Movement of Oil Under Sea Ice - RESTCo · THE MOVEMENT OF OIL UNDER SEA ICE I ... SURFACE TENSION 2 2.1 General 2 ... ture using the du NoUy ring detachment technique.

Movement of Oil Under Sea Ice L. w. ROSENEGGER Technical Report No. 28

THE MOVEMENT OF OIL UNDER SEA I CE I

L.W. Rosenegger .

Production Research and Technical Research Laboratory

Imperial Oil Limited Cal gary, Al berta

Beaufort Sea Technical Report #28

Beaufort Sea Project Dept. of the Environment

512 Federal Building 1230 Government St.

Victoria, B.C. V8W lY4

December) 1975

TABLE OF CONTENTS Page ABSTRACT 1

l . I NTRODUCTI ON 1 . 1 Objecti ves 1 1 . 2 Outl i ne of Present Work 1

2 . SURFAC E TENS I ON 2 2 . 1 General 2 2 . 2 Ana l yt i ca l Deve l o pment 2 2 . 3 Exper imental Procedure 4

2 . 3 . 1 Ses s i l e Drop Method 4 2 . 3 . 2 Ri ng Detachment Method 6

2 . 4 Res u l ts 7 2 . 4 . 1 Ses s i l e Drop Method 7 2 . 4 . 2 Ri ng Detachment Method 9

2 . 5 Concl us i ons 1 0 3 . MOVEM ENT OF O I L DROPS 1 1

3 . 1 Genera l 1 1 3 . 2 Ana l yt i ca l Devel opment 1 1 3 . 3 Experi mental Procedure 1 2 3 . 4 Resu l ts 1 3 3 . 5 Conc l u s i ons 1 3

4 . O I L MOVEMENT I N A L EAD 1 3 4 . 1 General 1 3 4 . 2 Spread i ng Ana l ys i s 1 4 4 . 3 Concl u s i ons 1 5

5 . SOLUTE RED I STRI BUTION AND O I L PENETRAT I ON 1 5 5 . 1 Sol ute Red i stri buti o n 1 5

5 . 1 . 1 General 1 5 5 . 1 . 2 Methods and Resu l ts 1 6 5 . 1 . 3 Conc l u s i ons 1 7

5 . 2 O i l Penetrat ion 1 7

6. CONCLUSI ONS 1 8

7 . REFERENCES 20

8 . B I BLI OGRAPHY 23

TABLES 1 to 6 i nc l u s i ve 24 to 28

F IGURES 1 to 39 i nc l u s i ve 29 to 65

APPEND I X A - DEV ELOPME NT OF THE EQUATI ON FOR THE PROFILE O F A SESS I LE DROP

APPEND I X B - TEST RESULTS FROM R I NG DETACHMENT METHOD APPEND I X C - DATA FOR CALCULAT I NG THE SPREAD ING COE FF IC I ENTS APPENDI X D - TEST METHOD USED IN DETERM I N I NG SALT CONTENT OF

CRUDE O I LS

P�ge 66 69 75

79

1

ABSTRACT Thi s report presents the res ults of laboratory tests to determi ne the i nter­fac i al tens i on and mot i o n of crude o i l bubbles u nder sea i ce . Two d i fferent crude o i ls were used i n these exper iments ( Swan H ills and Norman Wells ) . An assessment ha s also been made of; a ) the ab ili ty of o i l to penetrate sea i ce from beneath , b ) the equ i li bri um thi c knes s of a crude o i l f ilm on water u nder arcti c cond i t i ons , a nd c ) the red i stributi on of solutes i n the o il .

1 . I NTRODUCTION Thi s report presents the res ults of a laboratory i nvesti gat ion i nto certa i n a s pects of the beha v i o ur of o i l under i ce . The topi cs stud i ed were recommended by the Frozen Sea Research Group , Ocean and Aquat i c Sc i ences , Environment Canada , a s part of the Bea u fort Sea Project ( O i l i n I ce Stud i es ) . Fund i ng for thi s s tudy wa s prov i ded jo i ntly by the Frozen Sea Research Group ( DSS Contract OSZ4- 0344 ) and by Imperi al Oi l L imi ted . 1 . 1 Objecti ves

The major object i ves of thi s study can be s ummari zed as follows : 1 . To determi ne the i nterfac i al tens i on between o i l and water at

the temperature of freezi ng water for Norman Wells and Swan H i lls crude o i ls by the sess i le drop method .

2 . To study the movement and/or absorpt ion o f sess i le o i l drops at an i nterface between s ea i ce and water i n res ponse to grav i tat i o nal and drag forces prod uced by i ce sheet t i lt .

3 . To determi ne the movement o f an o i l f ilm i n a lead i n respo nse to a conti nuous o il i nput at a g i ven poi nt i n the lead .

4 . To determi ne whether o i l w i ll penetrate from beneath i nto a growi ng s ea i ce sheet due to buoyancy force s and to a s sess the effects o f the red i s tr ibut i o n o f solutes i n the o i l o n the i ce sheet .

The work outli ned above consti tutes a small port i on o f the i nput i nformat ion necessary for a better u nderstand i ng of the effects and ulti mate d i s pos i t i o n of arc t i c o ffshore o i l s p i lls . I ts ma i n p ur­pose i s to serve as u seful i nput i nformati o n to the Beaufort Sea Project cover i ng the ' Behav i our of O il i n an I ce-Co vered Area ' .

1 . 2 Outli ne of Pre sent Work In each Chapter , a bri ef d i scuss i on of prev i o u s work and of the s i g­n i fi cance of each study objecti ve i s presented . Experi�enta1 a nd a nalyt i cal procedures u sed i n thi s study are then described , followed by a presentat i o n of the results obta i ned . The s i gn i fi cance

2

of the resu l ts is discussed and concl u sions ba sed on the resu l ts are presented. A final chapter s ummarizing the overa l l res u l ts and concl usio ns of this wo rk is a l so presented.

2. SURFACE TENS ION 2.1 General

The s urface tension pa rameter is used extensivel y in cal cu l ations concerning the ris e and breakup of buoyant p l umes and in determin­ing whether one l iquid wil l spread on anot her . Bo th of t hese situations co u l d be encoun tered in the event of an accidental rel ease of oil in an arctic environment . The first wo ul d be t he ca se of a bl owout at the sea bottom , whil e the second wou l d pertain to the spread of oil under ice or on the water surface . For the bl owout case , in which a buoya nt gas and oil pl ume rises through a water col umn , one is interested in determining (Topham , 1 976); a ) whether the gas jet wil l penetrate the s urface , b ) t he fl uid vel ocity distribu tio n , c ) the entrained fl ow , d ) the inter­actio n with s urface cu rrents , and , e ) the behavio ur of the oil in the rising p l ume. Since there are two immiscibl e fl uids ( oil and wa ter ) present , one wou l d expect the surface tension ( pos sibly in the form of a Weber number ) to be an important pa rameter invo l ved in the so l ution of each of the above mentioned points ( Hinze , 1 955; Chris tia nsen and Hixson , 1 957 ) . The surface tensio n parameter wil l l ikewise p l ay an importa nt rol e in t he determination of whether and how extensivel y oil wil l spread at an interface . The current l iterature contains many exampl es of such appl icatio ns ( eg. Fay , 1 969; Chen et a l , 1 974; Gl easer and Vance , 1 97 1 ; Keevil and Ramseier , 1 975; and Ga rrett , 1 973 , to mention on l y a few ) . Our primary aim in this part of the work wa s to determine the inter­facia l tensio n between the two crude oil s ( Norma n Wel l s and Swa n Hil l s ) and brine at the interface between ice and brine u sing the sessil e drop method . An attempt wa s a l so made to asses s the effects of aging ( if any ) on the s urface tension . Severa l other parameters ( eg . eq uil ibrium bubbl e thic kness and diameter , contact angl e , etc . ) were a l so measured in the co urse of this wo rk .

2 . 2 Ana l ytica l Devel opment Surfaces can be cl as sified according to the physical s tate of the matter sepa rated by them . Thus , one is rea l l y dea l ing with inter­faces between l iquid-gas , l iquid- l iquid , so l id-gas , so l id- l iquid , and so l id-so l id surfaces . When one norma l l y ta l ks of the s urface ten sion of a s ubstance , one is actual l y referring to the interfacia l tension between two s ub­stances . I nterfacia l tension acts a l ong the interface and tends to minimize the interfacia l area . The concept of interfacial ( s urface ) ten sion can be devel oped from mechanis tic or energy consi­derations . Both yie l d t he same dimen sions , expressed as either force

3

per length , or energy per area . References c i ted i n the b iblio­grapy g i ve a detailed d i s c u s s i o n o f these v i ewpo i nts . The earli es t comprehens i ve work o n determi n i ng i nterfac i al ten s i ons u s i ng the ses s i le drop method wa s by Bashford & Adams ( B&A ) , 1 883 . B&A stud i ed the form of ses s i le l i q u i d drops (mercury ) and were able to rela te the coord i na tes ( x ,y ) of points P ( x ,y ) o n a mer i d i onal profi le of the drop to the fi rst and second deri vati ves y l = dy/ dx and y" = d 2y/dx2 and to a constant parameter , S , g i ven by

= gpb2 S Y . . . ( 1 )

where 9 i s gravi tati o nal accelera t ion p i s the dens i ty d i fference of the flu i d s y i s the i nterfac i al tens i o n between the

flu i d s b i s the rad i u s of curvature at the ori g i n

o f the coord i nate sys tem The above are related by the equa t i on [9J

dx dX] xdx dx . . . ( 2 ) � + {l + (� 2}.<!r = ( 2 + Sy ) {l +(.<!r)2}3/ 2

where x , y , are defi ned as � and t respecti vely , x and Y bei ng the actual phys i cally mea s urable d i mens i ons of the drop . The above equation i s appl i c able to drops res t i ng o n top of hori ­zontal surfaces and also to the case o f hang i ng drops as long a s the dens i ty d i fferences are ta ken as shown i n the mathemati cal development of the above equati on such that S i s po s i t i ve ( s ee Append i x A ) . B&A prepared a s eri es of tables so tha t S could be determi ned from the geometry of the s e s s i le drop . A second set of tables was u sed to obta i n values of b whi ch were then u sed to calculate y from equat ion ( 1 ) . More recently , Sta i copolus ( 1 962 , 1 963 , 1 967 ) and Parvati kar ( 1 966 , 1 967 ) have ver i f i ed and extended the B&A results to cover a larger range of S values . For the pres ent study , the emp i r i cal equat i ons expli c i tly relati ng the i nterfac i al ten s i on , y, to the exper imentally obta i nable quant i t i es x and Y as pres ented by Stai copolus ( 1 962 ) have been u sed . These are:

and

y _ � (x)", - 2 'I' Bl cp

2 . . . ( 3 )

. . . ( 4 )

4

where B¢ = S. F¢ = (x/b ) ¢ and G¢ = (Y/b )¢ ar� obta i ned i n terms of fourth order polynomi als of t he quant i ty l = (X!Y) p - Ap

C¢ as follows : B¢ = eXP{[PB (l )]¢} - D · . . (5 ) . ¢ F¢ = [PF (l]¢ · . . ( 6 )

and G¢ = [PG (l]¢ · . . ( 7 )

Table I. ta ken from Sta i copolu s ( 1 96 2) gi ves the values of the coeffici ents of the powers of l for ¢ = 45° and ¢ = 90° toget her w ith the correspond i ng constants A¢. C¢ and D¢ . Mea s urements of x and Y on the maxi mum peri phery of a s es s i le drop (¢ = 90° ) w i ll therefore yi eld valu es of the i nterfaci al ten s i on between o i l and water . I n h i nd s i ght. i t may be s tated that the ses s i le drop method of analys i s wa s perhaps not the best method to u se i n th i s study s i nce mea surement errors can have rela ti vely large effects on the res ults ( th i s i s d i scus sed i n the next secti on ) . As a check o n the work and in order to obta i n some add i t i onal i nformati on not ava i lable from t he ses s i le drop method , several tests were run at room tempera­ture us i ng the d u NoUy r i ng detachment techn i que . A deri vat ion of thi s method can be found i n Freud and Freud ( 1930 ); Harki ns and Jordan ( 1 930 ); and Fox and Chr i sman ( 1 95 2 ) . and a general expla na­t i on i n any of t he s urface chemi stry boo ks li sted i n t he b ibli ogra phy .

2. 3 Experimental Proced ure 2 . 3 . 1 Ses s i le Drop Method

The general procedure u sed i n t h i s study to determi ne the surface ten s i on by the ses s i le drop method can be s ummari zed a s follows : 1 . A bri ne solu ti on ( 1 2% 0 sali n i ty for thi s experimen t )

wa s prepared and allowed to freeze i n the cold room i n clear plexi glas s tan ks wh i ch were i ns ula ted on all s i des except t he top . A hea t i ng tape wa s placed d irectly below the ta nk so tha t a temperature grad i ent could be ma i n­tai ned i n t he wa ter .

2. The tank wa s removed from t he cold room o nce a n i ce s heet of approxi mately 5 cm t h i c kness had grown . A hole was drilled through the i ce and o i l was i njected u nder t he i ce wi th a clea n stai nless steel syri nge . Thi s u s ually res ulted i n several bubbles of vari ous s i zes a s i t was very d i ffi cult to control t he i nject i on process .

5

3 . The i ns ulat i o n was removed from one s i de of the tank and photographs were ta ken of the o i l bubbles . Duri ng thi s t ime� i nsula t i o n was placed o n top o f the tank so as to reduce heat losses and to keep the temperature as u n i form as po s s i ble .

4 . P i ctures were taken wi th a 200 mm fi xed-focus lens ( a t var i o u s reproduct ion rati os ) . Once develo ped� the f ilm was put i nto sli d e mounts . These were then projected on an opti cal comparator and meas urements were ta ken d i rectly from the sli des .

5 . Mea s urements were ta ken o f the d i ameter a t the meri d i onal plane� 2 X� of the he i ght from the apex to the mer i d i o nal plane , Y , of the hei ght from the apex to the i nterface between o i l and i ce , and of the contact angles a and B meas ured through the water phase . Average values of a

and B were used for calcula t i o n purposes . These para­meters are shown schemat i cally i n Fig ure 1 .

6 . The i nterfac i al tens i on wa s then calculated from the mea surement data accord i ng to the method descri bed i n the prev i o u s s ect i on .

The camera wa s mou nted on a cathetometer s tand to whi ch a spec i al brac ket had been added to allow for full 3-axi s po s i­t i on i ng o f the camera . Thi s wa s part i cularly u seful for rap i dly focu s i ng on a part i c ular bubble . Fi gures 2 and 3 show the camera and tan k i n pos i t i on ready for tes t i ng . A str i ng of 2 1 therm i s tors s paced 0 . 5 i nches apart was also constructed . I t wa s u s ually frozen i nto one o f the tanks , and when a test wa s run o n the parti c ular ta n k conta in i ng the thermi stors� read i ngs were ta ken both before and after the test to determi ne whether a s i gn i fi cant temperature change had ta ken place duri ng a test . Shown i n F i g ure 4 are the temperatures a s mea sured both before and after a test on Apr i l 1 1 , 1 975 . Even though the temperature of the top layer of i ce ha s warmed u p con s i derably duri ng the test� i t may be not i ced that both above and below the i nterface between the i c e and water the temperatures before and after the test d i ffer by at most 1 °F . The freez i ng temperature for the 1 2% 0 bri ne solut i on can be s een from the graph to li e between 29 . 1 and 2 9 . 6°F . Table 2 g i ves meas ured values and calculated s urface tens i ons for the two bubbles shown i n F i gures 5 and 6. These p i c tures were ta ken at 70 second s and 2 5 . 5 m i nu tes a fter i nject i o n respecti vely . The d i fferences i n the i nterfac i al tens i on s calculated for the bubble at 70 s econds c a n b e d u e t o errors i n x and y of ±0 . 001 i nch or les s . Thi s i s ver i fi ed by ta k i ng the above values and recalculat i ng for y us i ng x � 0 . 2896 i nch and y = 0 . 1 907 i nch. U s i ng thes e values i n equati ons ( 3 ) and

6

( 4 ) res ults i n values for y of 27 . 952 and 27 . 972 dyne/cm res pect i vely . Although accura c i es of ±0 . 001 i nch ( ±0 . 0025 mm ) can be obta i ned wi th the opti cal compara tor , the cri ti cal factor i n tak i ng mea surements i s the opera tor's abi l i ty to judge exactly when the cros sha i r on the comparator s creen i s ali gned w i th the i nterface . I t wa s fo und that i n pract i ce , an u ncer­ta i nty of up to 0 . 001 i nch could occur dependi ng on the sharp­ness of the p i cture . Such errors are then mult i pli ed by the i nverse of the reproduct i o n rat i o . Thu s , as shown i n Table 2 , the s urface ten s i ons calculated by u s i ng equat i ons ( 3 ) and ( 4 ) and the meas urement data for the f irst bubble ( 70 sec . , 1 / 2 x reproduct i on rat i o ) are i n error relati ve to the lower value of 1 1 . 5% whi le the same calculat i ons for the s econd bubble ( 25 . 5 m i n . , 1 x reproduc t i o n rat i o ) are w i thi n 0 . 1 % of each other . The i nab i l i ty to grow a truly flat i ce sheet cau sed d i ffi cult i es i n thi s pha se of the program . Flat i ce sheets could have been achi eved by grow i ng i ce and s ubsequent­ly melt i ng the bottom of the sheet u nt i l a flat surface was obta i ned . Thi s wa s not done , s i nce i t was des i rable to keep the structure of the unders i de of the i ce essenti a l ly the same as would be found i n nature . As a result , o i l bubbles u s ually settled i n a hollow maki ng i t d i ffi cult at t imes to make out the po s i t i o n of the i nterface between the i ce and o i l . Suffi c i ent bubbles were photogra phed , however , s o tha t a suffi c i ent number of good bubbles st ill rema i ned for mea surement purposes . A more annoyi ng problem i n thi s experi­ment wa s the li p of i ce that formed aro und the tank at the i nterface between the i ce and water . Thi s wa s u sually 0 . 5 cm i n depth and i n mo st i n stances had to be melted by playi ng a heat gun on the tank along i ts length . The bo nd between the i c e and plex i gla ss on that parti cular s i de wa s u s ually broken as a res ult and i f bubbles s ubsequently rolled over to that s i de , they wo uld sprea d up the i nterface due to cap i llary act i on and the presence of a i r .

2. 3 . 2 Ri ng Detachment Method Surface tens i ons were measured u s i ng the ri ng detachment method at a temperature of from 25 . 8 to 28 . 2°C . A F i sher Tens i omat model 21 was used for these tests . Fo ur or fi ve tests were run on each flu i d and the apparatus was thoroughly clea ned between each test ( s ee A . S . T . M . , 1970 , for cleani ng proce­dures u sed between o i l samples ) . The apparatus was cal i brated both before and after the tes ts accord i ng to the procedure recommended i n the users manual suppli ed . F i gure 7 shows the average cal ibra t i o n curve u sed i n thi s work . When u s i ng th i s i nstrument , the scale value i nd i cates only a n apparent s urface tens i on whi ch mu st be corrected for ri ng d i mens i on s and den­s i ty d i fference between the upper and lower phase bei ng tested . Thi s correct ion factor i s shown i n F i g ure 8 bas ed o n the d i men­s i ons of the ri ng u sed . The upper phase i n all the tests was a i r and Table 3 g i ves the dens i ti es and dens i ty d i fferences

2 . 4 Res ults

7 /

between the flu i d s and a i r . The s pec i fi c gravi ti es of the d i fferent flu i ds were mea s ured separately and the res ults are presented i n Fi gure 9 and 10 . S i nce the spec i f i c gravity of the 6% bri ne solut ion had not been determi ned , i ts value was i nterpolated as s hown i n Fi gure 1 1 . F i gure 1 2 i s bas ed on data g i ven by Kre ith ( 1 968 ) and wa s u sed to obta i n a value for the den s i ty of a i r .

2 . 4 . 1 Sess ile Drop Method A total of 200 o i l bubbles under an i ce sheet were meas ured i n thi s experiment ( 1 1 1 with the Swan H ills crude and 89 wi th the Norma n Wells crude ) . I nterfac i al tens i o n s were calculated accord i ng to the method descri bed i n Sect ion 2 . 2 . From these calc ulati ons , it wa s fou nd that 1 9 of the Swan H i lls and 7 of the Norman Wells i nterfac i al tens i o n s fell outs i de the range 5 to 50 dynes per cm . A n umber of these-mea surements were rec hec ked and i t wa s fou nd that the photographs were out of foc us or a low reproduct i o n rat i o ha d been used . Thi s data was then d i scarded before further calc ulati ons were made and althoug h the l im its i mpos ed may s eem somewhat arb itrary , i nter­fac i al tens i ons less than 5 and greater tha n 50 dynes/cm are cons i dered u nli kely for o i l/wa ter systems (T immons and Zi sman , 1 968 ) . Table 4 g i ves a s ummary of the experi mental res ults obta i ned showi ng the means and the ir s tandard devi ati ons . H i stograms are g i ven i n F i g ures 1 3 and 1 4 for the Swan H i lls and Norman Wells results respecti vely . As may be noted , the standard dev i ations are q u i te h i gh as are the mean dev i ati ons . Th i s led to some doubt as to the vali d i ty o f the results and to the sta tement i n Sect i o n 2 . 2 concerni ng more appropri ate mea s ure­ment techn i ques . These dev i at ions are of li ttle i mportance , however , and thei r causes and effects wi ll be expla i ned i n the d i scuss i o n that follows . The fi rst step i n the analysi s o f these results was to deter­mi ne the effects of vari at i ons i n the phys i cal d i mens ions on the surface ten s i ons . S hown i n F i g ure 1 5 and 1 6 are the res ults obta i ned for the i nterfa c i al tens i on us i ng equati ons 3 ( GAMMA X ) and 4 ( GAMMA Y ) plotted as a funct i on of X . Di fferences i n values between GAMMA X and Y are almost i nd i s­ti ng u i shable o n these graphs wh i ch i s rea sonable s i nce equati ons 3 and 4 y i eld the same results . Tha t they do , can be seen i n Fi gure 17 and 1 8 . More d i ffi cult to expla i n i s the large spread i n GAMMA for small X i n both F i gure 1 5 a nd 1 6 . Altho ugh F i gure 16 ( Norman Wells res ults ) does n't s how th i s decrea s i ng vari at ion too well due to the lac k o f data at i nter­med i ate values of X , one would expect the vari ati on to be s i m ilar to that of Fi gure 1 5 ( Swan H i lls results ) . These large dev i at ions a bo ut the apparent mean value are due to small errors i n the mea s urement proces s . To expla i n properly how th i s happens ,

8

one must fi rst loo k at the vari a t i o n of Y w i th X a s shown i n F i gures 1 9 and 20 for the Swan H i lls and Norman Wells crudes res pecti vely . One may note that X and Y vary li nearly up to X � 0.2 cm and tha t both X and Y are very nearly equal up to thi s poi nt . Thi s i s to be expected as the drops are st ill almo st spheri cal . Thi s means however , tha t X/V i s very nearly equal to +1 and as shown i n F i gure 2 1 and 2 2 , B ( eq uati on 5 ) approaches zero whi le F and G ( equati o n s 6 and 7 ) both approach one as X/V goes to one. Eq ua t ions 3 and 4 are there­fore , undefi ned at X/V = 1 . When X/V i s s t i ll very nearly' equal to one , a small error i n meas urement can lead to q u i te large errors i n the calculated value of the i nterfa c i al tens ion . Results of an error analys i s are g i ven i n Table 5 for Swan H i lls crude to demo nstrate thi s fact. Values of Y were f i rst calculated from the regres s i on equati on and error limi ts of ±O.003 cm (� .001 2 i nch ) were a s s i gned to both X and Y. Rati o s of X/V were ta ken so as to maximi ze ( i . e . X i ncreases a n d Y decreases ) and mi nimi ze Z . Fi gure 23 i s a gra ph of the data i n Table 5, and clearly shows that the calcula ted error decrea ses w i th i ncreas i ng X . If one allows for no error i n X and only a +0.003 cm error i n Y , the res ulti ng error i s actually i ncrea sed even though there i s a smaller rela ti ve error between X and Y . It i s , therefore , concluded that the equat ions rela t i ng the i nterfac i al tens i o n to the bubble d i men s i ons are not val i d for values of X less tha n 0.2 cm . Graphs showi ng the effects of an error i n the constant C ( equal to g'�p ) have also been prepared for both Swa n H ills and Norman Wells crudes ( F i g ures 24 and 2 5 ). These fi gures were prepared u s i ng valu es of Y calculated from the regres s i on equati ons g i ven i n F i g ures 1 9 and 20. For the Swan H ills crude ( F i gure 24 ) the values of the i nterfa c i al tens i on are approximately constant over the range 0.2 < X < 0.7. It can not be i nterpreted from F i gure 24 that the surface tens i o n decreases after X = 0.7 , s i nce the apparent red ucti o n i s a d i rect result of the uncerta i nty i n Y calculated from the regress i on. A d i fferent type of regress i o n procedure ( eg . a quadrat i c spli ne fi t wi th zero bend i ng moment at the end po i nts ) wo uld po s s i bly have g i ven a more u n i form curve but some fluctuati ons would s t i ll have been pres ent . Because of t he smaller number of data po i nts at the larger values of X , the value est imated for Y i n the regres s i on analys i s w ill be e i ther above or below the ' actual value' by a small amoun t , thi s error w ill cause a small error i n the rat i o X/V whi ch i n turn w ill res ult i n a large error i n B ( s ee F i gure 21 ) w i th l i ttle effect on F2 and G2. I t i s read i ly seen tha t B i ncrea s es much more rap i dly than e i ther F2 or G2 decrease by compari ng F i gures 2 1 a nd 22. As a res ult , the calculated surface ten s i on w i ll show the fluctuati ng vari ati ons shown i n F i g ure 24 and 2 5. For the Norman Wells crude , thi s effect i s even more apparent ( F i gure 25 ). The i nterfac i al tens i on calculated at X = 0.3 compares favourably , however , w i th the average of the tes t

9

pa i r res u l ts ( Tabl e 4 ) . Th i s was to be expected , however , because the majori ty of data po i nts for t h i s c rude are very c l ose to X = 0 . 3 . For va l ues of X > 0 . 3 , the ca l cu l ated val ues of Y from the regres s i on ana l ys i s and the res ul ti ng surface tens i ons wou l d therefo re , be h i gh l y s u s pect . The val ue of the constant (g . �p ) was ta ken to be 1 60 . 7 and 1 58 . 3 for the Swan H i l l s and Norman Wel l s crudes respect i vel y . Val ues for the i nterfac i a l tens i on o f the Swan H i l l s and Norman Wel l s crudes at the temperature of freez i ng water ( i n t h i s case , a pprox imatel y 2 9 . 5° F for a 1 2% 0 water sa l i n i ty ) have therefore been ta ken a s the average val ues obta i ned i n our experimental programme . These a re ( s ee Tabl e 4 ) 25 . 5 and 24 . 0 dynes/cm res pecti ve l y for the Swan H i l l s and Norman Wel l s crude o i l s . No effects on the i nterfac i a l ten s i o n due to agi ng of t he o i l s co u l d be determi ned . I f such effects do exi st , they are wel l w i th i n the l im i ts of accuracy of th i s method and wou l d have to be determi ned i n another manner . Some add i t i onal data rel ati ng the equ i l i bri um th i c knes s H to the parameters X and Y have a l so been prepa red . Shown i n F i gures 26 and 27 a re graphs of Y vs H for a l l mea surabl e data pa i rs . Us i ng these graphs and the asymptoti c val ues for Y i n F i gures 1 9 and 20 , one can read i l y a rri ve at an estimate for the eq u i l i br i um fi l m th i c kness of the two o i l s . These a re 0 . 80 and 0 . 88 cm for the Swan H i l l s and Norman Wel l s c rude o i l s respect i vel y . The l atter number wa s determi ned by extrapo l at i ng the ' best l i ne ' through the avai l ab l e data and p l otti ng on t h i s l i ne a val ue of Y = 0 . 6 whi c h was ta ken from F i gure 20 a s representa t i ve of the l imi ti ng Y val ue . F i gures 28 and 2 9 have a l so been i nc l u ded here for ease i n rel ati ng H back to X .

2 . 4 . 2 Ri ng Detachment Method The tes t res u l ts for th i s part of the s tu dy are gi ven i n Append i x B and p l o tted i n F i gure 20 . I t s ho u l d be noted tha t the i nterfac i a l tens i ons represented are w i t h respect to the ambi ent a i r . To ca l cu l a te o i l / br i ne i nterfac i a l ten­s i on , one can use Antonow ' s Law ( Adamson , 1 960 ) wh i ch states t hat for mutual l y saturated l i qu i ds

Ya b = I Ya, - Yb' I . . . ( 8 ) Th i s ru l e genera l l y ho l ds appro x i matel y , and i n the absence of s uffi c i ent data , i t i s u seful for est i mati on purposes . A more rel i ab l e method i s t hat of Good et a l , ( 1 9 58 ) who have o bta i ned a semi -emp i r i ca l equat i on for the i nterfa c i a l ten s i on wh i c h i s ,

,

Yab = Ya + Yb - 2� (YaYb )2 . . . ( 9 )

1 0

where

. . . ( 1 0 ) and V i s the mol ar vol ume of the phase i n quest i on . For the Swa n H i l l s crude � = 0 . 832 w h i l e for th� Norma n Wel l s crude � = 0 . 83 1 . Us i ng the average val ue shown i n F i gure 30 for the s urface ten s i o n of the 1 2% 0 bri ne so l uti on and the s urface ten s i ons of the Swan H i l l s and No rman Wel l s crude o i l s as gi ven i n Append i x B one obta i ns the fol l owi ng i nter­fac i a l ten s i ons between t he o i l / br i ne phases .

Yab = 23 . 91 dynes/cm for the Swan H i l l s crude , and

Yab = 24 . 32 dynes/cm for the Norman Wel l s crud e . These numbers compa re favourabl y wi th those arri ved at i n the prev i o u s sec t i o n but before a compar i son can be made , the above val ues must be corrected for the temperature d i fference . Ass umi ng a -0 . 1 cha nge i n tens i on per ° C ( Wash burn , 1 927 ) , and a temperature d i fference of 27 . 4°C res u l ts i n a correct i o n of +2 . 74- dynes/cm to be added to each of the a bove tens i o n s . F i nal i nterfa c i a l tens i ons arri ved at are therefore , 26 . 65 and 27 . 06 dynes/cm for the Swan H i l l s and Norman Wel l s c rude o i l s respect i vel y . Though these res u l ts cannot be cons i dered a s absol utel y correct , they shou l d be wi t h i n 1 0% o f the correct va l ue a l l owi ng for errors i n the ring detachment method itsel f and i n the temperature correc t i o n factor . These resu l ts do , however , l end c redence to the res u l ts obta i ned i n the prev i o u s sect ion and have , therefore , ful fi l l ed the i r purpose .

2 . 5 Concl u s i ons From the preced i ng d i scuss i o n , one may conc l ude the fol l owi ng: a ) The i nterfa c i a l tens i on between the Swan H i l l s crude and a 1 2% 0 bri ne so l u t i o n i s 24 . 5 dynes/cm at a pprox i ma te l y 29 . 5 ° F w h i c h wa s determi ned to be the average freez i ng temperature of the bri ne . The i nterfac i a l tens i o n between the Norman Wel l s c ru de a nd a s i mi l ar bri ne sol ut i o n i s 23 . 8 dynes/cm . These val ues are estimated to be correct to wi th i n 1 0% of the true val ue . b ) An a l ternat i ve method has been establ i s hed for determi n i ng t he i nterfac i a l tens i ons between the o i l s and bri ne at sa l i n i t i es other than 1 2% 0 us i ng the res u l ts obta i ned for t he r i ng det�chment method . One c ou l d a l so use F i gures 24 and 2 5 by a s s umi ng that the bubbl e d i mens i ons wi l l not change s i gnifi cantl y wi th a cha nge i n sa l i n i ty . S i nce the constant g . �p i s dependent on sa l i n i ty , one can therefore , arri ve at i nterfac i a l ten s i ons by i nterpcl at ion o n these graphs .

1 1

c ) Data a l l owi ng an est imat i on o f the equ i l i bri um t h i c knes s for use i n under- i ce spreadi ng ca l c u l a t ions has been presented . These a re 0 . 80 and 0 . 88 cm for the Swan H i l l s and Norman Wel l s c rude o i l s respect i vel y . I t i s est imated t ha t these va l ues are wi t h i n 6% a nd 1 0% for the Swan H i l l s and Norman Wel l s crudes respecti vel y . d ) The ses s i l e drop method repo rted here ( a s per Sta i copol o u s ) brea ks down at smal l va l ues of bubbl e rad i u s because the bubbl es at th i s po i nt are st i l l very near ly s pheri ca l . e ) The u se of po l ynomi a l l east s quares regress i on fi t on t he mer i d i onal hei ght Y can be appl i ed s ucces s fu l ly up to val ues of X < 0 . 8 cm approx i mate ly . Beyond th i s po i nt i nstabi l i t i es set i n due to the numerica l ca l c u l a t i o n techn i que .

3 . MOVEMENT O F O I L DROPS 3 . 1 Genera l

vlhen a buoyant gas and o i l p l ume r i ses through a wa ter col umn as wo u l d be the case in an underwater b l owout , the o i l w i l l brea k d own i nto smal l dropl ets . At the u nders i de of the i ce , mo st of these drop l ets wi l l coal esce to form an o i l s l i c k . S ho u l d th i s s l i ck move a l ong the bottom of the i ce , bubbl es of o i l wi l l brea k away from the peri phery due to i nterfaci a l i nstabi l i ty . Th i s fact has been demon­strated recentl y by Norcor Eng i neeri ng and Research L im i ted i n f i l ms of the i r work o n the behav i our of o i l u nder i ce whi ch i s part of the Beaufort Sea Project . I n th i s work , we have endeavoured t o determi ne the force req u i red to set suc h bu bbl es i n mot i on . Th i s force can then be u sed to est imate the c urrent necessary to i n i t i ate mot i o n of an o i l bubbl e . I t s ho u l d be noted that no attempt has been made here to defi ne the fl ow cond i ti ons to the i nstabi l i ty whi c h causes t hese bubbl es .

3 . 2 Anal yt i ca l Devel opment Cons i der i ng an o i l bubbl e u nder an i nc l i ned i ce s heet as s hown i n F i gure 3 1 , where R i s the res u l tant buoyancy force , one can eas i l y s how that at t he onset of mot i o n

R I = [ pw - 1 ] M g s i n a. po 0 . . . ( 1 1 )

I t i s th i s q uant i ty t ha t has been determi ned i n t he present study. I n an i dea l i zed fi el d s i tuati o n , the bottom of the i ce sheet wou l d be hori zonta l , and t he dr i v i ng force i n i t i at i ng the moti o n wou l d be the s hear stress T, exerted on the o i l by the mov i ng water . The s hear stress req u i red to i n i t i ate mot i o n can t hen be repres ented by

R I = J>dA . . . ( 1 2 )

where A , for l arge bubbl es , becomes t he area over wh i c h the o i l has

1 2

s pread . For bubb l es of the s i ze stud i ed here , a s s umi ng an area of TI·x2 a nd negl ecti ng the edge eff�cts wou l d y i el d a good fi rst approx imat ion to T. When con s i deri ng l arge areas of o i l , one can use the Bl a s i u s sol ut ion to fl ow o ver a fl at p l ate ( S c hl i cht i ng , 1 968 ) to rel ate the s hear stress to Uoo• I t s ho u l d be noted that both of these approaches to the probl em ass ume that the o i l has spread to i ts equi l i bri um s ha pe in t he a bsence of any shear stres s a n d that s udden l y such a s tress fi el d i s appl i ed t o t h e o i l .

3.3 Experi men tal Procedure Th i s seri es of tes ts wa s run i n a refri gera ted tra i l er us i ng essen­t i a l l y the same equ i pment a s descri bed i n Sec t i o n 2 . 3. One mod i fi ca­t ion req u i red wa s a mec han i sm for t i l t i ng the tan k . A pl ywood p l atform was b u i l t for t h i s purpose . I t was h i nged to the workbench at one end and at the other end a l ead screw wa s instal l ed through the bench whi c h a l l owed the pl atform to b e rai sed or l owered . The tan k s a t o n t h i s p l atform together wi th a l l the i ns u l at i on . Both tanks were a l so mod i ­fi ed by enc l o sing a dead a i r s pace , a l ong the l €ngth o f the tan ks , between the ori gi nal and add i t i onal outs i de wa l l s . Th i s wa s needed to prevent the freez i ng of bri ne down the i ns i de s urfaces of the tanks d u ri ng a test. Al tho ugh t he a i r gap d i d prevent the freez i ng of brine on the i n s i de wa l l s , i t prevented one from t hawi ng the l i p o f i ce formed d ue to conduc t i on a l ong the wa l l . Th i s made i t a l most i mpos s i bl e to obta i n a good p i cture from whi ch the hei ght of the bubbl e cou l d b e mea s ured accuratel y . Opera ti o n i n the tra i l er presented s everal add i ti o na l d i ffi cu l t i es . T h e act i on of the cathetometer s l i des became very st i ff i n the co l d ma k i ng accurate focu s i ng of the camera d i ffi c u l t . Growi ng fl at i ce s heets wa s a l so d i ffi cu l t . A way wa s however , found of growi ng p l ane s heets ( not paral l e l ) and these were then l evel l ed by bl oc k i ng up the ent i re wo rkbenc h . Once an i ce s heet wa s ready for testi ng , a quant i ty of o i l was i n­jected under the i ce. The bubbl es were then photogra phed and the tank t i l ted unt i l they started to move . Th i s wa s determi ned v i sua l l y and a measurement was t hen ta ken o f the e l evati o n of the tank at the l ead screw . On a number of occas i o n s , i t was not po s s i b l e to obta i n a p i cture of the bubbl e and so an est imate of t he vol ume wa s made ba sed on the amou nt of o i l i njected and the rel at i ve s i ze o f other bubbl es i f any . Vel oc i ty meas u rements were ta ken but were not i nd i cat i ve of the or ig i na l bubb l e i n mos t cases . Th i s i s because mo st o f the bu bbl es bro ke i nto a number of sma l l er bubbl es . I n such cases , a l a rge bubbl e wou l d start s l owl y , t hen nec k down , form a second bubbl e (wh i ch d i d not necessari l y move at t hat angl e ) and then conti nue . The vel oc i t i es t hat were recorded were for the fi rst bubbl e to reac h the fi n i s h marker and were , therefore, i n no way representat i ve of the ori gi nal bubbl e that wa s measured .

1 3

3.4 Res u l ts Tes ts were run o n bo th the Swa n H i l l s and Norman Wel l s crude o i l s . Res u l ts of the sess i l e drop experi ments were used to cal c u l ate the mas s of a bubbl e as a func t i on of X accord i ng to a method descri bed by Sta i copo l us ( 1 962 ). No s ign i fi cant d i fference cou l d be deter­mi ned between t he two o i l s as s hown i n F i gure 32 where the data for both o i l s has been p l o tted . F i gure 33 s hows th i s data on a l og-l og sca l e and revea l s that the ca l c u l a t i o n method i s l i m i ted to X � 1 . The exact reason for t h i s i s not known , but i s i s s u spected that i t i s due to the po l ynomi a l functi ons u sed i n the emp i ri cal equat i ons .9.i ven by Sta i copol us . The ma s s of a cyl i nder of o i l hav i ng a rad i us X and a height of 0 . 8 cm has a l so been p l o tted i n F i gure 3 3 for com­par i so n purposes. For s ubs eq uent ca l c u l ati ons , the s tra i gh t l i ne extrapo l a t i o n through the ca l c u l a ted data po i nts ha s been used . F i gures 34 and 3 5 were t hen prepared u s i ng the exper i mental data and c l ear ly show that the angl e of i nc l i nat ion neces sary to i n i t i ate moti on decreases w i th the mass of the o i l whi ch was to be expected . Appl i ca t i o n of equati o n 1 1 then resu l ts i n the-data presented i n F i gures 36 and 37 wh i c h g i ve the rel a t i o ns h i p between the force req u i red to i n i t i a te mot i o n of an o i l bubbl e a nd i ts ma ss . Equat i ons descri b i ng the best l i nes thro ugh the data are ,

and F = 48 . 5M o . 486

F = 23.4�10.659 ... ( 1 3 ) ... ( 1 4 )

for the Swan H i l l s and Norman Wel l s c rude o i l s respect ivel y . I t i s seen tha t the fo rce req u i red i ncreases wi th the mas s of o i l present .

3.5 Co nc l us i on s The fo l l owi ng concl u s i ons may b e drawn from t h e preced i ng d i sc u s s i on s: a ) The angl e a t whi c h an o i l bubbl e wi l l move up a n i nc l i ned i ce

s heet decreas es a s the mas s of the bubbl e i ncreases. b ) The force req u i red to i n i t i a te mot i on i nc reases wi th t he mas s

o f the o i l bubbl e a s s hown i n F i gures 3 6 and 37. For the Swan H i l l s crude , th i s force i s gi ven by F = 48 . 5M o.486 , wh i l e for Norman Wel l s crude i t i s gi ven by F = 23.4Mo.659.

4. OIL MOVEMENT I N A LEAD 4.1 General

How far o i l s pi l l ed on col d arcti c wa ters wi l l s pread i s of parti cu­l ar i mportance , espec i a l l y at amb i ent wi nter temperatures . I n an offs hore envi ronment such s pread i ng w i l l typ i ca l l y occur i n a l ead . The u l t i mate extent of the s pread w i l l , of cours e , be dependent on the amount of o i l rel eased . S i nce t h i s parameter can o n l y be determ i ned once a s p i l l has occurred and t hen o n l y approx imatel y , i t wa s thought that the best approach to th i s probl em wou l d be to determi ne the i n i t i al and fi na l s pread i ng coeff i c i ents from w h i c h an equ i l i br i um

1 4

fi l m t h i c knes s cou l d t hen be cal cu l ated . Th i s th i c knes s cou l d then be combi ned w ith the vol ume es t imate to gi ve an est imate of the maxi ­mum area l s pread t hat cou l d reasonabl y b e expected. Determi nat i on of the rate of spread duri ng the earl y l ife of the s p i l l may be made us i ng the methods descri bed i n t he open l i terature ( Chri s t i a nsen and H i xso n , 1 969; Fay , 1 97 1 ; Wa l d ham et a l , 1 972; and , Fannel op and Wa l dham , 1 972 ) and wi l l not be presented here.

4.2 Spread i ng Ana l ys i s The profi l e of a n o i l l ens fl oat i ng o n water i s s hown i n Fi gure 38 , where t repres ents the equ i l i bri um fi l m t h i ckness. Usua l l y t h i s quant i ty i s u sed to ca l cu l ate the s pread i ng coeff i c i ent S from a n equat ion gi ven by Langmu i r ( 1 933 ) as:

too 2 = - 2Sp/gPbllp · .. ( 1 5 )

where the subscri pt a represents t he water phase and b the o i l phase. I n the present work, equati o n 1 5 i s u sed- to ca l cu l ate too by ca l cu l at i ng val ues of S from the data presented i n Sec t i o n 2 of t h i s report. I t shoul d be noted here that equat ion 1 5 i s meani ng­ful on ly i f the spread i ng coeffi c i ent S i s negati ve. The s preadi ng coeffi c i ent i s defi ned as fol l ows (Adamson , 1 96 0 ) ,

· .. ( 1 6 )

where the s urface tens i ons of a and b i n equat ion 1 6 are those for the pure l i qu i ds. When two substances are i n contact however , t hey w i l l become mutual l y saturated , so t hat Ya and Y b wi l l become Yal and Ybl res pect ive ly. The correspond i ng spread i ng coeffi c i ent i s then wri tten as S bl/al or just S I . I t i s t h i s l a tter quant i ty wh i c h must b e used i n equat i o n 1 5 to determi ne too and i s gi ven by ,

· . . ( 1 7 )

Val ues of these parameters for t he Norman Wel l s and Swan H i l l s c rude o i l s ( c orrected to O°C ) are presented i n Append i x C a l ong wi th ca l c u l a ted val ues of S and S I . The s pread i ng coeffi c i ents l i s ted i n Append i x C are for a part i c u l ar po i nt i n t i me and for the cond i ti ons i nd i ca ted in the notes to Ta bl es C- l and C- 2 . S i nce t he sampl es were kept i n s ea l ed bottl es , negl i gi bl e agi ng occurred. I n rea l i ty , cons i derabl e evapora t i o n of the l i ght hydrocarbon fract i on s wou l d occur , thereby not on l y red uc i ng t he vo l ume of o i l present , but al so i ncrea s i ng the dens i ty of the rema i n i ng o i l . Th i s wou l d l ead to a l oweri ng of the s preading coeffi c i ent because th i s coeffi c i ent i s po s i t i ve o n l y for the l ower mol ecu l ar wei gh t hydrocarbons ( Pomerantz et a l , 1 967 ). In add i ti on , natural s u rfactants present i n the o i l wou l d d i ffuse i nto the water . S i nce these orga n i c s u rface act i ve consti tuents cause the spread i ng ( Garrett , 1 97 3 ) , one wou l d expect the s pread i ng coeffi c i ent to eventua l l y decreas e and become negati ve . That th i s d i d not happen for the Swan H i l l s s amp l es i s mo st l i ke l y due t o the fact t hat these sampl es were n o t thoro ugh ly mi xed wi th the bri ne as was the case wi th the Norman Wel l s sampl es . Attempts at

"

'c

1 5

mi x i ng the former sampl es y i el ded emu l s i on s that cou l dn ' t be bro ken e i ther by heati ng or centr i fugi ng t hem . Even i f a negat i ve spread­i ng coeffi c i ent had been measured for the Swan H i l l s samp l es , u se of t h i s i n ca l cu l at i ng an equ i l i br i um fi l m t h i c knes s woul d not be recom­mended because of t he rel at i vel y h i gh pour po i nt of th i s c rude (-9 . 5 ° C ) . Be l ow t h i s temperature , the o i l becomes non-fl u i d i n character ( i t gel s ) and spread i ng d ue'to s urface forces wou l d be essent i a l l y ha l ted ( Garrett , 1 973 ) . The fi l m th i c knes s ca l cu l ated for the Norman Wel l s crude (Append i x C ) s ho u l d be regarded as a mi n i mum to be expected . W i th evapo ra t i o n , etc . , th i s t h i c knes s wou l d be i ncreased . One s ho u l d a l so note that these val ues were cal cu l a ted for a O°C mean water and o i l temperature . At l ower temperatures , the s pread i ng coeffi c i ent wi l l be decreas ed even further whi ch wi l l a l so resu l t i n an i nc rea se i n t he equ i l i bri um fi l m th i c knes s .

4 . 3 Conc l u s i ons From the preced i ng d i scu s s i o n , one may concl ude the fol l owi ng: a ) The m i n i mum equ i l i br i um thi c knes s of the Norman Wel l s crude on

water under arcti c wi nter cond i t i ons w i l l be approximatel y 0 . 25 em (Appendi x C ) .

b ) A s i mi l ar mi n imum equi l i bri um t h i c knes s may be expected for the Swa n H i l l s crude tak i ng i nto account the effects of agi ng ( evaporat i on ) and the rel at i vel y h i gh pour po i nt of t h i s c rude .

c ) The equ i l i br i um fi l m t h i ckness quoted i s cons ervati ve i nsofa r as o n e may reasonabl y expect th i s f i gure t o i nc rease a s t h e o i l weathers .

5 . SOLUTE REDI STRI BUTION AND O I L P ENETRAT I ON 5 . 1 So l ute Red i str i but ion

5 . 1 . 1 Genera 1 I t i s genera l l y thought that many crude o i l s conta i n a certa i n amount o f d i sso l ved sa l t i n the i r produced form . The probl em pos ed , therefore , was whether o r not the sa l t produced a s a resu l t of an underwater bl owou t wou l d be tran sported by t he o i l to the under- i ce s u rface where i t may be rel eased caus i ng pos s i bl e rott i ng of the i ce s heet . Whether the i ce sheet ro ts i s a l so a funct i on o f the amou nt of sa l t , i n exces s of t he equ i l i br i um concentra t i o n a l ready presen t , ava i l a bl e at the u nder-i ce s u rface for l oweri ng the freez i ng temperature . Th i s probl em i s i mportant when one con s i ders a n o i l wel l b l owout i n the w i nter dur i ng wh i ch t i me o ne wou l d expect to be a bl e to move cons i derabl e heavy equ i pment over the i ce i n a c l eanup effort . Such movement cou l d b e i mpeded i f s i gn i fi cant rotti ng a nd weaken i ng of t he i c e were to ta ke p l ace .

16

5 . 1.2 Method and Resu l ts I t was deci ded that an experi mental i nq u i ry wa s unnecess ary as a l l of the perti nent i nformati o n was i n hand . The probl em was red uced to that of : a ) determi n i ng the amou nt of d i s sol ved sal t that may rea so nab ly be expected i n the o i l , b ) exami n i ng the method used i n arri v i ng at'these f igu res , and c ) d i scus­s i ng the probl em wi th severa l peopl e i n our ana l yt i ca l chemi stry group . The property record s of c rude o i l s from 2 2 d i fferent locations were rev i ewed . Sa l t contents of these crudes are presented i n Ta bl e 6 and i n F i gure 3 9 whi c h s hows t he resu l t s as hav i ng an expo nenti a l l y dec rea s i ng frequency d i stri but i on . A des­c r i pt i o n of the test method u sed in determi n i ng the sal t contents l i sted i n Tabl e 6 i s gi ven i n Append i x D . I t wa s poi nted out i n several d i scuss i ons ( persona l commu n i cati ons w i t h H . A . Jacobson , R . E . Heater and W . N . McKay ) that th i s tes t does not necessa r i l y i nd i c ate t hat-sa l t i s actual l y d i s­so l ved i n t he crude o i l . Mos t o f t he sa l t actual l y produced from a wel l comes i n the form of bri ne whi c h i s general l y removed i n t he fi el d before the crude a s s uc h i s ana l yzed . Whether the sa l t s ubsequentl y measured i s conta i ned as d i s­sol ved s a l t i n the o i l or whether i t i s present a s m i nute bri ne d rop l ets d i spersed i n t he o i l i s s t i l l s ubject to debate , a l tho ugh the l atter seems more l i ke ly ( personal com­mun i c at i o ns wi t h H . A . Jacobson and R . E . Heater ) . I n e i ther case the phys i ca l s i tuati on stro ngly m i t i gates aga i nst any of the sa l t reac h i ng the u nder-i ce s urface . Th i s i s d ue to the v i o l ent agi tat i o n expected as t he p l ume i s sues out of the pi pe and r i ses through 1 5 to 6 0 metres of water . The ensu i ng brea k­down of the o i l stream i nto very fi ne drop l ets duri ng i t s ascent shou l d rel ea s e any sa l t i n the s tream t o the water phas e . I f one a s s umes that th i s d i d not happen and that t here wa s d i s so l ved sa l t i n t he o i l t hat was carri ed to the u nder-i ce s urface , then one can ca l cu l ate an area l d i stri buti o n o f the sa l t due to t he movement of the o i l . Ta k i ng the average sal t content a s 27 . 1 pound s per M bb l , one o bta i ns 40 . 65 pou nds per day a s s um i ng t hat the o i l i s rel ea sed at a rate of 1 500 bbl /day . Ta k i ng an equ i l i br i um t h i c kness of O .B cm for the o i l and i gnor i ng any add i ti ona l spread i ng due to water cur­rents or i ce movement , one obta i ns a n area l spread of a pprox i­mate ly 29 .B3 x 1 07 cm2• One can a l so a s s ume t hat t he sa l t wou l d be qu i te s l ow i n comi ng out of so l u t i o n i f i t has n ' t done so dur i ng i ts ascent i n the o i l . On th i s bas i s , one obta i ns a fi gure of 6 . 2 x 10-5 gm of sa l t i n t he o i l per square cm of i ce covered by the o i l . Th i s can al so be repre­s ented a s a sa l i n i ty of approx imately O . OB% o' S i nce the o i l has now d i sp l aced the water , a reduct i o n i n the sal i n i ty at t he i nterface ha s ta ken pl ace . And a s s umi ng that a red i s­tri but i on of the sa l t does take p l ace ) through some concentra t i o n d i ffu s i o n proces s ) , t hen i t wou l d be rea sonabl e

1 7

to expect the concentra t i o n i n the o i l to i ncrease and that in t he i ce ( bri ne dra i nage c hannel s ) to decrease .

5 . 1 . 3 Conc l u s i o n s

Whether sa l t actua l l y exi sts a s a d i s so l ved spec i es i n the o i l i s open to quest i o n . Ass umi ng , however , t hat i t does , t hen one may concl ude from the precedi ng d i scu s s i on that t he d i sso l ved sal t wi l l not cause the under- i ce s urface to rot . I t i s a l so suggested , s i nce the sa l t i s more l i kel y to be present i n d i s persed water dropl ets , t hat the majori ty of the sa l t wou l d be rel eased to the water col umn d uri ng the o i l ·s a scent .

5 . 2 O i l Penetra t i o n

Thi s s ect i on cons i ders whether o r not the sma l l o i l bubbl es that ri s e t o the bottom s urface of t h e i ce wi l l penetrate t h e i ce structure due to their buoyancy forc e . I f t h i s were t he case , t hen a con s i derabl e amou nt o f t he o i l resu l t i ng from an u nderwater b l owout i n the w i nter wou l d be retai ned by the i ce . Th i s cou l d cons i dera b ly l i m i t the spread of the o i l but cou l d a l so hamper recovery operat i on s due to the l arge area l d i s tr i buti o n pos s i bl e for s uch drops . A combi nati on of observa t i ona l ev i dence and phy s i ca l reason i ng has been u sed i n formu l at i ng a so l ut i o n to th i s q uest i on . I n performi ng the s urface tens i on exper iments , a l arge number of oi l drops were stud i ed from an even l arger grou p of drops that had been i njected u nder the i ce . At no t i me wa s any bubbl e o bserved to d i s­a ppear i nto the i ce d uri ng an experi men t . An exami nati on was made of several i ce s heets a fter an experiment to determi ne whether some o i l had actual l y penetrated t he i ce . No traces o f o i l cou l d be found . Wo l fe and Houl t ( 1 972 , 1 974 ) i n their study of o i l under sea i ce a l s o observed t hat negl i gi b l e amo unts of o i l are entrapped i n the i ce bri ne matri x . What was noti ced , however , was that there were i n many cases , sma l l depress i ons l eft o n the u nder- i ce s urface where the o i l had been . These depress i ons are attr i buted entirely to our method of remov i ng the i ce from the test tanks whi c h was to remove the tanks from t he col d room and to a l l ow s uffi c i ent i ce around the edges to mel t so that the i c e b l ock cou l d be removed . Th i s norma l l y took q u i te some t i me and res u l ted i n cons i derab l e heat gai n by bot h t he i ce and the bri n e . Such behav i our i s eas i l y expl a i ned by cons i deri ng that the average rad i u s of the bri ne dra i nage c hannel s at the bottom i ce s urface i s of the order o f 0 . 1 mm ( As sur , 1 958; E i de a nd Mart i n , 1 975 ) and th i s s i ze genera l l y i ncrea ses somewhat as one moves up from the under s urface of the i ce . These bri ne c hannel s can be l i kened to i nverted i nk bottl es , havi ng narrow nec ks and bei ng w i der i ns i de . Even t hough the o pen i ngs are not tru l y c i rcu l ar , genera l l y they are more e l l i pt i ­ca l i n s hape (Assur , 1 958 ) , one can a s s ume that t hey are c i rc u l ar .

1 8

The press ure drop across the openi ng can be g i ven by (Adamson , 1 960 ). �P = ( 2y/r ) cos e

Ta k i ng y ( the surface tens i o n ) a s 25 dynes/cm and e ( the contact angl e ) a s 1 50° for th i s sampl e cal c u l a t i o n , o ne o bta i ns a press u re drop �P ( i . e . dri v i ng force req u i red to penetrate the men i scus ) dependent on ori f i ce s i ze as fo l l ows :

r ( mm ) �P ( dyne/ cm2 )

= 0.1 = 4330.

0.2 2 1 67.

0.5 866.

0.7 61 7.

If one now ta kes an o i l bubbl e th i c knes s of 0.8 cm , th i s i s eq u i va­l ent to a �P across the o i l ( P at the i nterface between o i l and i ce i s referenc e ) of approximately 636 dyne/ cm2 ( u s i ng a spec i fi c gravi ty of 0.81 for the o i l ) . One can ea s i l y see that i f the ori f ice s i ze i s i ncrea sed to 0.7 mm (�P equ i va l ent ·to 67 1 . dyne/cm2 ) that some o i l can then be expec ted to move i nto the c hannel . I n the case of Wol fe and Ho u l t ( 1 974 ) , i t seems that the o i l was not coo l ed pri or to i ts i nject ion u nder the i c e . Heat transfer from the warmer o i l to the i ce cou l d have caused some mel t i ng to ta ke p l ace wi th a res u l tan t i ncrea se i n t h e s i ze of some o f t h e bri ne dra i nage c hannel ori fi ces. Al terna t i vel y , the temperature gradi ent between the o i l and the i ce cou l d have res u l ted i n a therma l d i ffu s i on o f the sa l t i n the bri ne c hannel s towards the i nterface between the o i l a nd i ce . A reducti on of the freez i ng temperature of the br i ne wou l d resu l t wi th the effect that mel ti ng at the bri ne drai nage c ha nnel or i fi ces cou l d occur caus i ng an enl argement of some of these or i fi ces . A s im i l ar res u l t cou l d be expec ted i n e i ther case; a ) i nc rea se i n or i fi ce d i ameter , and b ) a correspond i ng decrea se i n �P . One can draw the fo l l owi ng concl u s i o ns from the preced i ng d i scu s s i on: a ) Ass umi ng the o i l to be i n therma l equ i l i bri um wi t h i ts surround­

i ngs , one s ho u l d no t expec t any s ign i fi cant penetrat ion of o i l i nto the i ce .

b ) The l i mi ted penetra t ion o f o i l tha t may occur wi l l l i ke l y resu l t from the o i l encounteri ng an over s i zed bri ne dra i nage o ri f i ce .

c ) Penetrat ion wi l l occur i n the spri ng a s mel t i ng proceed s and the dra i nage c hannel s open.

6. CONCLUS IONS An eva l uati o n of some of the parameters a ffec t i ng the fl ow and area l d i s­tri but ion of crude o i l u nder a s ea i ce canopy has been pres ented . I t was found that the i nterfac i a l tens i o n s between o i l a nd bri ne ( 1 2% 0 ) for Swan H i l l s and Norman Wel l s crude o i l s were 24.5 and 23.8 dynes/ cm respec t i vel y . I nterfa c i a l tens i ons a t sa l i n i ti es o ther t ha n 1 2% 0 have a l so been p re­sented . Effects of agi ng on the i nterfac i a l tens i o n cou l d not be determi ned due to the scatter i n the mea sured data . The equ i l i br i um thi c kness of these two crude o i l s under i ce was fou nd to be 0.80 and 0.88 cm for the Swan H i l l s and Norma n Wel l s samples respect i vel y .

1 9

Expres s ions rel a t i ng the force requ ired to initi ate mo t i o n of an o i l bubbl e have a l so been presented . For the Swan H i l l s crude , th i s force i s gi ven by F = 48 . 5Mo .486 whi l e for Norman Wel l s crude i t i s gi ven by F = 23 . 4Mo . 659 . I n these express i o n s , the force , F , i s i n dynes and the ma ss , M, i s i n grams . Data rel ati ng the mas s of o i l to the shape of the bubbl e has a l so been presented . Th i s w i l l enabl e cal c u l ati ons of the mi n i mum currents req u i red to i n i ti ate moti on of an o i l bubbl e to be made . When con s i deri ng the spread of o i l on wa ter under arcti c cond i ti ons , as wou l d be the case of o i l spread i ng i n a l ead , i t was fou nd tha t a m i n i mum equ i l i br i um fi l m th. i c kness .of 0 . 25 cm s ho u l d be expected for the two crudes tes ted . Ta k i ng i nto account the effects of evaporati on and the l each i ng of na tural s urface acti ve agents present i n the o i l i nto the water , i t i s reasonabl e to expect th i s f igure to be conservat i v e in mos t cases . A determi nat i on o f the maximum areal s pread of t he o i l , barri ng any externa l forces ( e . g . , effects of currents , etc . ) , i s therefore pos s i bl e . I t was a l so concl uded that the presence of d i sso l ved sa l t i n t he o i l , i f i ndeed i t does ex i s t a s a d i s so l ved spec i es i n the o i l , wo ul d not cause the under- i ce surface to rot . Penetra t i o n of the o i l i nto the i ce sheet i s not normal l y expected . When the o i l encounters a n overs i zed bri ne dra i n­age c hannel of approximatel y 0 . 7 mm rad i u s , l i mi ted penetrati o n "'ti l l l i kel y resu l t . As mel t i ng proceeds i n the spri ng , and t he bri ne dra i nage channel s open , a s i gn i fi cant amount o f o i l penetrati o n s ho ul d be expected .

20

7 . REFERENCES Adamson , A . W . , 1 960 . P hys i ca l Chemi s try of Surfaces . I ntersc i ence

Publ i s hers I nc . , New York , N . Y . Ass u r , A . , 1 958 . Compo s i ti on o f Sea Ice a nd I ts Tens i l e Strength .

Repri nt from Nat i o na l Academy o f Sc i ences , Nat i onal Researc h Coun­c i l Publ i cat ion 598 - Arc t i c Sea Ice .

A . S . T . M . , 1 97 0 . Standard Method of Test for Interfa c i a l Tens i o n of O i l Aga i nst Water by the Ri ng Method . Des i gnati on 0971 - 50 , Annual Boo k Boo k of ASTM Standards , Part 29 .

Ba shford , F . and J . C . Adams , 1 883 . An Attempt to Tes t The Theori es of of Capi l l a ry Act ion by Compari ng the Theo reti ca l and Mea s ured Forms of Drops of Fl u i d . Cambri dge Uni vers i .ty Press , Cambri dge , Engl a nd .

Chen , LC . , J . C . K . Overa l l and C . R . P h i l i ps , 1 974 . Spread i ng of Crude Oi l on an Ice Surface . Can . J . of Chern . Eng . , ·Vo l . 5 3 . p 7 1 .

Chri s t i ansen , R.M . and A . N . H i xson , 1 957 . Brea kup of a L i q u i d Jet i n a Denser L i qu i d . I ndustr i a l and Engi neer i ng Chemi stry . Vol ume 49 , No . 6 . pp 1 01 7 - 1 024 .

E ide , L . I . and S . Mart i n , 1 975 . The Forma t i on o f Bri ne Dra i nage Features i n You ng Sea Ice . J . of Gl aci o l ogy , Vol . 1 4, No . 70 .

Fannel op , T . K . and G . D . Wa l dman , 1 97 2 . Dynam i c s of O i l S l i c ks . AIAA Jo urnal , Vol . 1 0 , No . 4 . p . 506 .

Fay , J . A . , 1 969 . The Spread of O i l Sl i cks o n a Ca l m Sea . I n O i l on the Sea . D . P . Hou l t , Ed i to r . P l enum Pres s , N . Y . p . 53 .

Fay , J . A . , 1 971 . P hys i ca l Proces ses i n the Spread of O i l on a Water Surface . Proceed i ngs of t he Jo i nt Conference on Prevent ion and Con­trol of O i l Sp i l l s . Ameri can Petrol eum I nst i tute , Wa s h i ngton , D . C . p . 463 .

Fox , H . W . and C . H . C hr i sman , 1 952 . The Ri ng Method of Mea s uri ng Surface Tens i on for L i q u i ds of H igh Dens i ty and Low Surface Tens i on . J . Phys . C hern . 56 , 284 .

Freud , B . P . and H . Z . Freud , 1 930 . A Theory of t he R i ng Method for t he Determi nati on of Surface Tens i o n . J . Am . Chern . Soc i ety 5 2 . 1 77 2 .

Garrett , W . O. , 1 973 . The Su rface Acti v i ty of Petro l eum and i ts I nfl u ence on the Behav ior of O i l at Sea . I n Bac kgrou nd Papers for a Workshop on I nputs , Fates and Effects of Petro l eum i n the Mari ne Envi ro nmen t . NTI S Repo rt A D 783990 .

Gi r i fa l co , L . A . and R . J . Good , 1 957 . A Theo ry for the Es timati on of Sur-face and I nterfac i a l Energ i es . I . Der i va t i o n and Appl icat i on to I nterfaci a l Tens i o n . The Jou rnal of Phys i ca l Chem i s try . Vol . 6 1 , p . 9 04 .

2 1

Gi ttens , G . J . , 1 969 . Vari a t i o n of S urface Tens i on of Water wi t h Tempera­ture . J . of Co l l o i d and I nterface Sc i ence . Vol . 30 , No . 3 , p . 406 .

Gl easer , J . L . and G . P . Vance , 1 971 . A Study of the Beha v i or of O i l Sp i l l s i n the Arct i c . NTI S Report AD 7 1 7 1 42 , U . S . Coast Guard , Wa s h i ngton , D . C .

Good , R . J . , L . A . G ir i fa l co and G . Kraus , 1 958 . A Theory for Est imati on of I nterfa c i a l Energi es Appl i ca t i o n to Surface Thermodynami c s o f Tefl on and Gra ph i te . The Journa l of Phys i ca l Chemi stry . Vol . 6 2 p . 1 41 8 .

Hark i n s , W . D . and H . F . Jordan , 1 930 . A Method for the Determi nat ion of Surface and I nterfaci a l _Tens i on from the Max imum Pu l l on a R i ng . J . Am . Chem . Soc i ety 52 , 1 751 .

H i nze , J . O . , 1 955 . Fundamentals of the Hydrodynamic Mec han i sm of Spl i t­t i ng i n D i s persion Processes . A . I . Ch . E . Jo urna l , Vo l . 1 , No . 3 pp 289- 295 .

Keev i l , B . E . and R . O . Rams e i er , 1 97 5 . Behav i or of O i l S p i l l ed Under Fl oati ng Ice . I n Proceedings 1 975 Conference on Preventi o n and Con­trol of Oi l Po l l ut i on . p . 497 . Sponsored by EPA , AP I , USGG , San Franc i s co , Marc h 25-27 , 1 97 5 .

Kre i th , F . , 1 968 . Pr i nc i p l es o f Heat Transfer . 2nd Ed . I nternati onal Textboo k Company , Scranton , Pennsyl van i a .

Langmu i r , I . , 1 933 . J . Chem . P hys . 1 , 7 56 . Parvat i kar , K . G . , 1 966 . Veri fi cat ion of Emp i ri ca l Equat ions i n Computing

Surface Ten s i on by the Sess i l e- Drop Method . J . of Co l l o i d and I nter­face Sci ence 2 2 . pp 298- 299 .

, 1 967 . Ver i fi cati on of Emp i ri ca l Equat i o n s i n Compu ti ng --�--;:,--.,.---:-the Contact Angl e by the Sess i l e-Drop Method . J . of Col l o i d and

I nterface Sc i ence 23 . pp . 274- 276 . Pomerantz , P . , W . C . C l i nton and W . A . Z i sman , 1 967 . Spread i ng Pres s ures

and Coeffi c i ents , I nterfaci a l Tens i o n s and Ad hes i on Energi es of the Lower Al kanes , Al kenes and Al kyl Benzenes on Water . J . o f Col l o i d and I nterface Sc i ence . Vo l . 24 , No . 1 . p p 1 6- 28 .

Schl i cht i ng , H . , 1 968 . Bo undary Layer Theory . S i xt h Ed i ti on . McGraw H i l l Boo k Co . , New Yor k , N . Y . p . 1 28 .

Sta i copo l us , D . N . , 1 96 2 . The Computati o n of Surface Tens i o n and of Contact Angl e by t he Ses s i l e -Drop Metho d . J. of Col l o i d Sci ence , 1 7 . 5 39-447 .

, 1 963 . The Computa t i o n of Surface Tens i on and of Contact ------A�n-g�l e�b-y�t�h -e Ses s i l e- Drop Method ( I I ) . J . of Col l o i d Sc i ence 1 8 .

793- 7 94 . , 1 967 . The Computati o n o f Surface Tens i on and of Contact ---=-A-ng--=l:'""" e----:-"by----,t-:-h-e Ses s i l e -Drop Method ( I I I ) . J . of Col l o i d and I nterface

Sc i ence 23 . 453-456 .

22

Timmons , C . O . and W . A . Z i sman , 1 968 . The Rel ati on of I n i t i a l Spread i ng Pres s u re of Pol ar Compounds on Water to I nterfac i al Tens i on , Work of Adhes i o n and So l ub i l i ty . J . o f Col l o i d and I nterface Sci ence . Vo 1 . 28 , No . 1 . p . 1 06 .

Top ham , D . R . , 1 974 . Hydrodynami c Aspects of an O i l wel l Bl owout Under Sea Ice . Beaufort Sea Project Study G2a , I nteri m Report , December 1 974 . Dept . of the Envi ronment , V i ctori a , B . C .

Wa l dman , G . D . , T . K . Fannel op and R . A . Johnson , 1 97 2 . Spread i ng and Trans­port of O i l Sl i c ks o n the Open Ocea n . Repr i nts o f the 1 97 2 Offs hore Tec hnol ogy Conference . Vol . 1 , p . 3 53 . ASME , Housto n , Texas .

Was hburn , E . W . , 1 927 . I nternat i ona l Cri t i ca l Ta bl es . Mc Graw H i l l Boo k Co . , New Yo rk , N . Y . Vo l . 2 , p . 1 46 .

Wol fe , L . S . and D . P . Hou l t , 1 97 2 . Effects o f Oi l Under Sea I ce . F l u i d Mec h . Lab . Publ i cat i on . No . 72-1 0 . Dept . Mech . Eng . M I T .

Wol fe , L . S. and D . P . Hou l t , 1 974 . Effects of O i l Under Sea I ce . J . o f Gl aci o l ogy . Vol . 1 3 , No . 6 9 .

8 . B I BL IOGRAPHY B i kerman , J.J .

Davi es , J . T . R i deal , E . K. Hark i ns , W . O .

Adams , N . K.

23

Surface Chemi s try Theo ry and Appl i cat i ons . 2nd Ed i t i on , Academi c Pres s , New �or k , 1 958 . I nterfac i al P henomena . Academi c Pres s , New Yo r k , 1 96 1 . The P hys i ca l C hemi stry of Surface F i l ms . Rei n ho l d Pub l i s h i ng Co . , New York , 1 952 . The Phys i cs and Chemi stry of Surfaces . 3rd Ed i ti on Oxford Un i vers i ty Pres s , London , 1 941 .

Gregg , S . J. The Surface Chemi stry of Sol i ds. 2nd Edi t i on , Chapman and Hal l Ltd � , Londo n , 1 96 1 .

Gou l d , R . F. ( ed i to r ) Co ntact Angl e , Wettabi l i ty and Adhes i on . Advances i n Chemi stry Ser i es 43 , Ameri can Chem i ­cal Soc i ety , Wa s h i ngto n , D.C. 1 964 .

Burdon , R.S. Surface Tens i o n and Spread i ng of L i q u i d s . 2nd Edi t i on. Cambr i dge Uni vers i ty Pres s , Cambr i dge , Engl and , 1 949 .

Adamson , A . W . P hys i ca l Chemi stry o f Surfaces. I ntersc i ence Publ i s hers I nc. , New Yo rk , 1 96 0 .

TABLE 1

�I 2R x=R y B90 F90 . G90 y ( eq 3 ) Elapsed i ( in . ) ( in . ) ( in . ) dyne /cm Time :

70 , sec . 0 . 5772 0 . 2886 0 . 1917 7 . 7607 0 . 65 57 0 . 4 598 28 . 141 25 . 5 min . 0 . 5902 0 . 2951 0 . 1955 8 . 1312 0 . 6371 0 . 4218 29 . 745

Numerical Values of Coefficients (ak) and C ons tants Acp

' Ccp

' Dcp

used in Equations ( 5 ) , (6 ) and ( 7 ) . (From Ref . 1 0 ) .

Quant ity

Br/J Fr/J Gr/J

TABLE 2

C/J Polynomi al Coeffi cient s Const ant s

aO

a1 a

2 a3 a4 ' A r/J

45 3 . 1713 1 . 596 -0 . 1064 -0 . 0526 0 . 0464 3 . 11 90 2 . 5924 2 . 1838 -0 . 1302 -0 . 1347 0 . 1141 1 . 5922 45 0 . 4443 -0 . 2027 0 . 0509 -0 . 009 --- 3 . 11 90 0 . 5864 -0 . 3512 0 . 0859 0 . 00898 -0 . 01415 1 . 5922 45 0 . 1425 -0 . 0979 0 . 0408 -0 . 0124 --- 3 . 11 90 0 . 3684 -0 . 3555 0 . 1857 -0 . 07188 0 . 01838 1 . 5922

Measurements and Calculated Interfacial Tension of Sess i l e Drop a t Two Times Aft er Inj ection (Fi g . 5 and 6) .

Cr/J

0 . 6958 0 . 5922 0 . 6958 0 . 5922 0 . 6958 0 . 5922

y ( eq It ) dyne/cm

25 . 237 29 . 770

D¢

4 . 8 1 . 7 ----- -

------

N -I==>

25

TABLE 3 : F lu i d Dens i t i e s Used in C a l cul at i on s ( at 2 5°

F )

F LU I D SPEC I F I C DENSITY DEN S ITY DENS I TY 3 2

DI FFERENC Et GRAVI TY 1b . m / ft grn/ em

Ai r - - - - 0 . 074 0 . 0 0 1 1 8 5 - - - -

No rman We l l s 0 . 8 2 4 5 5 1 .4 4 8 8 0 . 8 2 4 1 0 . 8 2 2 9

Crude

Swan Hi l l s 0 . 8 1 9 0 5 1 . 1 0 5 6 0 . 8 1 8 6 0 . 8 1 74

C rude

6 % Brine 1 . 0 0 4f 6 2 . 6 5 5 8 1 . 0 0 3 6 1 . 0 0 2 4

1 2 % Brine 1 . 0 0 8 2 6 2 . 9 1 1 7 1 . 0 0 7 7 1 . 0 0 6 5

1 8% Brine 1 . 0 11 2 6 3 . 0 9 8 9 1 . 0 1 0 7 1 . 0 0 9 5

2 4 % Brine 1 . 0 1 6 3 6 3 . 4 1 7 1 1 . 0 1 5 8 1 . 0 1 4 6

3 0 % Brine 1 . 0 2 0 1 6 3 . 6 5 4 2 1 . 0 1 9 6 1 . 0 1 8 4

3 6 % B rine 1 . 0 2 4 1 6 3 . 9 0 3 8 1 . 0 2 3 6 1 . 0 2 2 4

3 t Den s i ty di fference = F lu i d Den s i ty - Ai r Dens i ty in gm . / crn .

T Taken from Fi gure 1 1 .

�rude No .

Type Observ .

Swan 9 2 Hi l l s

No rman 82 We l l s

.

TABLE 4 : Interfaci al Tensions Between O i l and

1 2 % Brine C a l cu l ated From Experimental Data

Me an Standard Max . Min .

Deviation Value Value

2 5 , 45 6 7 . 009 4 7 . 693 6 . 9 2 5

2 3 . 9 8 2 7 . 9 04 49 . 1 5 1 5 . 686

-

, ,' "

Range

4 0 . 768

4 3 . 465

Mean

Deviat ion

4 . 960

6 . 1 0 3

N 0)

TABLE 5 : Error Analysis For Swan Hi l l s Crude

X Y ( cm . ) (cm. )

a . 1 . 0966

. 3 . 2680a

. 5 . 3874a

b . 09 7b . 0996

. 10 3b b . 0936

b . 297b . 2 710

b . 30 3b . 2650

. 49 7b b . 3904

. 503b b . 3844

c . 1 00 . 0996

Xlv z

1 . 0352 - . 9406

1 . 1 194 - . 7984

1 . 2907 - . 5092

0 . 9739 - 1 . 0441

1 . 1 004 - . 8 304

1 . 0959 - . 8 380 - ------

1 . 1434 - . 7579

1 . 2 731 - . 5 389

1 . 3085 - . 4790

1 . 0040 - . 9932

B F2

. 1 675 . 9490

. 11 29 . 8 303

2 . 6585 . 6 168

- . 1 1 70 1 , 0398

. 5 7 1 7 . 8564

. 5400 . 8626

. 90 8 8 . 7980

2 . 3894 . 6 368

2 . 95 1 9 . 5969

. 0 125 . 9948

a Y Value taken from regress ion equation ( s e e F igure 19) b Error o f ± 0 . 003 cm . Maximum error di fference used .

c Error of + 0 . 003 cm . in Y on l y .

G2

. 8860

. 6636

. 3700

1. 0934

. 7082

. 7 192

. 6 1 1 3

. 3928

- . 3483

. 9 860

d C is the constant g · np in equations 3 and 4 . Taken here a s 160 .

X2/ BF2 y2/ BG2

. 0629 . 0629

. 1520 . 1 5 1 8

. 1 525 . 1 526 ------------

- . 0773 - . 0775

. 02 1 7 . 02 1 6

1894 . 1 891

. 1 266 . 1 264

. 1 623 . 1 624

. 1 436 . 1457

. 8009 . 80 1 5

CX2/ BF2 d Cy2/ BG 2 d

1 0 . 066 1 0 . 061

24 . 326 24 . 291

24 . 394 24 . 4 1 1

- 1 2 . 372 - 1 2 . 404

3 . 467 3 . 462 N '-I

30 . 297 30 . 253

2 0 . 255 2 0 . 227

2 5 . 973 2 5 . 984

22 . 974 22 . 995

1 2 8 . 1 3 7 1 2 8 . 242

28

TABLE 6 : SALT CONTENTS OF VARI OUS C RUDE O I LS

CRUDE O I L ORIGIN

Mi dale (S ask . )

Leduc -Woodbend (Al t a . )

Sturgeon Lake (Al t a . )

Norman We l ls (N . W . T . )

Pembina (Alta . )

Redwat er (Al t a . )

Ros e lea (Man . )

Stett l er (Alt a . )

Smi l ey (Sas k . )

Aches on (Al t a . )

Rat c li ff (Sask . )

C antuar (Sask . )

Bonni e Glen (Al t a . )

Forget (Sask . )

Wape l l a (Sask . )

Success (Sask . )

Fos terton (Sask . )

Co l evi l l e (Sask . )

Virden (Man . )

Wi z ard Lake (Al t a . )

I vik (N . W. T . ) +

Atkinson Point (N . W . T . )·

GRAVITY

AP I

2 7 . 9

39 . 7

3 7 . 1

40 . 8

3 7 . 3

34 . 7

35 . 5

2 7 . 2

3 3 . 0

36 . 6

3 1 . 3

20 . 3

4 2 . 5

3 1 . 4

2 6 . 5

2 1 . 1

2 4 . 1

1 3 . 8

3 2 . 6

3 7 . 2

*

SALT AS NaC l

lb . 1M bb1 .

1 . 5 8

2 8 . 6

1 4 . 8

2 . 3

1 . 2

40 . 8

N I L

32 . 6

1 5 . 1

1 4 . 2

1 9 3 . 0

5 7 . 0

1 . 8

2 3 . 3

N I L

4 . 0

1 . 0

64 . 0

14 . 5

1 . 0

5 . 9

80 . 0

* These analyses are from the mid 5 0 ' s . These fi gures change w ith time .

+ From re cent analys es .

2 9

SC H EM AT IC OF MEASU RE M E N T PA R A M E T E R S

I N T H E S E SS I L E D ROP M ET H O D

F I GUR E 1

30

Figure 2. Close-up of Camera and Lens.

Figure 3. Camera and Tank in Test Position.

+ I -

c '- 0 ) -' w > -w -' 0:: _ w

I

�

I � � -3 -'

l <t - 4-Z � -0:: j o , .-: - 6 a:: , � - 7 Z o _

.... <t

I

, U - 9 o -' I 0::- 1 0 �

• , I • , : i

I

; " ;

T Y PICAL TEMPERATU R E D I S T R I B U T IO N I N T E ST TA N K

b RIGI NAL WAT E R I L E V E L k - .1. x , ,

_. f--- -i • X 0 ! o j I

, x _ , • , BOT TOM OF ICE i

e<� I I i I I , ! , , .K 0 I i i ,

l i : I � \ 0 I I • x 0 I

I � �v i Ie x 0 I I I

I I � - 1 I t-- . T E M PERATURES W I T H TA N K I N COLD ROOM I 1 1 05 AM I I 0:: w J: - 12 -I-

- 13 18

x T E M PE RA T U R E S W I T H TAN K O U T O F COL D ROOM AT STA RT O F T E ST I 1 2 40 PM - I-.

o T EM P E RAT U R ES W I T H TA N K ? UT OF COL D ROOM AT E N D OF TEST I 1 :55 PM I I I I L J J I

20 22 £4 26 28 30 32 TEM PE RAT U R E (O � )

34 36 38

I .2 �

-

i !

I I - -i w

I I I i I -! , ----- ,

I

I i

I

40 42

F I GURE 4

32

F i gure 5 . Bubb l e No . 6 o f test on March 2 7 , 7 0 s econds after inj ect ion ( 1 / 2X R eproduct ion Ratio ) .

Fi gure 6 . Bubb l e No . 6 of test on March 2 7 , 2 5 . 5 minute s after inj ection ( I X Reproduction Ratio ) .

w ::::> ....I

90

80

70

� > 50 Q w '" ::::> ." � w � AO

30

20

10

3 3

CALIBRATION CURVE FOR F ISH E R TENSIOMAT SCALE READING

V

AVERAGE SLOPE : 1 .006645 V

V

V

V

/

MASS CALCULATED INIT. CAl. GM DYNE/CM DYNE/CM.

0.19962 16 .275 16,5

. 0.39721 32.385 32.6

V

/ 10 20

0.67886 55 .348 55.9

0.99343 80.995 81 .7

30 40 50 60 CALCULATED VALUE ( dyne. /em.)

FINAL CAl. DYNE/CM.

16.4

32.5

55.6

81 .5

70

V

80 90

F IGURE 7

34

.

940

r----"-----,...-c-o-RR"'T"""�

C-T-lo-Jr---

FA-C-rT�-R-F-o"'T"""�---r-----r--/-+-----­

.932 1----+- TENSIOMAT READI NGS ---+-/--+I'-F---+--�

.92

4 �--+----+---+----+---+---t-/

-+-

.

--+----I---�

/ . 916 J-----+---+---t----+---+

V

-

.

-7---+----+----+----1

/ '90a ��--r-;/ --+--------f--+--------f

.900J-----+---+---t---....-ht----I----+----f-----+--� ��.2�-�1--�-������-�--�-�-�-� Q / �.aur----+---+-/--+-t-----+ CORRECTION FACTOR EQUATION :: I F : 0 725 +J 0.01452 P + 0.045l4 _ 1.697r 8 I . CI (D-d) R

.876 I WITH C = 6.01 e m . / R I. • 53. 8384846

.868 1-----------+--/ --+--------f--+--------f--+--------f-----1 i I I

I

'860 I--j-¥------+-----+-----t----'-t--'----t------+---t-----'---i

852

V .UA �-�---+--_+--�--�---+_---r_--�--�

�%�--���---2�0-----��--�A�0 ----�50�---��--�7�0----�a�0 ----·�90 P/D- d FIGU�E 8

>-�

> 4:( IX C) U u.. -U w � \I)

3 5

.S55

I I I I I I I I VA R I ATION OF SPEC I F I C G R AV I TY

W I TH TEMPE R ATU R E

. 850 6 . '\

, \ '\. x �

o - NORMAN WELLS CRUDE _ r\. X - SWAN H I LL S C R U D E -.� .845

\\ \ \.

\0\ \ \ .840

\ 1\ A \ \ i\ \ \ 1\ \ \ .S35

\ \ 1\ f\ \ \

\ \ " 2 1\ .S30

1\ \ \ \ \ � \ \0

\ .825

\ \ " 1\ \X .820

.815 -10 0 10 20 30 40 50 60 70 SO 90 1 00

TEM P E R ATU R E (e F. ) F I G U R E 9

> I-

� IX C) U IL

U w a.. Vl

. .;

1 .028

1.026

1.024

1 .022

1 .020

1.018

1.016

1 .014

1 .012

1 .010

1.008 2 0

x

0

6

*

0

x

SpJC I F l lC x

U

0

6 i W

6

IW * *

-'-"

0 0

30 40

3 6

I I I I I I I J GRAV I T Y OF B R I N E SOLUT IONS

x x

0

0

6

6

**

1'l -

0

50 60 TEM P E R ATU R E (-F. )

o -* -II -o -X -

70

u -,.

1 2 Y ••

18 ·/_ 24 -/ .. 30 -/ .. 36 %.

--

...

,... 80 90

F I G U R E 10

1.024

1.022

1.020

1.018

1.016

> � > 1.014 < CC C) U i:&: 1.012 U w Q.. I/)

1.010

1.008

1.006

1.004

1.002

37

I V

SPEC I F IC GRAV I T Y VI BR I N E CONCENT RATI� AT 77 ° F. (250 C . ) , , ,

,

/ V

i

. (

/

V /

/ V

/

V I

/ V / j

5 10

,' ,

15 20 25 30 BR I N E CON C E NT RAT ION ( e/ee)

35 40

FIGURE 1 1

>­I-

38

AI R D E NS I T Y vs T E M P E R AT U R E

AT ATM OS P H E R IC P R E SS U R E [20J

� .O�� ____________ � w ·074 o

u..: .065 • ...... ......

. 060 0 20 40 60 80 1 00 120 140 160 180 TEMPERAT U R E ( O F. )

F I GURE 1 2

200

39

S U R FACE T E N S IO N F R E QU E N C Y DI ST R I B U T ION SWAN H I LLS C R U DE

45

40

35

30 -� o

o 25 Z w ::> C � 20 u..

15

10

5

-

l-

I-

I-

-

I-

....

I-

I-

o o

44.44 %

23.33 %

10.00 %

7.77 %

2.22 % 3.33 % 3.33% 3.33 % 2.22 %

J J 5 10 15 20 25 30 35 40 45 50

S U R FACE T E N S ION (dynes /em . )

F I GU RE 1 3

40

S U R FAC E T E N S ION F R E Q U E N C Y D I ST R I B U T IO N

N O R M A N W E L L S C R U D E

45 ,....

40 �

35 ""'"

3 2 .93 %

30 -

28.05 -'0

-

-

15.85 -'0 1 5 I-

1 0.98 -'0 10 l-

5 I- 3.66 -'0 3 .66 % 2.44 %

1.22 %1 1 .22 % I I

o 5 10 15 20 25 30 35 40 45 50 SURFAC E TENS ION ( dynes ' em. )

F I GURE 1 4

.,

+ ..

.

+ ... ... +

... ..

. +

+

+

+

41

+ ...

...

... ...

...

+ +

... ...

+ +

+-

+ ...

+ +-...

+ +

++ +

++ ...

... +-... ...

++ ...

... +-... +-

+-... + .

...... ++

..

. ..

FIGURE 15

+

4-

...

o

o

'"

6 ><

o

C"".I

6

01-0

r-

----

��

----

_r--

----

�--

----

r_--

--_,

----

--_r

----

--�

----

--�

----

_,--

--:-

_h6

o

o

.n Ll)

o

CI

.n ..

o

. CI

.n

CO>

o

o

.n N

W:lI AD A � X VWWV9

o

.9

Ll)

o

o

.n

§' z

........

X

ei.)

>

>-\:b

X

«

::E

::E

«

C!)

+ +

++i-. 42

+

FIGURE 16

+ +

+ ++

+

+ .,.

++ +

+-of-

+ ... +

+ +

+ ++

... + �

... "'" ... +� " '"!.. �

+

... ++ +

+

+

+

0

r-6 �

�

X

0

Ln

6

o

M

6 ;:

�--

--�

----

--�

----

� __

----

�--

----

�--

---T

----

--�

----

�--

----

'---

--�

6 o

o

an

..,

g an

.....

C)

C)

an

M

C)

C)

an

N

W:lI AO A '.3 X VWWV9

. 0

o

� o

o

an

55-00

45- 00

35-00 � U ......... > C > « ' � � « (.!J 25-00

1 5 -00 ++

+

...

GAMMA X VS S IGMA Y [SH]

+

/#' /�

j'/

� ... .,...+

...

.... + -t

+

"" ...

++

+

+

+

...

+

5·00 -+1 --�----r----Y---�----r----Y-----r----.----Y-----'

5·00 15 ·00 25·00 35·00 GAMMA X Dy/eM

45·00

"Tl

GJ c: ;;0 JT1

'-l

..". LV

>­«

�

(.!J

CI.)

CI.)

>

x

«

:E

�

«

(.!J + +

oj. +

44 + +

..

FIGURE 18

+.j.

<:>

o

Lb U")

o

o

Lb:::2:

MU

..

...... >­C

x

«

:::2: :::2:

0«

c:;:>

C!)

U")

N

o

o

Lb

oj. 0

�----�------�------�----�r-----�-------.�----'-------r------.------;-%

o

o

Lb U")

o

o

Lb

..,

o

o

Lb N

W31 A[J A '1WW'19

o

o

Lb

0.6

0.5

0.4

E u

1>- 0.3

0.2

0.1

"T1 G) L ___ c 0 ::>0 0 m

-0

._- _ .... _ _ ._ ..

// -- 0-'--------/ 0.1 0.2

.. _ ..

I I

.;.

GR APH OF Y vs. X FOR SWAN H ILLS C R U D E ( FOR ALL TIMES )

x

1-.------ -.-. . --.- .. . - --.- . �� --- -. -._ .

xy..----'p- 'X

...- x D

l�-'! x ! 0

x -- 0 ---' AT X= 1 .80

'" _ .. _-

I

-- --- --J --><-- �A 1 _._.. -.--- , . --.- - - . - i i

. ._-

!

,�r )()( / x / ..,lEx

0.3

I � xx�

/ �. I-----�-.-.-

0.4

x O B S E R V E D DATA . " CALC U LAT E D F ROM REGRESSION EQUAT ION Y = : 0.0 1 1 7 191 + 1 . 165253X.- 0.83929045( 2+ 0.2104865X i! o OBSERVED DATA FROM MOV E M E NT STU DIES.

I --- ·-·---- --

T

.

l �m 1 . I I

I I 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1 . 2 13

X (em.)

-P> Ul

0.7

0,6

; O,S

0.4 I -: E

,-; 0.3

0,2 �

o. I

I 0

0

G R A P H ' OF V V S . X F O R N O R M A N W E L L S C R U D E ( FO R A L L T I M E S )

0 0

--� r-s..... V l- O U � I--� ""

0 0 0 _I---I---� r- 0 � '\

,"""" � X ' 9 � .... ./ D 1\ 0

/ -

,.-vJ 1\ I), V · �

X OBSERVED DATA

0 CALCULATE D FROM REGRESSION EQUATION Y= 0,00!530755 + 1 .027235X- 0.53688281(2

; • BASED ON OBSERVED DATA X , I I I I I I 0 OBSERV E D DATA FROM MOVEMENT STUDIE S , 1/0 6 CALCU LATED FROM REGRESSION EQUATION y = 0 , 04197806 + 1 .395072X- I,400440X%+ 0, 6470588� - 0, 1076159r

BASED ON OBSERVE D DATA X AND 0 USED FROM C A LCULATIONS ,

I I I I I I 0.1 0,2 0,3 0,4 O.S 0,6 0,7 0,8 0,9 1.0 1 . 1 1.2 1.3 1.4 I,S 1,6 1.7 1 .8 1.9 2.0 2.1 2.2 2.3 2,4 2.5 2.6 2.7 2.8 2.9 3.0 Xlem.l

Fi �:u re 20

� m

47

1,000 -G RA P H OF (3 vs. X / Y

-- - - - - -1--------- -

------- -

100 .v / / L /" JV

/ /'

/ V

/ V / Qt 10

/ )II / V

/ / / V

I I I J r I

/ / 7

/ \0 1 .2 1 .4 1 . 6 1 .8 2.0 2.2 2.4 x / v FIGURE 21

48

1 .0

�

\\

G R AP HS OF ' F ! AN D G 2 vs. X I Y

\ \ \ \

\ \ \. �

.9

. 8

. 7

.6

.4 \ 1\ \ \ \ 3 f\

\ � \ � 2 .� � � '" 1

" x..... � � !--x_ x -

01 1.2 1 .4 1 .6 1 . A 2.0 2.2 2.4

x / v F IGU R E 22

AO

36

-E 32 u

...... ; � 28 '""tI

Z 2.4 Q V) z � 20 w U « 16 u.. a.: ::> V) o 12 w I-� 8 ::> U � « .4 U

o

- 4

- 8

- 12 o

49

GRAPH OF ERROR L I M ITS F RO M TAB L E 5 vs

CAL CU LAT E D S U R FAC E T E N S IO N S

i at 128.

0

• )(

+

$

. 1 .2

:

I • 1 J I I

CALC U L ATED US I N G REGRESSION VA L U E FOR Y -M A X I MUM N E GA T I V E E R ROR

M AX I M U M POS I T I V E E R ROR

I NT E R M E D IATE E R ROR

.3 .4 .5 6 .7 X (em. )

.

. 8 . 9

- - ---

i : I

i 1

I

i I ! I I

1 .0 1 . 1 1 . 2

F I G U R E 2 3

/

-

E u

...... '" •

37

35

3 1

� 27 .:!!. z o V) Z � 23 w U � c.:: :> V) 0 19 w .... � ..-j :> u ..-j

� 15

I I +

0 0

50

GRAPH OF SURFAC E T E N S ION vs.

X FOR SWAN H I L L S C RUDE

VAL U E S CALC ULAT E D U S I N G T HE REGRESSION EQUAT ION FOR V I N TERMS OF X F ROM F IGURE 19 T H E DI F F E RENT C U RVES R E P RESENT D I F F E R E N T VALUES OF T H E CONSTANT C USE D IN THE EQUATION A S INDICAT E D

C XZ

y = �

+

I� 0 I

0

0 a +

+

0

0

.2 .4

+

0

0

C = 140 C = 160 C = 180

+

0

0

.6 .8 i ( em.)

+

+

0 0

0 0

1.0 1 .2

F I GURF 24

..

0

0

1.4

51

G R A P H OF S U R FAC E T E N S I O N v s .

32 r------ X FOR NORMAN W E L L S C R U D E -___ --.

VA L U ES CA L C U L AT E D U S I N G T HE R E G R E SS IO N E Q U AT ION FOR Y I N T E R M S O F X F RO M F IG U R E 20

T H E D I F F E R E N T C U R V E S R E P RE S E N T D I F F E R E NT VALUES OF T H E

E 2 8 t- CONSTANT C US E D I N T H E E Q UAT ION AS I N D ICAT E D u C X 2

� Y = {:3 F2 c:. >­

."

+ I-i I

Z 24�------T---�--+-------�-------+-------4-------�--� o (/') z W I-w U 22 �------r-------�--�--�-------+-------4------���

� c.::: ::> (/') c w � 1 6�-

-----r-------+-------�-------+-------4------�--� ..J ::> U ..J « u

1 2 �------�-------4--------�-------+-------4--------�--0 C = 140 x C = 1 6 0 I

+ + C = 180 x

8 0

4 I

� 0

0 .2 . 4 . 6 .8 1 .0 1 . 2 1 .3 X (em. )

F IGU R E 25

.--.

::r::

CI)

.........

I­::r::

C!J

j::ij

::r::

CI)

>

>-

+

+ 52

FIGURE 26

+

+ + +

+ +

+

o

co

6 o

co

6 0

N

6

o

r-----��-----r------�------r_----�------_r------,_------r_----�------_r;6

o

co

6 o

co

6

o

N

6 o

o

6

53

o

o

o

co

o

o

�-------r--------�--------�------�---------r--------'---------r--------4_0

a

co

a

o

N

6

o

o

'? a

.--.

::::t:

CI)

"--'

>-..........

X

CI)

:::-

t-::::t:

C!J

w

::::t:

+ + :t ... + +

..

. +

++

.1-�++� �

+ +

54

+

+ ... "" +

$I-++

+

+ FIGURE 28

+ o!-+

>­........

><

o

!

+ +-

0

�------�----��----�-------T------�------�-------r-------r------�------��

o

CX)

6 o

cc

6 [W3] lH913H

o

..,.

6 o

N

6

o

o

6

55

FIGURE 29

§' 0

�

z

........

>->< >-"-

CI.)

x

>

r-:::z::::

0

(.!:)

1: w

::

:z::::

#-+--+

.. t 0

+ �

+ ..

+ +

+ + + t+++ ..

....

+;:::+ + .. +

+ 4-• +++

++

�

+ +-

+ ++ +

+

-+++ 0

�

0

0

0

0

0

0

co

co

... '"

0

2

6

6

6

.6

6

[W3] lH9I3H

70

66

62

r----

,� "

56

-:- 54 E u ........ en Q) c: >-

� 50 Z o V) Z � 46 w U il! a:: ::) 42 V)

38

34 -.--

56

x , J- --

x

! B R I N E AGAI N ST A I R

x x

S U R FAC E T E N S IO N vs . BR I N E CON C E N T R AT I O N -F ROM R I N G D E TATC H M E N T M E T H OD

v v B R I N E S AT U RATE D WITH SWAN H i l LS O IL AGA I N ST A I R

'J -v V

I�

Ifl � � � b

V--..... �

� I-'" ----n ------ r!J

BR I N E SAT U RAT E D W I T H NOR M A N WE LLS O i l AGA I N S T Oi l AGA I N S T A I R

[

28 �----�----�--- �-----�-----4�----�----�--�

SWAN H I LS O i l SAT URATE D W I T H BR I N E AGA I N ST I R

24�----�----�----�----�----+-----��--+---�

NOR MA N WEllS Oil SAT U RATE D W I TH BR I N E AGA I N FT A I R

20 �----�----�----�----�----�----�----�--� o 5 1 0 15 20 2 5 30 35 40

B R I N E CONC E N T RAT IO N (%0 ) F I GURE 30

57

SC H E M AT I C OF T H E FORC E R '

C AU S I N G MOT I O N I N T H E X D I R E C T I O N

R y

WAT E R : Pw

Pw R = ( - - 1 ) Mo 9 Po

R ' = R sin a

F I GURE 3 1

2.0

1.8

1.6

1.2

j II> 1.0 II> <I: �

0.8

0.6

0 .•

0.2

o

�� �/" x �

-� 0.1 02 0.3 0..01

58

I I BUBBLE MASS vs X

/ / iJ'/ �

0.5 _ 0.6 X lem.)

j/

0.7

I

I I

/ xl xl V

0.8 0.9 1.0 1.1

F I GURE 32

100

1 0 )

E u

1><

I"" �./ I.---

........ f ..... � ..........

k k V

.�' ..... 1 · :01 .1

G R A PH OF X vs . MASS

./

,,;0 �.I"'" ........-V

./ �I---I--7 7 ./

� i;

P-': I---� V--

V � I:::: � t:::1-'

� .,..

)( CALC U LATE D ACCORDING TO ME T HOD GIV E N IN ( 10) • EXT RAPOLAT ED FROM F IGURE 32 I o VOLUME OF C YL I N DE R OF RADI U S X A N D I Cr iiUtLllrfll UM HjI G,T· 1 I I I I I I I

10 100

MASS (gm.)

F I GURE 33

1000

U1 <.0

60

MASS , S I N a vs M ASS FOR SWA N H I L L S C RUDE

10

J,... x V v

k v 0 V � �

./ _x :11

....

L..r ; � v/ V

> v

� �V' x

.... � 0

v � V V ./

./ . 1 o F ROM MEASUR E D BUBBLES

•

x F ROM VOLUME ESTI MATES

10 100

MASS ( gm.)

F IGURE 3.4

10

E til -a

Z V) . V) II') « :E

. 1

/' V

61

M A SS , S IN a vs M A SS FOR N OR M A N W E LLS C R U DE

� � y v-0

/ V 0 ./ /'

./ 0 ./ � v

0 0 ...".'" c /' 0 I ./

x y / K n

O � V ..... � X

� v V v o F ROM M EA S U R E D B U B B L ES

./ I X F ROM VOLU M E EST I MAT E S

...".'/ � V 0

10

10 100 MASS ( gm.)

F IGURE 35

62

FORC E TO I N I T I ATE MOTI ON v s MASS FOR SWAN H I LLS C RU D E

100

." QI c

0

�O 0 z Q I-o :E w � I-Z

� IIV w U 0:: 0 1 LL

0

1

.......... ./ �

...... V

.

x x / / ............... x

.... � 0

F = 48. 5 M 0.486 f

I'

..... � x � ......

./ c/ Vx :It

0 ..� pX ...... �

� ./ � � . .,...

o F ROM MEAS U R E D BU BBLES

x F ROM VOLUME ESTI MATES

I 10 100

MASS ( gm.)

F IG U R E 36

63

FORCE TO I N I T I AT E MOT ION vs MASS F O R NORMAN W E L L S CRUDE

1000

-;; QI c

�lao z o � o � w � !f t: Z e w u 0 10 u..

1 .1

1/ ./

./

� I .....

V V

V V V Pc

/' 1 0 I n

l...-V ;-

V 7 1 4 0 ./ c V V x

K X

0

, 0 0 ,

a ..-/"

./ I n O /' " � /-' ...,

) X

I...-� F = 23.4 M 0.659 0

, o F ROM M E A S U R E D B U B B L ES , X F ROM VOLUME E ST I MAT E S

I 10

MASS OF O I L (gm . )

F IGURE 37

100

64

P ROF I L E OF AN O I L L E N S O N WAT E R

WAT E R

F I GURE 38

60 -

55 I-

50 �

45 45.5 %

40 l-

I-

�

I-

65

H I ST OGRAM OF SALT CO N T E N T

F ROM 22 LOC AT I O N S

X 52 5

M E A N : 27. 1 I b l VAR I A NCE : 1845.

M bb l 75

42.96 S T D. DE V I AT IO N :

20 � 18 .2 %

1 5 -

10 -

5 i-

o o 10

9. 1 %

4.5 % 4.5% 4.5 % 4 .5 % 4.5 %

20 30 40 50 60 70 80 90 SALT CON T E N T ( I b l M bbl )

4.5 %

I J , 100 �OO

F I GURE 39

66

APPEND I X A

DEV ELOPMENT OF THE EQUAT I ON FOR THE PROF I LE OF A S ESS I LE DROP

67

APPEND I X A . DEVELOPMENT OF THE EQUAT I ON FOR THE PROF ILE O F A SESS I L E DROP .

y

¢ II I

WAT E R D2 Pl D l o P2

Con s i der fi rst a po i nt A on the s urface of the bubbl e w i th P l the pres­s ure on t he concave s i de of A and P 2 the pres s u re o n t he convex s i de . Let Dl be the dens i ty on t he l ower s i de and D2 that on t he upper . Let Pl - P2 = C , C be i ng determi ned by the curvature at A . Now the pre s sures at P wi l l be

P l - gD2y and P 2 - 9D 1 Y where P i s a po i nt on the su rface at a l evel y above A .

1 1 LlP Now from ( 9 ) -R + -R l = Yo/w

where R and Rl are the pri nc i pa l rad i i or curvature at any po i nt , then at po i nt P

or 1 1 R + Rl =

P l - P2 + gy ( D l - D2 ) Yo/w

. . . (A- l ) Let x be the hori zontal and y the vert i ca l coord i na te of any po i nt i n a mer i d i o na l s ect ion o f the s urface of the fl u i d , r the rad i us of curva­ture of the meri d i ona l sect i o n at that po i nt , and � the a ng l e whi c h the normal to the s urface ma kes w i th the axi s of revo l uti on ( i . e . , y axi s ) . Then the l ength of the norma l termi nated by the axi s i s x/ s i n� and

00

0 R = r and Rl = � s l n� Eq . ( 1 ) then becomes

. . . ( A- 2 )

68

Let b be the rad i us of curvature at the o ri g i n , so that at that po i nt we have r = b and l im ( x/ s i n� ) = b

00

0 when y = ° one gets

� = f or C = � b y b Substi tuti ng i nto ( 2 ) we get

Let

= 2 + gb2 ( Pw - Po ) (f) Yo/w

- s

Equa t i o n ( 3 ) then s i mpl i fi es to

. . . (A- 3 )

. . . ( A-4 )

1 + s i np = 2 + sY . . . (A- 5 ) r x Al so , when � = 0 , Y = 0 , r = and l i mi t (s�n�) = 1

d 2y

and

and

dx2 r -{l + (�l 2}

.QY dx

3/ 2

s i n� = ;;-------:----{l + (�) 2} 1 / 2

cfi7 d2-l etti ng Ql = y l and � dx dx2 = y "

Then equat ion ( 5 ) becomes

or

y l -

+ [ 1 + (y l ) 2J l/ 2x - 2 + sY

Y II = ( 2 + Sy ) [ 1 + (y 1 ) 2 J 3 / 2 - � [ 1 + (y 1 ) 2 ] X

. . . (A-6 )

69

APPEND I X B

TEST RESULTS FROM R I NG DETACHMENT METHOD

70

Abbrev i at i ons u sed i n t h e fo l l owi ng tab l e are : a , A

m , �

nc

The a u tomati c mode was u sed for ten s i oma t operati on . The ten s i omat wa i operated manual l y . I nd i cates that the a ppara tus wa s not c l eaned between tests .

Notes to the fo l l ow i ng tabl e are : +

*

From data i n Ta bl e 3 . P i s the cal i brat ion co rrected val u e of the meas ured tens i on i n dynes/em .

- Ta ken from F i g ure 8 .

I F LUI D TYPE

Brine 6%

6%

1 2 %

1 2 %

18%

18%

24%

- 24%

24%

24%

30%

30%

36%

36%

36%

36%

N . W. /S6%B

� . W/S 6%

I TEHP . °c

26 . 65

2 7 . 2

2 6 . 65

26 . 25

2 6 . 5

-

25 . 95

-

-

-

-

26 . 85-

26 . 7

-

-

-

2 7 . 15

TEST RESULTS FOR RING DETACHMENT METHOD

TEST VALUES ON TENSIO��T DYNES/CM AVG :OP-'rEST I CALiBRATION

TEST 2 TEST 4 TEST 5 ' VALUES CORRECTED TEST 1 TEST 3 DLICM_ YALULP.YL01 67 . 6 (m 6 8 . 4 (m 6 8 . 3 (m) 68 . 0 (m - - 6 8 . 075 6 7 . 626

70 . 7 (m 7 3 . 6 (m 74 . 7 (A) 75 . 2 "(A 75 . 4 (A) 73. 92 7 3 . 4 32

68 . 7 (m 69 . 3 (m 6 8 . 8 (m) 6 8 . 7 (m - 6 8 . 875 6 8 . 42 0

75 . 4 (m 76 . 0 (A 76 . 0 (A) 76 . 0 (A 75 . 5 (m) 75 . 78 75 . 2 8

71 . 2 (m 69 . 6 (m 69 t:ffi) 6 8 . 7 (m - 69 . 70 6 9 . 240

65 . 5 (m 72 . 8 (A 74 . 0 (A) 76 . 8 (A 75 . 5 (m) 72 . 9 2 72 . 439

7 3 . 6 (m) 7 3 . 7 (M 73 . 5 (M 73. 8 (M 73 . 65 7 3 . 164 -

66 . 9 (m) 6 8 . 2 (A) 7 3 . 5 (A 75 . 9 (A 75 . 9 (M 72 . 0 8 71 . 604

76 . 3 (A) 76 . 4 (A 7 6 . 4 (A 76 . 5 (M 76 . 1 (M 76. 34 7<; R7;f>

76 . 0 (m) 75 . 9 (M 76 . 4 (A 76 . 4 (A 75 . 9 (M 76 . 12 75 . 6 1 8

7 3 . 8 (M) 7 3 . 9 (M 73 . 9 (M 73 . 8 (M -7 3 . 85 73 . 363

7 1 . 9 (M) 75 . 0 (M 75 . 7 (M 76 . 4 (A 76 . 2 (A 76. o iA

75 . 2 74 . 704

7 3 . 2 (M) 73 . 0 eM 73. 2 (M 7 3 . 1 (M 7 3 . 125 72 . 642 -

66 . 8 (M) 7 1 . 6 (A 75 •. 1 (A 76 . 4 (A 75 . 8 (M 76 . 4 (A

7 3 . 68 73 . 194

7 1 . 5 (M) 75 . 2 eM 75 . 7 (M 75 . 8 (M 75 . 7 (M 74 . 78 74 . 2 86

75 . 8 (M) 76 . 0 eM 76 . 2 (M 76 . 0 (M 76 . 6 (A 10 . 1 1 75 . 667 7 6 . 4 (A

24 . 8 2 4 . 8 25 . 0 2 5 . 4 (A 25 . 0 25 . 00 24 . 835

25 . 0 (M) 25 . 6 (A 25 . 6 (A 25 . 6 (A 25 . 0 (M 2 5 . 19 25 . 36

D-d +

GM . /CW

1 . 0024

1 . 0024

1 . 0065

1 . 0065

1 . 009-5

1 . 0095

1 . 0146

1 . 0146

1 . 0 146

1. 0 1 46

1 . 0 1 84

1 . 0 1 84

1 . 0224

1 . 0224

1 . 0224

1 . 0224.

0 . 8229

0 . 8229

P / (D-d) :f DY . 012 I GM 67 . 4617

73. 254

6 7 . 9761

74 . 79 1

6 8 . 5 861

71 . 755

72 . 1089

70 . 5 72

74 . 74 3

74 . 52 8

72 . 0 350

73: 352 ]

71 . 0489

71 . 5 89

TL. . 0';) 7

74 . 00 7

�0 . 1 790

�0 . 610

ORRECfION �����6� I FACTOR* DY . l ot 0 . 9280 62 . 7 &

9 . 9339 6 8 . 5 7

0 . 9286 63 . 5 .3

0 . 9353 70 . 4 1

0 . 9292 64 . 3 4

0 . 9 324 67. 54

0 . 9 325 68 . 2 3

0 . 9 312 66 . 68 -....J --'

0 . 9 352 70 . 92

0 , 9 350 70 . 71

0 . 9320 68. 37

0 . 9 339 69 . 77

0 . 9 3 1 7 67 . 6 8

0 . 9 32 2 68. 2 3

0 . 9 332 69 . 33

0 . 9345 70 . 71

0 . 8871 2 2 . 03

0 . 8877 2� . 36

I I TEHP . I F LUI D TEST VALUES ON TENSrOHAT DYNES/Ol

i TI'PE °c _ TESJ 1 TEST 2 TEST 3 TEST 4 [rEST 5

N . W . /S 1 2%1 27 . 8 24 . 8 25 . 3 (A 25 . 3 (A) 24. 5 (M ) 25 . 1 (A)

N . W . /S 18!li 27 . 65 25 . 4(A) 25 · 4 (A 24 . 8 (M) 25 . 0 (111 ) 25 . 4 (A)

N . W . /S 24%1 25 . 8 25 . 1 (M) 25 . 7 (A 25 . 0( M) 25 . 8( A) 25 . 7 (A)

N . W . /S 30% 25 . 95 27 . 9 (M) 28 . 5 (A 28 . 6 (A) 2 7 . 9 (M: 28 . 0 (M)

N . W . /S 30% 27 . 1 28 . 0 (M) 28 . 6 (A 28 . 5 (A) 28 . 6 (A 28 . 1 (M)

N . W . /S 36% n . 1a 28 . 0 (M 28 . 6 (A 28 . S (A) 28 . 7 (A 28 . 2 (M)

N . W . /S 36%1 26 . 2 27 . 6 (M 27 . 9 (A 28 . 0 (A) 27 . 5 (M 27 . 5 (M)

6% B/S N . W 26 . 85 46 . 4 45 . 0 44 . 4 43 . 5 - - -

1 2 % B/SN . W 25 . 9 40 . 2 39 . 7 36 . 6 3 7 . 0 - - -

2% B/S N . W 26 . 7 49 . 1 (M 49 . 4 (A 50 . 6 (A) 50 . 9 (A 50 . 8 (� 8% B/S N . W 26 . 0� 47 . 7 46 . 8 46 . 6 46 . 0 - - -

�4%B/S N . W 26 . 4 45 . 2 4 3 . 5 4'; ,) 42 . 8 - - -

174% B/S N . W 26 . 7 48 . 1 (M: 56 . 7 (A) 55 . 6 (A 55 . 5 (A 5 5 . 4 (�

30% B/S N . W 26 . 5� 50 . 2 . 49 . 2 48 . 8 48 . 7 - - -

36% B/S N . W 26 . 8 50 . 8 50 . 7 (11 50 . 6 52 . 0 (A) - - -51 . 9 (A) 51 . 4 (M)

S . H . /S (,6% B - - - 27 . 7 (M 28 . 4 (A) 28 . 4 (A 28 . 4 (A) 27 . 7 (M

S . H . /S12% B 27 . 9 27 . 7 (M 28 . 3 (A) 28 . 3 (A 28 . 4 (A) 2 7 , 8 (M

S . H . /S18% B 28 . 1 2 7 . 3 (M 2 8 . 1 (A) 2 8 . 1 (A 2 8 . l (A:) 27 . 6 CM:

AVG�.OF-tESTI CALI1lRATrON D- d + o VALUES 0 1 CORRECTED

GM. lcW -DL/CM._ XALUL!2YLCM 25 . 00 24 . 835

25 . 20 25 . 034

25 . 46 25 . 292

28 . 1 8 27 . 994

28 . 36 28 . 17 0 . 8.229

28 . 42 28 . 23 0 . 8229

27 . 70 27 . 5 17

4 4 . 825 44 . 529 1 . 0024

3 8 . 375 38 . 12 2 1 . 0065

50 . 16 49 . 83 1 . 0065

46 . 775 46 . 466 1 . 0095

43 . 625 4 3 . 337 1 . 0146

) 54 . 26 5 3 . 90 1 . 0 146

49 . 9raS 4 8 . 900 1 . 01 84

51 . 233 50 . 895 1 . 0224

28 . 1 2 27 . 93 0.,8 174

28 . 10 27 . 91

2 7 . S4 ,21 . 66

PI CD- d) :j:: DY . c:W /GM

30 . 1790

30 . 4209

30 . 7344

34 . 0178

34 . 232

34 . 305

33. 4 382

44 . 4208

3 7 . 8745

49 . 506

46 . 0272

42 . 71?1

5 3 . 123

4 8 . 0152

49. 7787

34 . 168

34 . 144

3 3 . 8 38

ORRECI'rON FACTOR*

0 . 8871

0 . 8874

0 . 8878

0 . 8918

0 . 8921

0 . 8922

0 . 89 1 1

0 . 9039

0 . 8964

0 : 86 38

0 . 9057

0 . 9020

0 . 9 134

0 . 9079

0 . 9098

0 . 892

0 . 892

0 . 892

1 ����t5� DY 10\ 22 . 03

22 . 2 2

22 . 45

24 . 97

25 . 13

25 . 19

24 . 52

40 . 2 5

34 . 1 7

46. 56

42 . 08

39 . 09

49 . 2 3

44. 40

46 . 31

24 . 9 1

24 . 89

24 . 66

I

'-l N

I FLUID TEHP . TEST VALUES ON TENSIOHAT DYNES/CM TIPE °c IE.ST 1 TEST 2 TEST 3 TEST 4 �EST 5

S . H . /S 24%1 2 8 . 05 2 7 . 8 (M) 12 8 . 4 (A) 28 . 5 (A) 2 8 . 4 (A) 2 7 . 8 (M)

S . H. /S 30%1 2 8 . 1 2 7 . 8 (M) 128 . 5 (A) 2 8 . 4 (A) 2 8 . 4 (A ) 2 7 . 8 (M)

S . H . /S 36'1 2 8 . 1 2 7 . 8 (M) 2 8 . 5 (A) 2 8 . 4 (A) 2 8 . 4 (A 27 . 7 (M)

6%8/S S . H. 27 . 4 5 8 . 3 (M 5 8 . 2 (M) 60 . 0(A) 60 . 9 (A 61 . 7 (A)

1 2%8/5 S . H 2 7 . 4 5515 (M 55 . 1 (A) 55 . 9 (A) 56 . 4 (111 56 . 1 (M)

1 8%8/S S . H. 27 . 5 56 . 6 (M] 57 . 3 (A) 57 . 4 (A) 57 , 8 (A 57 . 8 CM)

24%8/S S . H 27 . 55 58 . 7 (N:: 59 . 7 (A) 6 0 . 5 (A) 60 . 9 (A 6 1 . 0 (M)

30%8/S S . H 27 . 65 56 . 9 (M 57 . 3 (A) 5 7 . 7 (A) 58 . 4 (A 57 . 9 (M)

36%8/S S . H 2 7 . 75 56 . 7 (M 5 6 . 9 (A) 5 7 . 1 (A) 5 7 . 7 (A 56 . 9 (M)

N . W. 26. 9 2 7 . 5 (A 2 7 . 4 (A) 26 . 8 (M) 26 . 8 (M 27 . 5 (A)

S . H . 26 . 5 28 . 0 (M 2 7 . 7 (M) 2 8 . 4 (A) 2 8 . 4 (A - - -

CENTRIFUGE S . H . /S 24% - - - 2 7 . 8 (M 2 8 . 4 (A) 2 8 . 3 (A 2 8 . 5 (A 2 7 . 8 (lI DISTI LLED

H .. O 28 . 2 72 . 3 (M 7 3 . 3 (M) 14 . 1 (M 73. 9 (M 73 . 9 (M) DISTILLED

H'JO 2 8 . 2 74 . 4 (A 74 . 4 (A) 74 . 4 (A 74 . 4 (A 74 . 5 (A) DISTI LLED

HJl 27 . 5 74 . 6 (M 74. 7 (A) 'Z4 . 4 (A 'I5 . 2 (A 74 . 6 (M)

6%8/8 S . H . 24 . 7 57 . 0 III 5 7 . 1m- 5 7 . 9a- 58 . 1a 57 . 5m·

1 2%8/S S . H . 25 . 6 5 1 . 4m 5 1 . 9a 5 2 . 0a- 5 1 .9�; '5H6n�

1 8%8/8 S . H . 26 . 65 55 . 4 m 56 . 2a 56 . 4a 56 . 4a 56 . 1m.

AVG . OP--rEST1CAL!BRATION D-d + " VALUES " \ CORRECTED GM . /cW -DY./-CM_ Y_ALJ,JLP.YLCM 28 . 1 8 27 . 99

2 8 . 1 8 27 . ,99

28 . 1 6 27 . 97

59 . 82 59 . 425 1 . 0024

55 . 862 55. 432 1 . 0065

57 . 38 57 . 001 1 . 0095

60 . 16 59. 763 1 . 0146

57 . 64 57. 260 1 . 0184

57 . 06 56. 683 1 . 0224

27 . 20 2 7 . 020 0 . 8245

2 8 . 125 27 . 9393 0 . 8190

2 8 . 1 6 27 . 9.741 0. 8 190

73 . 50 7 3 . 0148 0. 9988

'14 . 42 4 3 . 9287 0 . 9988

74 . 7 74 . 207. 0 . 9988

57 . 52 5 7 . 14 1 . 0024

5 1 . 82 5 1 . 48 1 . 0065

56. 1 0 55 . 73 1 . 0095

P / (D- d) :f DY. c:M1 IGM 34. 242

34 . 242 "

34 . 2 1 7

59 . 281

55 . 072

56. 463

58. 901

56. 2 24

55 . 440

32 . 7719

34 . 1 1 39

34 . 1564

73. 1014

74 . 0164

14 . 295

57. 00

51 . 15

55 . 20

,",ORRE Cf ION FACTOR*

0 . 892

0 . 892

0 . 892

0. 920

0 . 91 6

0 . 917

0. 920

0 . 9 1 7

0. 916

0 . �903

0 . 8919

0. 8920

0 . 9337

0 . 9345

0. 9348

0 . 9 1 76

0 . 9 1 1 2

O. 156

TENSION DY . /OI. 24 . 97

24 . 97

24 . 95

54. 66

50. 75

52 . 27

54 . 9 5

52 . 49

5 1 . 92

24 . 06

24 . 92,

24 . 95

68 . 17

69 . 09

69 . 37

5 2 . 43

46 . 91

5 1 . 03

I

-......J W

I , I F LUID I TEHP . TEST VALUES ON TENSIOHAT DYNES/Ol . TYPE °c T.ESJ'-.l �T 2 �ST 3 TEST 4 trEST 5

24%B/8 S . H 25 . 75 59 . 4 m 60 . 1 ac 60 . 3ac 60. S ac 60 . 0mc

30%B/S S . H 25 . 2 59 . 2m 59. 1m 59 . 9a : 60 . 9rn 60 . . 8rn

36%B/S S . H 25 . 8 56 . 1m 56 . 8a 5 6 . 7a 57 . Ga, 55 . Bm.

AVG . OF-TEST' CALTT3RATION D-d + , VALUES I CORRECTED G�l . / CW -IlY../CM_ YALUL R.YLCM .

60 . 06 59 . 66 1 . 0 1 46

59 . 9 8 59 . 58 1 . 0 1 84

56 . 62 56 . 25 1 . 0224

P / CD- d) :j: ORRECTION DY . c�t2 /GM FACTOR*

58. 80 0 . 9 194

58 .·50 0 . 9 1 9 1

55 . 02 0 . 9 1 54

! �����6� DY fO!.

5 4 . 85

54 . 76

5 1 . 49

I

'-I +=::0

7 5

APPEND I X C

DATA FOR CALCULATI NG THE S PREADI NG COEFF I C I ENTS

76

APPEND I X C I n prepa ri ng the fol l owi ng ta bl es , val ues of the surface tens i ons o f the pure l i q u i ds ( ya and Yb ) a n d of the mutua l l y saturated l i q u i d s (Ya ' and Yb ' ) have been ta ken from t h e best l i nes through the data i n Fi gure 30 . The val ues s hown have been corrected to account for the 5 percent error i n the average val ue of the mea sured surface tens i on of d i sti l l ed wa ter ( 68 . 88 dynes/em ) compared to an average va l ue o f 72 . 44 dynes/em as reported on page 4 3 of Adamson ( 1 960 ) . For the mutual l y sa turated l i q u i d s wh i c h were a l l owed to s tand for twenty-fo ur hours before the mea surements , the notat ion Ya ' and Yb l has been u sed wh i l e Ya " represents those sampl es measured after o ne week . These val ues were then corrected to O°C and the temperature corrected val ues are shown a s Y ( TC ) , Yb ( TC ) , Ya ' ( TC ) , �tc . Spread i ng coeffi c i ents co rrespond i ng �o the u nsa turated and mutual l y saturated s tates are rep resented a s S , S I and S " , res pecti vel y .

Sal inity Ya Ya l ppt dynes/em dynes / em

6 0/00 69 . 9 55 . 6

1 2 0/ 00 69 . 9 55 . 6

1 8 0/00 69 . 9 55 . 6

2 4 0/00 69 . 9 55 . 6

30 0/00 69 . 9 5 5 . 6

3 6 % 0 69 . 9 55 . 6

o i l only -- - - ----

TABLE C- l : DATA USED FOR CALCULATING THE SPREADING COEF F I C IENTS FOR SWAN H I LLS CRUDE

Ya" Yb Yb ' Ya (TC) 1 Ya I (TC) 1 Ya" (TC) l Yb (TC) 2 Yb I (TC) 2 dynes/em dynes/ em dynes / em dynes/ em dynes/em dynes /em dynes/ em dynes/em

54 . 6

5 4 . 6

54 . 6

54 . 6

54 . 6

54 . 6

1 .

2 .

3 .

4 .

5 .

26 . 2 74 . 2 60 . 0

26 . 2 74 . 8 60 . 0

26 . 2 75 . 4 60 . 0

26 . 2 76 . 0 60 . 0

26 . 2 76 . 5 60 . 0

26 . 2 77 . 1 60 . 0

24 . 9

Using �i = - 0 . 16 d�:s /em [ 32] .

U · ay

= _ 1 dynes/ em [25] SIng y T o . bC .

58 . 7 29 . 0

5 8 . 7 29 . 0

58 . 7 29 . 0

5 8 . 7 29 . 0

5 8 . 7 29 . 0

58 . 7 29 . 0

28 . 9

Calculated using Ya (TC) and Yb (TC) i n equation 9 with � = 0 . 832.

C a l culated using Ya (TC) , Yb (TC) and Yab in equation 1 6 .

Calculated using Ya l (TC ) , Yb ' (TC) and Yab in equation 1 7 .

6 . Ca lculated a s fol lows : S " Ya" (TC) - Yb ' (TC) - Yab .

Yab 3 dynes / em

26 . 0

26 . 3

26 . 6

26 . 9

2 7 . 2

27 . 5

S 4 S I 5 dynes /em dynes/em

19 . 3 5 . 0

19 . 6 4 . 7

1 9 . 9 4 . 4

20 . 2 4 . 1

20 . 4 3 . 8

20 . 7 3 . 5

S" 6 dynes /em

3 . 7

3 . 4

3 . 1

2 . 8

2 . 5

2 . 2

- - -----

-....J -....J

S a l i n i ty Ya Ya ' Pr: � dynes/em dynes / em

6 0/00 69 . 9 4 1 . 9

1 2 0/00 70 . 5 4 3 . 2

1 8 0/00 7 1 . 1 44 . 4

24 % 0 71 . 7 45 . 7

30 0/00 72 . 2 46 . 9

36 0/00 72 . 8 48 . 2

I o i l only

TABLE C- 2 : DATA USED FOR CALCULATING THE SPREADING COE F F I C I ENTS FOR NORl-1AN WELLS CRUDE

Yb Yb ' Ya (TC) 1 Ya ' (TC) l Yb (TC) 2 Yb ' (TC) 2 Yab 3 dynes / em dynes / em dynes/em dynes/em dynes/em dynes/em dynes/em

1 .

2 .

2 3 . 2 74 . 2

23 . 2 74 . 8

2 3 . 4 75 . 4

23 . 6 76 . 0

24 . 7 76 . 5

26 . 9 7 7 . 1

24 . 1 -

Using aTy = - 0 16 dynes/em Y . d �

Using �i = - 0 . 1 d��s/cm

<16 . 1 25 . 9

4 7 . 4 25 . 9

48 . 6 26 . 1

49 . 9 26 . 3

51 . 1 27 . 4

5 2 . 4 29 . 6

2 8 . 1 - --- -

[ 32 ] .

[25] .

3 . Calculated using Ya (TC) and Yb (TC) i n equation 9 with � = 0 . 831 .

4 . Calculated using Ya (TC) , Yb (TC) and Yab in equation 1 6 .

5 . Cal culated using Ya , (TC) , Yb , (TC) and Yab in equation 1 7 .

26 . 4

26 . 7

27 ; 0

27 . 3

27 . 5

27 . 8

6 . Calcu1 at�d us ing S ' i n equation 1 5 and density data from Table 3 .

S 4 dynes/em

1 9 . 7

20 . 0

20 . 3

20 . 6

20 . 9

2 1 . 2

S ' 5 dynes/em

-6 . 2

- 5 . 2

- 4 . 5

- 3 . 7

- 3 . 8

-5 . 0

too 6 em

0 . 29

0 . 2 7

0 . 25

0 . 2 2

0 . 22

0 . 25

� CO

79

APPENDI X D

TEST METHOD USED I N DETERMI N I NG SALT CONTENT OF CRUDE O I LS

80

APPEND I X D - MODI F I ED BLA I R METHOD FOR DETERM I N I NG TOTAL CHLORIDE CONTENT OF CRUDE O I L

Scope : Th i s method i s i ntended a s a rap i d and rea sonably accurate method o f determi n i ng the to ta l ch l ori des content o f · crude o i l . Reference : I IB l a i r Metho d l l i nd o Engl Chem . , Anal . Ed . , 1 0 , 207 ( 1 938 ) . Appa ratus :

a ) Sepa ratory funnel , 500 ml . b ) Beaker , 250 ml . c ) P i pette , 1 00 ml . d ) Graduate , 1 00 ml . e ) Graduate , 50 ml . f ) Funne l . g ) No . 41 fi l ter paper .

Regeants : a ) Benzene , C . P . b ) S i l ver n i trate 0 . 05 N . c ) Potas s i um c hroma te i nd i cator . d ) Tret- O- L i te Destabi l i zer "A l l or " B " . e ) Sod i um b i carbonate , 1 0% so l n .

Procedure : 1 . 1 00 ml . of crude o i l sha l l be p i petted from a wel l s ha ken samp l e

i nto a 500 ml . separatory funnel . 2 . 1 00 ml . o f Benzene s ha l l be added and a drop o f Destabl i l i zer A or

B i n concentrated form . The funnel and contents s ha l l be s ha ken for 30 seconds .

3 . Exactly 1 00 ml . o f bo i l i ng d i s ti l l ed water s ha l l be added to the funnel and the contents s ha ken gent l y , rel i ev i ng pressure unt i l safe .

4 . The separatory funnel and contents s ha l l be s ha ken v i gorous l y for 5 mi nutes and a l l owed to settl e . Any i nterface sha l l be c l eared w ith a warm w i re .

5 . The separated aqueous so l u t i o n s ha l l be drawn off through fi l ter paper i nto a 50 ml . g raduate , unti l exact ly 50 ml . a re o bta i ned .

6 . The contents o f the g rad uate sha l l be trans ferred to a 250 ml . bea ker and the graduate ri nsed w i th 2 5 ml . of d i st i l l ed water . The was h i ng s are added to the bea ker .

81

7 . The p H of the so l ut ion s ha l l be regu l a ted to approx . 6 . 5 wi th sod i um b i carbonate and the extract ti trated wi th 0 . 5N s i l ver n i trate u s i ng 5 drops of a 5% so l ut i on of potass i um chromate as i nd i cator .

8 . The endpo i nt so obta i ned shal l be matched wi th a t i trat ion u s i ng 75 ml . of d i st i l l ed water and 5 drops of i nd i cato r . The vol ume of s i l ver n i trate req u i red sha l l be s ubtracted from the prev i o u s t i trat i on and the resu l t expressed as po unds of NaCl p e r 1 000 bbl s . of crude by mul ti pl yi ng the rema i n i ng vol ume o f s i l ver n i trate by 1 9 . 8.

Note 1 : The factor , 1 9 . 8 takes i nto account the d i fference i n vo l ume between the water added and the extract wi thdrawn , due to d i fference i n temperature . Note 2 : An experi menta l determi nati on wi l l q u i c k ly s how wh i c h type of destab i ­l i zer i s mo st s u i tabl e to the crude under tes t .