil iii '.,.
10' o
FINA AC·l0 AND BLENDS
ENVROOM·AGED AT 140 of UNDER AIR _ ~~~ v
_o-::':::::_:&-" 0- .-0- =.~E·;"'-
o ~-:iI2:!?-' ti - _ _ &
o
oX J::: IJ = Initial Jump Region
• FINA AC-10, TK III 5110, FINA AC·l0, TG -40, TK
o o
Blends cured at 375 of and 500 rpm for 12 hours under N,
Blends cured at 375 OF and 500 rpm : ~~A~~?R1G-46?TkK ~ for 12 hours under N2 v 10190. EXX AC-1 0, RS -40, TK "'- -e-EXXAC·10 -~~e::::::,.. -e -5I95,EXXAC-10,TG-40
Figure 3-6. Effect of Binder Curing TIlDe on Gyratory Compactibility Index (GCl)
time on compactibility can be seen in Figure 3-6 which indicates that no difference was detected
in compactibility between I-hour and 6-hour mixes.
The gyratory stability index (GSI) is calculated as the ratio of the maximum gyratory angle
to the minimum gyratory angle. A GS! in excess of unity indicates a progressive increase in
plasticity during densification. An increase in this index indicates an excessive bitumen content
for the compaction pressure employed and foretells instability of the bituminous mixture for the
loading employed. A mix GS! in excess of unity also indicates the likelihood of the mixture to
permanently deform. The effect of CRM particle size on GS! is shown in Figure 3-7. All of
these mixtures have acceptable GSIs. The effect of CRM concentration on GS! is shown in Figure
3-8 which indicates instability in the mixture containing 25 percent CRM in the binder. The
binder curing time (Figure 3-9) does not appear to affect the GS!.
45
Gyratory Stability Index 1.2,-----------------------------------------,
0.8
0.6
0.4
0.2
O~~~~----~~~----~~~----~~~--~
Control -#10 mesh -#40 mesh -#80 mesh
CRM Particle Size
Figure 3-7. Effect of CRM Particle Size on Gyratory Stability Index (GSI)
Gyratory Stability Index 1.2r---------------------------------------~
0.8
0.6 ,:.: ..
0.4 .'::"""
0.2
O~~~~----~~~--~~~----~~~~
10% 18% 25% Control
CRM Concentration
Figure 3-8. Effect of ClRM Concentration on Gyratory Stability llndex (GSI)
46
Gyratory Stability Index 1.2r-----------------_
1
0.8
0.6
0.4
0.2
0
• 0·. ~ ...
-#10 mesh -#40 mesh
CRM Particle Size
Ed 1-hour Cure
o 6-hour Cure
. ...... ;.
-#80 mesh
Figure 3-9. Effect of Binder Curing Time on Gyratory Stability Index (GSI)
Based on the results presented herein, the GTM did not measure adverse compactibility
properties of the mixtures tested. However, the GTM measurements are made during the
compaction process (while the sample is under load). When the load was removed from the
sample, some significant changes were observed. The samples which were compacted containing
the -#10 mesh rubber swelled significantly within the first24 hours after compaction. One sample
made with the -#10 CRM even disintegrated upon removing it from the mold. Sample heights
were taken at the end of the compaction process and again 24 hours after removing them from the
mold. These data are shown in Figure 3-10. Note that the -#10 CRM mixture was significantly
taller 24 hours after extrusion. Intuitively, this characteristic would be highly undesirable in a
field mixture. It indicates that CRM mixtures may certainly compact in the field under the weight
of the roller; however, when the roller is removed the density may become unacceptable.
47
Sample Height. inches 3.51-----~ _______ ~ _ ___,
3-- ---
2.5
2
1.5
1
0.5
El Before Extrusion
o 24-hrs After
......... ~ ....
.... , ..•. --.~ .... ,.
O~~~~--~~~---L~~--~~~~ Control -#10 mesh -#40 mesh -#80 mesh
CRM Particle Size
Fagure 3-10. Sample Height Before Extrusion from the Mold and 24-Hours After Extrusion
In the past, it has been standard practice to allow CRM asphaltic concrete samples to cool
in the mold to prevent the sample from swelling; however, this practice may be deceptive. It
may be better practice to extrude the sample after molding and then measure if swelling occurs.
If swelling does occur, then adjustments to the mixture design (in particular, the aggregate
gradation) should be made.
The following are some preliminary conclusions based on this effort:
8 The TxDOT mixture design procedure used in this study, in general seems
to be acceptable for the design of CRM asphaltic mixtures.
CRM mixtures designed _ according to the TxDOT mixture designed
procedure appear to be resistant to permanent deformation, with the
48
exception of the mixture which contained a large concentration of CRM
(25%).
The CRM binder curing times evaluated in this study did not affect
compactibility or permanent deformation characteristics of the asphaltic
concrete mixtures.
CRM particle size is the main variable evaluated in this study which affects
its compactibility, based on measurements of sample heights taken after
extrusion from the mold. CRM particle size of -#10 or greater is of
greatest concern. While it may be possible to design a mix to accommodate
CRM of this size, the mixture design in this study was not adequate.
EVALUATE DEFORMATION AND FAILURE OF COMPACTED MIXTURFS
In the compaction study, a methodology for designing good mixtures was verified. In
Subtask 2.3, the methodology is being applied to design and compact mixtures having different
asphalts and binder preparation methods. These materials will be aged and tested for deformation
and failure using three test methods: (1) "nondestructive" sinusoidal frequency sweeps (fully
reversed tension-compression), (2) creep and recovery, and (3) tensile strength to failure. A
summary of the experiment plan is shown below.
Aging CRMBinder Time
1 2 3 4 5 6 7 8 9 10 11 12
0
1
2
3
49
Preparation of binder samples to be tested for this subtask is complete. Compacted
samples will be prepared using six different asphalt cements combined with 18 % -#80 CRM cured
under two different conditions for a total of 12 binders. The six asphalt cements to be used in the
experiment are as follows:
Diamond Shamrock Resin,
Fina Resin,
Fina AC-lO,
Fina AC-5,
Exxon AC-5,
Exxon AC-lO.
Two curing conditions were used for preparation of the binders:
375°F, 1550 rpm, 6 hours, and
350°F, 500 rpm, 1 hour.
All of these twelve binders have been prepared. Aggregate has also been prepared for preparation
of the compacted samples and compaction of samples is ongoing.
50
CHAPTER 4
ADHESION TEST PROPERTIES
The strength of an asphalt-aggregate composite mix and its performance under varying
loads and environmental conditions are key factors in determining pavement lifetimes. These
factors strongly depend on the cohesive properties of the asphalt constituent and the adhesive
properties at the asphalt-aggregate interface (Labib, 1992).
The work of cohesion is the work required to create two interfaces from one phase (see
Figure 4-1):
(4.1)
where aGo is the Gibbs free energy of cohesion, and Y I is the surface energy of phase 1. The
work of adhesion is what is needed to create two interfaces from two phases in contact (see Figure
4-2):
(4.2)
where aG'12 is the Gibbs free energy of adhesion and Y 12 is the interfacial surface energy of phase
1 and phase 2.
The cohesive and adhesive bonding interactions in asphalt-aggregate systems arise mainly
from two effects: (1) the Lifshitz-van der Waals interactions of electron shells df neighboring
molecules and (2) the acid-base interactions between constituent molecules (Good and van Oss,
1992). The acid-base interactions are generally dominant for asphalt-aggregate composites and
are particularly critical in establishing strong adhesive bonds as well as bonds that are resistant to
water enhanced stripping (Labib 1992). The acid-base interaction term is further partitioned into
a Lewis acid parameter and Lewis base parameter. Thus, three parameters must be determined
to calculate the surface energy of the material: Lifshitz-van der Waals, Lewis acid, and Lewis
base. The surface energies of the materials, the asphalt and aggregate, are used to calculate the
work of cohesion and the work of adhesion. It should be noted that the Lifshitz-van der Waals
force is present in all molecules, but acid-base interaction are not. In fact, the acid-base
interactions will be the key in determining the compatibility of asphalts and aggregates.
51
1
Vacuum
1
1
Figure 4-1. Work of Cohesion
1
1
Vacuum
2
2
Figure 4-2. Work of Adhesion
52
The usefulness in being able to calculate the work of cohesion and adhesion by measuring
the surface energies of asphalts and aggregates are threefold: (I) The mechanical work required
to crack an asphalt-aggregate interface can be predicted. This theoretical work of adhesion would
correspond to a fracture test measured below the brittle-ductile transition· temperature. (2)
Numerous crack propagation models exist; but, these models require difficult experiments to
determine the parameters. It can be shown that these parameters are functions of the surface
energies of the materials. By measuring the surface energies of asphalts and aggregates, these
parameters can be calculated. (3) Asphalt-aggregate systems can be evaluated for the propensity
to be water susceptible. Suitable asphalts may be found for the so-called" stripping aggregates."
The focus of this portion of the project is establishing the framework for predicting adhesion and
cohesion in asphalt-aggregate systems. The main goals of this investigation are: (1) to develop
test methods that predict adhesion and cohesion in asphalt/asphalt-rubber and aggregate systems,
(2) to develop test methods predicting adhesion and cohesion in asphalt/asphalt-rubber and
aggregate systems in a water environment, and (3) to use the test methods to evaluate
asphalt/asphalt-rubber and aggregate mixes.
ADHESION TESTS
Wilhelmy Plate Method
The Wilhelmy plate method (Wilhelmy, 1863) is an established technique for measuring
contact angles of liquid/solid systems and is being used in this study. The contact angle is
determined by measuring the change in force during immersion and emersion cycles (see Figure
4-3):
~F = py cose (4.3)
where ~F is the change in force; p is the perimeter of the plane; y is the surface tension of the
liquid; and e is the contact angle between the solid and liquid measured through the liquid.
. The apparatus consists of three main components (see Figure 4-4): the Cahn C2000
balance for measuring the force; the moveable platform to advance and recede the liquid;
53
(al before plate contacts with liquid
Liquid
(b) plate just contacts the liquid-air surface
Liquid
(c) plate is partially immersed in the liquid
. " Liquid
Figure 4-3. Dynamic Wilhelmy Plate Method Force Balance
54
glass plate
Cahn C2000 balance
step motor
shielding case
computer data acquisition
& control
lFigwre 4-4. Wilhelmy Plate Apparatus
55
and a computer for data acquisition and control. Glass plates are coated with asphalt to produce
a smooth solid surface. The contact angle of four fluids (water, glycerol, ethylene glycol, and
formamide) are measured against the asphalt coated glass plate. From the advancing and receding
contact angle measurements of two fluids, the three parameters of the surface energy can be
determined. The two remaining fluids are used to verify the results.
An inherent problem with the Wilhelmy plate technique is contact angle hysteresis.
Hysteresis is the difference between the advailcing contact angle and the receding contact angle.
It is believed that for heterogeneous materials that the hysteresis effect can be explained. While
the liquid "sticks" on the high energy acid-base regions and only wets the Lifshitz-van der Waals
regions. Therefore, the acid-base interactions are negligible while advancing. While receding,
the liquid preferentially wets the high energy regio}ls; the receding contact angles are a measure
of the total surface energy.
Figure 4-5 is an example of a typical set of experimental results. The lower straight-line
region corresponds to the advancing angle while the upper straight-line corresponds to the receding
angle. For each experiment, a total of 5 advancing and receding cycles are measured. As can be
seen from Figure 4-5, the reproducibility for a single plate is quite good.
A total of 19 asphalts have been characterized using the Wilhelmy plate method (see Table
4-1). Figure 4-6 is a graphical representation of the two surface energy components: yLW is the
Lifshitz-van der Waals force while yAB is the acid-base interactions. All of the asphalts exhibit
a wide range of values.
Gas Adsorption
The aggregates pose a more difficult problem than the asphalts. Since a smooth surface
cannot be made from the aggregates, gas adsorption must be utilized (see Figure 4-7). A total of
four gases are needed for this process; three to characterize each aggregate with the fourth being
used to check the results. The gases used are hexane, water, methyl propyl ketone, and
chloroform.
Six aggregates have been characterized with one being separated into 7 size gradations
56
Asphalt Coated Plate AAG-1 in Glycerol
350
300
-I/) 250 CI) I: >-'C 200 ...... CI) u 150 ... 0 u.
100
50
0 -0.4 -0.2 0 0.2 0.4 0.6 O.B
Depth (em)
Figure 4-5. Example Experimental Results
60 -- 1 8 (\I < 50 E -.., E 40 -->-i:7)
30 "-til I: W
til 20 u m - 10 ... ::::I rn
0
Figure 4·6. Surface Energy of Various Asphalts
57
Pressure Gauge
High
Hand Pump
Nitrogen In!d.
Go to Fume Hood forRc1easc
Go to Fume Hood
Safety Disc
J" ransducer
1tfj....,r-tf;I-Pressure Release
Figure 4-7. Gas Adsorption Experimental Apparatus
58
988 Funoe!
(see Table 4-2). Figure 4-8 represents the two components of the surface energy. Both
components are much larger for the aggregates than for the asphalts.
Adhesion and Cohesion
From the measured surface energies, the work of cohesion and the work of adhesion can
be calculated. The work of cohesion for various asphalts in a vacuum and in water can be seen
in Figure 4-9. For less polar asphalts (having small acid-base parameters), water increases the
work of cohesion.
If the work of adhesion is greater than the work of cohesion, fracture should occur in the
asphalt. If the work of the adhesion is less than the work of cohesion, fracture should occur at
the asphalt-aggregate interface. Two aggregates were analyzed with various asphalts in a vacuum.
According to Figures 4-10 and 4-11, fracture should occur in the asphalt binder. Figure 4-12 is
a comparison of the two aggregates in a vacuum. Both JG11a and JG21 have similar works of
adhesion.
WATER SUSCEPTIBILITY TESTS
According to Figures 4-13 and 4-14, water will decrease the work of adhesion. For
aggregate JG 11a, some asphalts have the potential to strip (see Figure 4-15) since the work of
cohesion is greater than the work of adhesion. For aggregate IG21, virtually all the asphalts have
the potential to strip (see Figure 4-16). In Figure 4-17, the aggregates are compared by the work
of adhesion in water. The results suggest that IGlla is better aggregate than JG21 for all asphalts.
Summary of Adhesion Tests
Using the work of adhesion and the work of cohesion, the asphalt-aggregate systems can
be ranked according to their potential to water strip (see Table 4-3). These are presented from
highest to lowest with the acceptable systems having a positive difference and the unacceptable
systems having' a negative difference.
59
Table 4-1. Characterized Asphalts
Code Description AAB-J SHRP WYoming Sour AC-lO AAD-J SHRP California Coastal AR 5000 AADN AAD-J + 6% PE. Novophalt AAG-J SHRP California Valley AR 4000 AAM-l West Texas Intermediate AC-20 EX 10 Exxon AC-lO
hXlOa EXIO + 5% CaO EX5 Exxon AC-5 EX5a I EX5 + 5% Rouse -40 Mesh F110 . FinaAC-lO JA: 1 RTFO A edJG3l
II-J..;;;;;G-'~j li-nF.Ku~wT.~O/lOi JUjL I Saudi /Iv, 100 [Esso) JU33 Venezuela 80! 100 (Nvnas)
JC 33a J(i33 Jlijjb I J( .33 BUDble Al!;ed tor 24{J hours JU34 I KuWaIt 45/60
AASHTO TP3, "Standard Test Method For Determining the Fracture Properties of Asphalt Binder in Direct Tension (DT)," AASHTO Provisional Standard, Edition lA, 1-13 (September 1993).
Allison, K., "Those Amazing Rubber Roads," Rubber World, 47-52 and 91-106, respectively, (March, April 1967).
Anderson, D.A., D.W. Christensen, R. Dongre, M.G. Sharma, J. Runt, and P. Jordhal, Asphalt Behavior at Low Service Temperatures. Report FHWA-RD-88-078, FHWA, U.S. Department of Transportation (1990).
Andrade, E.N. da C., "The Viscosity of Liquids," Nature, 125, No. 3148,309-310 (1930).
ASTM D113, ·Standard Test Method for Ductility of Bituminous Materials," Annual Book of ASTM Standards, Vol. 04.03, 23-25 (1994).
Bahia, H.U., D.A. Anderson, and D.W. Christensen, "The Bending Beam Rheometer; a Simple Device for Measuring Low-Temperature Rheology of Asphalt Binders," Proc. Assoc. Asphalt Paving Technol., 61, 117-153 (1992).
Blow, C.M. and C. Hepburn, Rubber Technology and Manufacture, Butterworth Scientific, 2nd ed., Boston (1982).
Blumenthal, M., ·Using Scrap Tire Rubber in Asphalt,· BioCycle, 32(10), 85-86 (1991).
Bullin, J.A., R.R. Davison, C.J. Glover, and T.C. Billiter, Optimization of Rubber Content in Asphalt Pavement, Federal Highway Administration Research Report FHW A-RD-94-001 (1994).
Chehovits, J.G., R.L. Dunning, and G.R. Morris, ·Characteristics of Asphalt-Rubber by the Sliding Plate Microviscometer,· Proc. Assoc. Asphalt Paving Technol., 51, 240-261 (1982).
Corbett, L.W., ·Composition of Asphalt Based on Generic Fractionation, Using Solvent Deasphaltening, Elution-Adsorption Chromatography, and Densimetric Characterization," Anal. Chern., 41, 576-579 (1969).
Davison, R.R., I.A. Bullin, C.I. Glover, 1.R. Stegeman, H.B. Jemison, B.L. Burr, A.L.G. Kyle, and C.A. Cipione, Design and Manufacture of Superior Asphalt Binders, Texas
73
Dept. of Trans. Research Report No. 1155 (1991).
Davison, R.R., J.A. Bullin, C.J. Glover, H.B. Jemison, C.K. Lau, K.M. Lunsford, and P.L. Bartnicki, Design and Use of Superior Asphal! Binders, Texas Dept. of Trans. Research Report No. 1249 (1992).
Davison, R.R., J.A. Bullin, C.J. Glover, J.M. Chaffin, G.D. Peterson, K.M. Lunsford, M.S. Lin, M. liu, and M.A. Ferry, Verification of an Asphalt Aging Test and Development of Superior Recycling Agents and Asphalts, Texas Dept. of Trans. Report No. 1314 (1994).
Dempster, D., "America Finds New Uses for Scrap," European Rubber Journal, 161(4), 22-26 (1979).
Dempster, D., "Rubber Could Give The Road 100 Year Old Road Surface," European Rubber Journal, 160(4), 47-48 (1978).
Estakhri, C.K., S. Rebala, D. little, LaboralOry Evaluation oj Crumb-Rubber Modified (CRM) Binders and Mixtures, Texas Department of Transportation Report #1332-1, written by Texas Transportation Institute, Texas A&M (1993).
Ferry, J., Viscoelastic Properties oj Polymers, John Wiley and Sons, 4th ed., New York, NY (1985).
Franta, I., Elastomers and Rubber Compouding Malerials, Elsevier, New York, N. Y ., 302-315 (1989).
Gagle, D. W., H.L. Draper, and R.J. Bennett, • Asphalt Rubberizing Compositions," United Stales Palent 3, 779,964 (1973).
Good, R.J. and C.J. van Oss, "The Modem Theory of Contact Angles and the Hydrogen Bond Components of Surface Energies,· in Modem Approaches to Wettability, Ed. E.M. Schrader and G. Loeb, Plenum Press, New York, 1-27 (1992).
Heitzman, M., "Design and Construction of Asphalt Paving Materials with Crumb Rubber Modifier," Transp. Res. Rec., 1339, 1-8 (1992).
Huff, B.J. and B.A. Vallerga, "Characteristics and Performance of Asphalt-Rubber Material Containing a Blend of Reclaim and Crumb Rubber," Transp. Res. Rec., 821, 29-36 (1979).
Hveem, F.N., E. Zube, and J. Skog, "Proposed New Tests and Specifications for Paving Grade Asphalts," Proc. Assoc. Asphalt Paving Technol., 32, 271-327 (1963).
74
lemison, H.B., B.L. Burr, R.R. Davison, I.A. Bullin., and C.I. Glover, "Application and use of the A TR, FT-IR Method to Asphalt Aging Studies," Fuel Sci. Techno/. Int., 10, 795-808 (1992).
Labib, M. W., "Asphalt-Aggregate Interactions and Mechanisms for Water Stripping," Pre prints of Papers, 37(3), 204th ACS National Meeting, Washington, D.C., American Chemical Society, Div. of Fuel Chemistry, 1472-1481 (1992).
Lalwani, S., A. Abushihada, and A. Halasa, "Reclaimed Rubber-Asphalt Blends Measurement of Rheological Properties to Assess Toughness, Resiliency, Consistency, and Temperature Sensitivity," Proc. Assoc. Asphalt Paving Technol., 51, 562-579 (1982).
Lau, C.K., K.M. Lunsford, C.I. Glover, R.R. Davison, and I.A. Bullin, "Reaction Rates and Hardening Susceptibilities as Determined from POV Aging of Asphalts," Transp. Res. Rec., 1342, 50-57 (1992).
Linden, R.N., I.P. Mahoney, and N.C. Jackson, "Effect of Compaction on Asphalt Concrete Performance," Paper No. 880178 presented at the Transportation Research Board 68th Annual Meeting, Washington, D.C., January 22-26 (1989).
Liu, M., K.M. Lunsford, R.R. Davison, C.I. Glover, and J.A. Bullin, "The Kinetics of Carbonyl Formation in Asphalt," AIChE J., 42(4), 1069-1076 (1996).
McDonald, C.H., "Elastomeric Pavement Repair Composition for Pavement Failure and a Method of Making the Same," United States Patent 3,891,585 (1975).
McQuillen, J.L., H.B. Takallou, R.G. Hicks, and D. Esch, "Economic Analysis of RubberModified Asphalt Mixes," J. Transp. Eng., 114, 259-277 (1980).
Nadkarni, V.M., A.V. Shenoy, and J. Mathew, "Themomechanical Behavior of Modified Apshalts," Ind. Eng. Chern. Prod. Res. Dev., .24, 478-484 (1985).
Ohta, K. and translated by E.A. Inglis, "Modified Asphalt for Asphalt Road Surfaces," Int. Polymer Sci. Techno!., 10(7), T/44-T/54 (1983).
Oliver, J. W .H., "Modification of Paving Asphalts by Digestion with Scrap Rubber," Transp. Res. Rec., 821, 37-44 (1979).
Pearson, C.D., G.S. Huff, and S.G. Gharfeh, "Technique for the Determination of Asphaltenes in Crude Oil Residues," Anal. Chern., 58, 3266-3269 (1986).
Peters, M.S. and K.D. Timmerhaus, Plant Design and Economicsfor Chemical Engineers, 4th
75
edition, McGraw-Hill, Inc., New York, 210 (1991).
Roberts, F.L., P.S. Kandhal, E.R. Brown, R.L. Dunning, "Investigation and Evaluation of Ground Tire Rubber in Hot Mix Asphalt," In Florida DepaT7l11ent of Transponation Repon, written by National Center for Asphalt Technology, Auburn University (1989).
Roberts, F.L., P.S. Kandhal, E.R. Brown, D.Y. Lee, and T.W. Kennedy, HO[ Mix Asphalt Materials, Mixture Design and Construction, NAPA Education Foundation, 1st ed., Lanham, Maryland, 68 and 382 (1991).
Sainton, A., "Advantages of Asphalt-Rubber Binder for Porous Asphalt Concrete: Paper No. 890163 presented at the Transportation Research Board's 69th Annual Meeting, Washington, D.C., January 7-11 (1990).
Shuler, T.S., R.D. Pavlovich, and J.A. Epps, "Field Performance of Rubber-Modified Asphalt Paving Materials," Transp. Res. Rec., 1034, 96-102 (1985).
Stroup-Gardiner, M., D.E. Newcomb, and B. Tanquist, "Asphalt-Rubber Interactions," Preprint of paper presented at the Transportation Research Board 72nd Annual Meeting, Washington, D.C., January (1993).
Takallou, H.B., R.G. Hicks, and D.C. Esch, "Effect of Mix Ingredients on the Behavior of Rubber-Modified Asphalt Mixtures," Transp. Res. Rec., 1096, 68-80 (1986).
TakalIou, H.B., and M.B. Takallou, "Recycling Tires in Rubber Asphalt Paving Yields Cost, Disposal Benefits, n Elasromerics, 123(7), 19-24 (1991).
Thenoux, G., C.A. Bell, and J.E. Wilson, "Evaluation of Asphalt Physical and Fractional Properties and Their Interrelationship,· Paper No. 870531 presented at the Transportation Research Board 67th Annual Meeting, Washington, D.C., January 11-14 (1988).
Wilhelmy, L., 'Ueber die Abhlingigkeit der Capillaritats-Constanten des AIkohols von Substanz und Gestalt des benetzten festen Korpers," AnnalenDer Physik Und Chemie, 119, 177-217 (1863).
76
AASHTO
AC M1 AR ASTM ATR BBR CA CRM CRMA DOE-OIT DOT DS ENV ES FTIR GCl GPC GSl GTM HPLC D IRR MP P POV PTFE RA ROSE RPM RS S SA SC SHRP TG THF TK TxDOT WC
ABBREVIA TIONS
American Association of State Highway and Transportation Officials
asphaltene content aromatic material asphalt rubber American Society for Testing and Materials attenuated total reflectance bending beam rheomater carbonyl area crumb rubber modifier crumb-rubber modified asphalt Department of Energy Office ofIndustrial Technologies Department of Transportation Diamond Shamrock environmental exposed surface Fourier transform infrared spectroscopy gyratory compactibility index gel permeation chromatography gyratory stability index gyratory test machine high performance liquid chromatography initial jump internal rate of return metering pump pressure or pressure gauge pressure oxygen vessel poly tetrafluoroethylene recycling agent residual oil supercritical extraction revolutions per minute Rouse separators superior asphalt supercritical Strategic Highways Research Program Tire Gator tetrahydrofuran tank Texas Department of Transportation water cooled condenser
77
NOTATION
cross-sectional area effective gauge length elongation at failure Lifshitz-van der Waal's component of surface energy load at failure low frequency limiting viscosity
78
APPENDIX A
EXPERIMENTAL METHODS
SUPERCRITICAL FRACTIONATION
A brief description of the supercritical fractionation process, operating conditions, and
apparatus modifications follow. The following description is taken primarily from the TxDOT
Study 1249 (Davison et al, 1992) report with appropriate modifications. The unit operates at a
constant pressure above the critical pressure of the solvent. The SC fractionation unit separates
heavy petroleum products into up to four fractions according to solubility in SC solvents .. The
temperatures of the separators determine the density of the solvent and, consequently, the solvent
power in each vessel. Components of the feed precipitate when no longer soluble in the solvent.
The lightest, most-soluble materials are removed by decompression during solvent recovery.
Figures A-I and A-2 illustrate schematically the SC unit. The solvent is pumped to the
operating pressure in S 1-S3 by MPl. Several hours are required to bring the temperatures to the
desired steady-state values. The steady-state operating temperature in S4 determines the steady
state pressure for S4. Once steady-state conditions are achieved, MP2 is activated, introducing
feed material into the circulating solvent stream. The temperature in each separator determines
the solubility in the SC solvent. The insoluble material is transferred from the separator to its
.corresponding collector periodically to avoid potential plugging problems while the soluble
material travels to the next separator. Finally, the overhead mixture from S3 passes through the
control valve, where the pressure is reduced to significantly subcritical value. At these gaseous
conditions, none of the asphaltic material is soluble and complete separation of the solvent is
achieved. The solvent then passes overhead, is condensed in WCI and flows back into the solvent
reservoir. For this DOE effort, n-pentane is the solvent used for supercritical fractionation.
The four asphalts fractionated during the first year of this DOE effort were fractionated
in two passes through the unit. The lightest fraction from the first pass was fed through the unit
79
00 o
Al
','
Filler
P6
C3 C4
C2
Figure A·1. Supercrltlcal Unit Process Diagram WC2
so
C1-C4
III
C?
Solvent Tank
Collectors
In-line Filter/mixer
Control Valve
Tubing Wall Temperature Thennocouple
Pressure Controller
Heating Tape Heater
Dual Purpose Heater/Cooier
S1-S4 Separators
A1 Asphalt Tank
Valve
<y Thennocouple
~ Temperature Monitor and Controller
Pressure Gauge
Metering Pump
WC1
~ Water Cooled Condenser
Figure A·2. Legend for Supercritical Extraction Unit Diagram'
81
a second time yielding eight fractions that may be analyzed. The lightest fraction from the second
pass is designated as fraction F I and the heaviest fraction from the first pass is designated as
fraction F8 (fraction F5 is the feed material for the second pass through the unit).
PRESSURE OXYGEN VESSEL (POV)
The original unit is described by Lau (1992) and Davison et al. (1992). In order to
improve on aging simulation capacity, four additional units were constructed and a central control
panel was installed as shown in Figure A-3. Later, to eliminate temperature gradient problems
with the initial design, the vessels were placed in glycol/water baths.
Figure A-4 shows a schematic of one of the pav s. The vessels are located behind a steel
wall in an explosion proof hood. Each vessel is contained in an aluminum barrel filled to the
bottom of the top flange with a mixture of triethylene glycol and water. The vessel is monitored
and controlled from a panel outside the explosion proof hood. The control panel houses a
compound pressure gauge to monitor the pressure, a variable transformer to control the amount
of electrical power to the heating elements in the water/triethylene glycol bath, a temperature
controller which controls the temperature of the bath, and a recorder to monitor the temperature
within the pav. A stirrer is employed in the bath to insure that the temperature distribution in
the bath is uniform. A vacuum pump is used to evacuate the vessels before charging with oxygen
or to remove oxygen depleted air once per day. Three valves per vessel, as labeled in Figure A-I,
are used for venting to atmospheric pressure, evacuating to low pressure to remove the gas inside
the vessel, and charging with oxygen. The oxygen feed valve isobites the pays from the oxygen
cylinder when closed.
Asphalt samples are prepared in aluminum trays. The dimensions of the tray are 7.0 cm
(2.75 in) by' 3.5 cm (1.38 in). Typical ftlm thicknesses of less that 1 mm (0.039 in) are used to
minimize potential diffusion problems at low pressure; however, diffusion studies may be
performed with thicker films. After preparing the asphalt samples, loading the sample rack, and
allowing the temperature in the pav to reach equilibrium, the operator places the rack inside the
pav and bolts the cover flange to the top. The vent valves, oxygen feed valves, and vacuum
Figure A·4. Pressure Oxygen Vessel and Control Panel
83
valves are closed. A vacuum pump evacuates the air in the vessel to a pressure of 0.03 atm
absolute. The vessels are slowly pressurized to the desired level by manipulating the oxygen
cylinder regulator and oxygen feed valves for pure oxygen aging, or by slowly opening the
atmospheric venting valve for aging with air (note 0.2 atm oxygen is equivalent to atmospheric
air aging). Once the desired oxygen pressure is reached, the cylinder, regulators, and feed valves
are closed.
During the experiment, samples are periodically removed. To obtain samples, the pressure
in the vessel is decreased by slowly venting off the oxygen to the atmosphere until the pressure
gauge reads zero. The operator removes the top insulation, unbolts the cover flange, and collects
the samples. Samples to be aged further are loaded back into the vessel, and the process is
repeated. The aged samples are saved for chemical and physical analysis.
CORBETT ANALYSIS
A description of the traditional Corbett (1969) analysis can be found in the standard
method ASTM 04124. Corbett analysis separates the components of an asphalt according to
polarity. Some modifications of the Corbett procedure were implemented to reduce sample size
and increase efficiency as suggested by Thenoux et al. (1988).
MIXING APPARATUS
To produce the asphalt-rubber binders, asphalts and rubbers were 'cured' or mixed at high
temperatures (> 177"C (350"F). Curing, for the purpose of this paper, is defined as an increase
in viscosity without oxidation, with oxidation being measured by the carbonyl peak area of the
infrared spectrum. The curing process, as carried out in this laboratory, involved mixing at high
temperatures with a 5.1 cm (2") diameter blade driven at variable speeds, 500-1550 rpm by a
variable speed motor. The blends were cured in either 1 quart or I gallon paint can, depending
on sample size, under a nitrogen blanket to prevent the binder from oxidizing.
84
BENDING BEAM RHEOMETER
Low-temperature properties of the asphalt-rubber binder, were evaluated using a bending
beam rheometer (BBR). Anderson et al. (1990) concluded that the BBR is the best instrument for
measuring low-temperature properties of binders. Furthermore, both Set) and the m-value, the
properties obtained by utilizing the BBR, have been correlated with the low-temperature thermal
cracking of binders (Bahia et aL 1992). All bending beam results were obtained at a beam testing
temperature of -15DC (5"F). The beam specimens were produced and the bending beam rheometer
was utilized as specified in AASHTO Designation TPI.
DYNAMIC SHEAR RHEOMETER
The intermediate-temperature rheological properties were tested with a Carri-Med CSL-500
dynamic shear rheometer configured in the parallel plate geometry. This instrument may be
operated in either a constant stress-mode (its natural mode) or a constant-strain mode over a
temperature range from O"C (32"F) to 9O"C (l94"F). This instrument was operated in the constant
stress oscillation mode for analysis of neat asphalt samples but the constant-strain mode was
necessary for analyses of asphalt-rubber samples.
The behavior of asphalt samples is non-Newtonian at intermediate oscillatory frequencies.
However, by utilizing the constant-stress mode, a limiting complex viscosity, TJ:, can usually be
obtained at low frequencies. For highly aged samples the low frequencies are obtained by
utilizing temperatures greater than the reference temperature and the time-temperature
superposition principle (Ferry 1985). For asphalt-rubber samples, however, at low frequencies,
a limiting complex viscosity can not be obtained. To complicate matters further, the strains
induced in the asphalt-rubber binders at low frequencies are quite large and may cause partial
destruction of the bonds' formed between the asphalt and rubber during the curing process.
Therefore, it is necessary to operate the rheometer in the constant-strain mode for asphalt-rubber
samples.
To analyze the asphalt-rubber samples in this study it was necessary to determine the strain
85
level which corresponds to the linear viscoelastic region. Theorectically, the linear viscoelastic
region exists in the strain level range from 0% to some maximum percent strain level. However,
a rheometer cannot accurately measure linear behavior at and slightly above the 0% strain level,
thus narrowing the range of the linear viscoselastic region. In reality the measureable linear
viscoelastic region exists from a stain level range of slightly above 0%, a minimum strain level,
to a maximum percent strain level. This range was determined by specifying several different
strains and observing the strain response wave. Linear viscoelastic behavior is encountered when
the strain response to sinusoidal stress input is also sinusoidal. The strain level for measurement
was chosen to be the minimum strain level at which measureable linear viscoelastic behavior
occurred. This minimum strain level was found to be highly sample dependent and ranged from
approximately 0.5% to 200%, depending upon the temperature.
An additional complication to the measurement of asphalt-rubber properties is the presence
of the rubber particles. As a result, it was necessary to determine the gap width for the parallel
plate geometry. This gap width was found to be strictly a function of the rubber particle size and
rubber content. The gap width for a given rubber size and content was determined by measuring
the rheological properties of a given asphalt-rubber at multiple gap settings. To insure the
elimination of the 'gap effect', the gap width was chosen such that the rheological properties taken
at as wide or wider gap widths, were independent of the gap width.
BROOKFIELD ROTATIONAL VISCOMETER
A Brookfield rotational viscometer Model RVF 7 was used to obtain the high-temperature
(> 121"C) (250"F) viscosities of the asphalt-rubber binders. Torque is applied to spindle placed
in the binder sample which is contained in a thermostatically controlled beaker. The relative
resistance to rotation is measured for a given rotational speed. The relative resistance, the spindle
size, an the rotational speed are then used to calculate the viscosity, ".
86
GEL PERMEATION CHROMATOGRAPHY (GPC)
GPC analyses were performed using a Waters 712 sample processor and a Waters 600E
multisolvent delivery system. Helium-sparged HPLC grade tetrahydrofuran (THF) at a flow rate
of 1 mLimin was used as the carrier solvent to efficiently separate the asphalt-rubber binders.
Three columns with pore sizes of 1000A, 100A, and 50A were connected in series. The lO00A
and lOOA columns are 30.5 cm (l foot) in length and are packed with ultrastyragel particles. The
50A column is 61.0 cm (2 feet) in length and is packed with PLgel particles. A Waters 410
Differential Refractometer and a Visctoek HS02 Viscometer was used to monitor sample elution.
The column and detector temperatures were controlled at 40·C (104"F). Samples were prepared
by dissolving 0.20 to 0.25 grams, depending upon the rubber content, in 10 mL of THF and
filtering through a PTFE syringe fllter with a membrane pore size of 0.45 p.m (0.45 micron).
Thus, sample preparation removes all rubber particles greater than 0.45 microns, since asphalt is
soluble in THF and rubber is not.
FOURIER TRANSFORM INFRARED SPECTROSCOPY (FTIR)
A Mattson Galaxy series 5020 Spectrometer at 4 cm·1 resolution and 64 scans is used to
measure the infrared absorbance spectra of asphalt samples. In particular, the Attenuated Total
Reflectance, ATR, method with a Zinc Selenide prism is used (Jemison et aI., 1992). To quantify
the changes in the spectra, the carbonyl content is defined as the integrated absorbance from 1820
to 1650 cm-I with respect to the baseline at the absorbance of 1820 cm-I. This area is called the
Carbonyl Area or C4. The range of wave numbers includes the following carbonyl compounds:
esters, ketones, aldehydes, and carboxylic acids. The primary absorbance peak for the oxidized
asphalt is located at 1700 cm-I and corresponds to ketone formation. The carbonyl area has been
shown to be a good measure of oxidation (Liu et aI. 1995).
At low aging pressures of2 and 0.2 atm oxygen and for thick (-=Imm) fllms, oxygen
diffusion may be significant. To partially eliminate this diffusion problem, only the exposed
surface, ES, of the film is analyzed for kinetic data. For analysis, a quarter of the material in the
87
aluminum tray is removed and the ES placed on the prism face. For samples that have been aged
in thinner films, diffusion is probably not significant, so it is possible to measure the spectra of
a stirred sample. To insure good contact at the sample/prism interface, the sample is compressed.
Heating of the sample is avoided, if possible.
For measuring the spectrum of asphaltenes, the material is dissolved in THF and the
solution deposited on the A TR prism drop by drop allowing the THF to evaporate. When the film
on the prism is sufficiently thick it is further dried with a heat gun.
MICRODUCTILITY MEASUREMENTS
A detailed description of the microductility measurements can be found in Chapter 2.
, ,
c 88
APPENDIX lB
OIT SPREADSHEETS
89
Economic Analysis for a
ROSE Supercritical Unit for Producing
Aromatic Material for eRMA
90
'" .....
Projecf Name:
Filename:
Anlll/ysf:
olr Project Benefit Analysis Worksheet Version 2 . .,
Development of Asphalts 3. Pavements Using Recycled Tire Rubber
1trr95_7.XIS
ICharles Glover I This syst.m of apraadaheet. was developed 10 old In "'" 1InmncIal, morkot, ond bonoll! """Iyo'" of orr proJocts. Tho u •• r '" .sked for. number of Input. relatlog 10 both the new and """"log technolagles for. """"In proJocI, auch .. : capIIol coats, onnual coals, onorgy us., wam. reduction. oqulpmonlllratlma, discount rate and merkollnformatlon. The system will calculatotho following proJoci "''''latlc9: InRIaI capftallnves1ment. Ialol onnual coals, proJoci rmmnclollnformollon, mmrkol pm_Ion Infor",allon, not """'lIY BIIIIIr1gs and not Wlllolo reduction. Tho spread,heel, ore doolgned In ouch a way thol "'" uaor can change lnfor",,,,11on In "'" cello wIlh _1m only. The cello wIlh red Ie'" contain formulas and are locked.
Industrial Partner: Tellaa Deplllrtment of Trlllnsportation
ROSE unit for producing recycllnglllgenVauperlor IIIsphalt. The Installed cost of III ROSE unit I" $30 million for a 10,000 bblfday recycling agent produced. At 330 operating dmymlyr, this Is $9.09/bbl of RA (capital Investment). The annualized coat of producing the recycling agent Is $2.381bbl of RA (eKcludlng energy costa) leslI "n allllllumed sales price of $11.251bbl. The aneflllY savings Is sloss due 10 the anergy cost of processing the RA. This Is cslculatmd eccordlnglo 100,000 Btu/bbl of faed and 330 operating dayBlyr for the 30,000 bblfday of feed unit which produces 10,000 bblfday of RA. The energy cost la antared In the anergy &lIvlng8 work .. h .... t These assumptions plullla10% discount rata glvaan Intarnal rata of rmturn of 28% and a dlscountad payb .. ck period of 4.4 y .... ra. Ths en .. rgy u ... for this pracellllllis 990,000 million Btu/yr (100,000 Btu/Bbl of faed, 30,000 Bbl of
faed/day, and 330 opsrmtlng dayelyrl·
5130196
\0 N
1 2 3 4 5 e 7 8
9
10
SheelA
Capital Investment Worlcsheet
Capital Cost Component
FInIt Cost of Equipment SR. Prepormllon and Englneorlng Installation Contingency Allowance Construction indirect Coste Interest During ConstrucIIon Start-up Expen ..... ~orkIng C.pIlmI Qther.,
TOTAL: Initial Capital Investment
Coste ehould be onIorod In 1994 cIo!mJ.
5130196
DlIII8lopment of Asphafls & Pavemenllo Using Recycled TIre Rubber
ConvenUonal New Unit Incremental Incremental Net Cost Unit Capital Costs Savings Increment
S - S -. - $ - $ -S - $ - $ - $ - $ -$ - $ - s - $ - $ -S - . - S - S - $ -$ - s - $ - $ - $ -S -. - S - $ - $ -• -. - $ - $ - $ -• - S - $ - $ - $ -S - S
OIT Project Benefit Analysis Worksheet Version 2.1
Development of Asphalts 3. Pavements Using Recycled Tire Rubber
ITRf95_3.xIS
ICharies Glover I This systom 01 spreodshoels was dovotoped 10 old In the nnsnclal, marlo!!, and bonefilonBlysls 01 OIT projects. Tho user Is asked for s number of Inputs relating to boIh the new and mdsIfng technologies for. certain proJod, such .s: capRal costs, .nnual costs, anorgy USB, waste reduction. oqulpmont 1Ifotlme, discount l1IIa mnd morlo!!lnI'ormalion. Tho systam \'An calculalathe foIto\'Ang proJmct statlsllca: Inftlsl capftBllnwstment, totalonnual costs, proJmct Onanclallnflll,,_, _ plliObatloillnformallon, nat anergy sol.!ngs and nat wasta roductlon. The spreadsheets .'" dasIgnod In ouch. waytha! the user can chongo hirom_11n the caIIo __ tmt only. Tho caIIo _ rod telt contsln lonnulas snd .", locked.
Industrial Partner: Tellms Department of Transportallon
Comport ..... bet ... on .,. con_I pavement plac&nentlhavfng an ........ ed"r.. 01 12 yo ..... ,_ b,. ouperlor a""haHlCRM pavement pI ....... ent Iv.fth an ........ ed or.. of 11 yeara'. AI ... , II I ......... ed thollha annuo' coolin both ca .... equo'. the Iioial <splta' cool 01 Iha conventional pavement)llyra _.114 and thollha CRM cententln tho binder ,. 10% and dI""loc •• an equlv.lent .... ounI of aophaH binder, thereby ",mulling In both Wllole reduction and energy savlnglL Ho ... v .... It ,. osoumed Ihol Iha pr_ng coo! wm offoot Iha b'nder energy oavfnglL For th' •• nary .... 10 mesh crumb rub ..... '. asoumed, 01. coot 01 $O.1411bm. A'ao, e CRMA binder procasalng cool of $2&IIon of binder ,. aeaumed lor blending and curing the CRM and .""han and. price 01 $103.l!IIon Iplu.the procesalng cool". placed on Iha .""holt molerla', .nowng lor 1110 be. designed molerlal of 00% .""".11101 $IOD/Ion,ond 10%"",,,,0110 malerla'ial $13&/Ion~ Tho required blnderlo"""all plu. rub ..... ' 'n the pavanent'. taken 10 be 8% 1IIIHh th .. " aoaumptlons pluo 0 10% discount not. ,Iha Internal not. of return ,. 14.0% and the dIscounted
paybeck period ,. D.3 ye .......
5130196
.... o w
2 3
4 5 6 7 6
9
10
SheelA
Capital Investment Worksheet
Capital Cost Component
First Cost of Equipment Sfte Preparation and Englnoerlng InstaRmllon Contingency AIIowanca Construction indirect Costa Interest During Construction S1art-up Expon ..... Working CopRal other.
TOTAL: Initial Capltallnvesiment
Cost. should .... ontored In 1994 doIIare •
Development of AsphaRs & Pavemant. Using Recycled Tire Rubber
Conventional New Unit Incremental Unit Capital Costs
OIT Project Benefit Analysis Worksheet Version 2.1
Development of Asphalts & Pavements Using Recycled nre Rubber
ITRf95_ 4.xls
ICharles Glover I Thlo ayaIom of opreadsoom. _ d .... 1oped 10 mid In the IInImcIoI, ma"".I, ond benall! .""!yolo of OfT p!UJocts. The user Is .sked for a number of
Inpuls roIoIIng t. both the new and exlotlng loChnoIogIae for a certain p!UJee!, much as: ""plio! costs, annuel costo, energy u .. , wasle reduction,
equipment lifetime, dIocount rate and merltGt Infonnatlon. The ayaIom will cafcuIotethe following p!UJee! sI_: InRIaI capftallnveslment.
lotal.nnuol costs, proJee!linanclollnfonnellon, markllt paIftItJ"lI.iI Infonnellon, not anergy oavIngtI and not WIlsie reduction. The spreadsheets 0 .. designed In 8UCh a WfIY _the ....... ClIO ""-Moo",.II.h In the ___ taxi only. The __ nod text conIaln fonnulas and
0 .. !ockad.
Indumtrlml Partner: Tel!!!e Department of Trmnllportntlon
Comp .... son between ., • conlllHlllonal pm""""",t pI .... ment (having an assumed me of 12 Y" ..... , WIth b, •• uperlor asph.,tlCRM pa""""",t placement (wtIII an ossumed life of 18 Y"I!I'S'. Also, It I ••• sumed that the annual cost In both cases equals the (Iolal capital cost of the conlllHlllonal pmvemenl)l(yrs """"")14 ond that the CRM conlent In the bind ... I. 10% and displaces an equivalent amount of .sphalt bind ... , thEreby resulUngln both wasta reducHon and enorgy .... vlng.. HoweVEr, It Is assumed that the processing cost will _I the bind ... enorgy .. Vlngs. Far this analysis, 10 mesh crumb rubbor I. a.sumed, at • cost of SO.1411bm. Also, a CRMA binder .....,., ••• ,ng cost of $Z5Iton of binder I. assumed far blending and clBlng the CRM and asphalt .nd • price of $103.5Iton (plus the processing cost, I. placed on the asphalt ma_', .noWlng far It to be • designed m.ter1al 01 90% a.phalt (al StOOIton, and 10% .... mallc mater1al (at $1351ton,. The requlrod binder (a'phalt plus rubbor, In the pavement Is taken to ba 8%. WIth \hess assumpHon. plu. a 10% discount rata, the Internsl reta of retum I. 29.0% and a discounted payback
period of 4.4 yes •.
5/30/96
..... ..... ..,.
1 2 3 4
5 8 7 a 9
10
Shonl A
Capital Investment Worlcsheet
Capital Cost Component
FIrot Cost of Equipment SMa Praparmllon and Engln80flng Installation ContIng&ncy AIIowanca Construction IndImct Costs Il1Ierest During Construction Start·up Expenses Working Capftsl
Other:
TOTAL: Initial Capital Investment
Costs ""auld be entorad In 1994 doIIsrs •
Development of Asphafts & Pavements Using Recycled Tire Rubber
Conventional New Unit Incremental Unit Capital Costs
Annual Cost Component Payroll plus labor Indlrects Opemllng Supplies Mmlntenrmce Supplies Tnonspor1atlon PolluUon Control and Waste Disposal other Costs/Credits
TOTAL: Annual (non-energy) Costs
Costs should be entered In 1994 dollars.
Conventional Unit
• S - $ • - S
$ 4,770 $
$ 4,770 $
5/30/96
Development of Asphalts & Pavements UsIng Recycled Tire Rubb
New Unit Incremental Incremental Net Cost Annual Costs Savings Increment
ODT Project Benefit Analysis Worksheet Version 2.1
Development of Asphalts 3. Pavements Using Recycled Tire Rubber
ITRf9S_S.Xls
ICharkls Glover I This ayslom of ep<eedsheeta WIllI developed to aid In IIIe I'InsncIIII, markof, end beneIII analysla of OIT pmJocIs. Th. usar Is .sked for. number of Inputs rMItIng to both the new end 8ldoIIng technclogloa fer" certain "",Joct, such am: c:apbl costs, annual costs. onorgy U98, wast. reduction, equipment lIfollme, dIscounIl8Ie and marItet Inflll1rlOlhJlt. '"'" aysIem will cmlculote IIIe folluwlng pmJoct _1otIcs: InRIaI capRallnvaslment. total an""'" costs, ptJIjoct IInIIrtcIaIInfClllllii1ltHt, markof "",""'mllott 1nfllfllU!lllon, IitIt onorgy aovfngs and IitIt wasto reduction. Tho spreadsheets ano d ... lgned In ouch a waylltslthe _ CIIh chImga ..v",,,tIIIIIott In the ..... wIIIt green t8lII only. '""' ..... wIIh red toxt cont.1n formulas and oro locked.
Industrial Partner. Tellas Depl!lrirnent of Transportation
Compmrlson between a) • convenllonal pavement placement (having an assumed IIfa or 12 Y"ar.) with h) • supmrlor asphaltlCRM pavemtHtt placement (with an a.sumed IIfa or 21 years). Also, It Is assumed that tha annual cost In both cases equals the (Iotal capital cost or the con\llllillonal pavemtHtt)l(yrs ..vIce)l4 snd that tha CRM content In tha bind ... I. 10% and displaces an equivalent lIiiiOunt or asphalt bind .... th .... by ..... uHlno In both wasta nducllon and energy savlnos. Howev ... , It Is assumed Utat the pnJCf!ss\ng cost will offset the bind ... energy savings. Far thl. analysiS, 10 mesh crumb ntbber Is •• sumed, .ta cost 01 $0.1411hm. Also. II CRMA bind ... pnJCf!sslno cost or $2!11ton or binder ... assumed for blending and curlno the CRM and asphalt and a prlce or S103.!IIton (plus the processing cost) Is placed on the asphalt rna""'.I, .nowlno for It to be a designed material of 80% a.phalt (at S1001tonland 10% ......... Oc ma""'.1 (at $13!11ton). '"'" required binder (msphalt plus ntbber) In the pavement Is laken to be G%. WHIt these assumpUons plus a 10% discount rata,the Intern.1 rate or .... turn I. 38.0% .nd a discounted payback
period or 3.2 Y"l!!NI.
5/30/96
..... N V>
2 3 4 5 8 7 8
9
10
ShoelA
Capital Investment Worksheet
Capital Cost Component
FlmI Cost of Equipment SHe Preparation and EngInaaring I"uI.1lsl1on Centlngancy AlIowanca Construction Ind~ect Costs Int"""" During ConstrucIIon Start-up Expanses wortdng CopR11
other:
TOTAL: Initial Capltallnveslment
, costa ohould be antGI8d In 1994 cIoIIIaN •
Development of Asphalts II Povamants Using Recycled Tire Rubber
Conventional New Unit Incremental Unit Capital Costs