UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Extraction of oil-shale Brock, Thomas Leith 1937

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-UBC_1937_A7_B7_E8.pdf [ 9.68MB ]
JSON: 831-1.0061978.json
JSON-LD: 831-1.0061978-ld.json
RDF/XML (Pretty): 831-1.0061978-rdf.xml
RDF/JSON: 831-1.0061978-rdf.json
Turtle: 831-1.0061978-turtle.txt
N-Triples: 831-1.0061978-rdf-ntriples.txt
Original Record: 831-1.0061978-source.json
Full Text

Full Text

• m.WABWim OF OtL-SB&LB" f!  Bj  mam  t .  mocs:  ffiMKJAffi 0 S S I O A L MOGSEERlHa 1SESJS m f f f f l s i f f  OF miTim  QoaxmtL  ttELB  OF  OQjsraMgS  1  Importance of Fully Investigating Oil-Shales Shaio In-TOstigatod ...........  .  ..... S  Trial Extractions ..............«. ............. 5 Analysis of th© 'Shale ............................... 4 Construction of Extraction Apparatus ................ S Operation of the Extraction Apparatus  ?  Treatment of the Extract ............................. 8 Interpretation of Data Obtaiasd ..................... 10 Conclusion .......................................... 11 Appendis 1 ................ Extraction Theory ....... 12 Appendix 11 ............... Sketch for Continuous Counter-Current Extractioa Plant for Oil-Shai© ............................... IS  LIST OF 1IMJSTOATI«S  Soshlot Extractors Vacuum  Distillation  «.  ...................  ..............  ...... 45  Turneco for Carius Determination .................... Combustion Analysis Train ..........................« Diagram of Extraction Apparatus  Wiring  „ Q>  Diagram  Thermocouple  Wiring ................................  Extraction Apparatus ............................... Propane  Apparatus  ..........................a.......  Flow Diagrams for Various Methods of Extraction .... Sketch  S  of Continuous Oil-Shale Plant  ~J  -  X  -  3HP0RTAHGE OF FULLY IM15TI6&TMG 0XL-SBALE3 The vjorld supply of petroleum i s a natural resource that i s rapidly being consumed. The fact that the supply i s not inexhaustible, and the gravity of the situation that vri.il result from the supply becoming so depleted as to be inadequate to most demands have been generally appreciated only within the past fen years. As a result there i s nox~ a widespread effort to conserve petroleum and i t s products, and to seek possible substitutes for it« Even to-day, when the proven world sources of petroleum w i l l last for several decades, many countries fearing that they w i l l be cut off from the supply through war or other reason are frantically 9  searching for net? fuels. For Great Britain especially, who imports a l l her o i l , the question of possible substitutes for petroleum ia of national importances As a source of substitute for petroleum oils, the reserves of o i l shale in many countries throughout tho world stand out as most important. O i l shale deposits in Scotland, France and Australia hav© been the source of products similar to those obtained by petroleum, and i t is -highly probable that the vast deposits of the United States ana other countries w i l l bo extensively worked in the future. Under present conditions the shale industry i s unable to compete with the petroleum industry, largely because the only system employed i s that of retorting. A retorting process for oil-shale cannot compete with petroleum, for though the investment cost of a retorting plant i s much less than the average "well cost" of petroleum par barrel  e  the cost of producing petroleum is much lower than that of producing an equivalent quantity of shale-oil,, Difficulties involved in a retorting process are:(1) Maximum yield by distillation for average good grade oil-shale i s 40 gallons per ton of shale. (2) Retorts and equipment become rapidly corroded, by reason of the high temperatures involved in the process. (3) The high temperatures cause a certain amount of decomposition or cracking of the o i l which affects adversely the products of the refining process. (4) The retorting process i s not a continuous process. If howevar, a continuous solvent extraction process could be developed tshich rauld remove a larger fraction of the bituminous constituent of the shale the cost of production could be reduced. High temperatures would be unnecessary, corrosion avoided, extracted matenial would not undergo decomposition, and the solvent recovered. The cost of mining the shale rihleh i s 55$ of the total cost in the retorting process would be 5.ncreased  slightly for i t would bo necessary to secure fine  grinding for a solvent process. Whether or not a nev; process for oil-shale could be siaae a coBsaercial success within the next few years or Hhethor i t will have 5  to await further decreases in the world petroleum supply the fact rsmains 8  is  that incalculable work  yef  gfmaS&B  to be done i n the Investigation of the  chemistry of shale-oil; that this field has been barely touched by tTorkers in the past said that no time mast be lost i n pressing further investig=B  ations.  So AH LET EXTRACT**?  SHALE A  INSTIGATED  representative high-grade  - Albert shale = obtained  by  Canadian oil-shale  the Department of Mines, Ottawa-,  vicinity of Rosevale,  Albert  County,  however, are  not as  good- grade as most  samples,  Albert shale i s  dull  Albert  in colour  from the  These particular  Hew Erunsnick.  black  was chosen  shale.  when mined, but  becomes brownish on exposure to the air„ and i n many  ways  corresponds  to  the Scottish shales. The shale grinds and pulverizes easily  with  little  dirt  0  and leaves no o i l matter on the machinery.  TRIAL ISCTRACTICaS 40-mesh shale was dried at 90 degrees, dessicated, and 55  gram  samples used.  Samples  i c acid, and butyric a c l d  5  were extracted  with acetone, benzene, proprion-  using Soxhlet extractors on hot plates.  extraction was run for 200 hours, changing the solvent Minimum  heat was  employed.  using  asbestos padding  occasionally.  where necessaryo  The solvents were d i s t i l l e d o f f under reduced  the  extract residues dried i n evaporating dishes  e r a l hours.  Each  on  pressure„  water-baths,  and  for ssv=>  The brown-black asplalt products i*ere dessicated and weighed. Portions of the extracts vrere calcined to determine the  percentage of inorganic material  which was  present.  Qualitative analysis  of the inorganic portion indicated the presence of A l , Fo, Zn, Ca, Mg„ 1. For a f u l l e r description see: Bailey and E l l s = "Report oa. the Albert Shales of Albert and Westmorland Counties, H.B." Gool. Sur. Can. 1876: also A. A. Sninnerton, " O i l Shale from Rose vale, H.B. Mines Branch Re. 689, Page 104. n  VAC«"M  DISTILLATION  and: M&a The butyric extract was dissolved i n benzene, f i l t e r e d solvent d i s t i l l e d , and extract dried on a water bath.  0  Extract vxas weighed  and calcined. The following table was obtained;  TOTAL EMAGTICM  SOLVENT  TOTAL  ORQ«I0  Acetone  2.4  3*4 - •  Benzene  5*0  3.0  • 23.4  18.8  20.7  16,5  Proprioslo . Butyric  TOTAL SOLUBLE  M BMZHE  SOLUBLE M BMZM1  OROAHIC  14.7  ULo8  (Figures are percentages of original shale). ANALYSIS OF THE SHALE A method for the determination of the organic matter employed by Galssan and Bader"  vjbe  triad.  e  F i n s l y powdered samples of ths  shale was fused with s i s times their weight of sodium hydroxide, u i t h vigorous s t i r r i n g over a low flams. filtered.  The melt was poured into water,, and  The greyieh-brora. residua contained most of the s i l i c a i n ched  addition  to the organic  natter-. The inorganic' setter i?as leaj&ttfi. out with  d i l u t e hydrochloric acid, and th© residue was dried to eonetant weight at 100 degrees.  2.  This gave organic matter 36.92$ which i s high, probably.  Gaissen, P.O. and Bader, H. - Wurtemberg O i l Shales 11, Chem.. Ztg. 50 (1926) Pages 277 - 273 »  CAR^S  Corns  DCTE R WIN ATI  TRAIN  5 <=,  e  Samples  in  procelain  of shale  dried In an oven, and then calcined  crucibles. Moisture  «.  .85$  Organic Matter »  2915$?  Ash  7Q„00$  a muffle  6  -  Samples ibles in  wars  of the shale  furnace  were  heated in  partially  at 45G°C«, dessicated and  material at that temperature  weighed.  was 19.6$ of which 18.75  covered  Volatile  i s organic.  Carius determinations f o r sulphur and chlorine  cruc-  nocs-f-or-c)  (i'e l***  wore made  9  follovrf.ng Gattermaim' a' procedure. Although micro analysis i s considered simpler and more accurate  for this type of work,  macro analysis.  The  there was available only  Liebeg method  was  tubs nas heated et white heat f o r four  of air. The  average  of four runs  t7as  employed, using  hours,  taken,  apparatus for-  a quartz  passing over  tubs.  oxygen  She  instead  the values not cheeking  welli  TABLE OF ANALYSIS  .  &  1.004  3.  0L Oo  0 9.8?  • H  c/a  HflTTEB  X o S3 -  5.86  • 29.15  Qattermann - Laboratory Methods of Organic Chemistry. Uieland, i932, Pages 65 - 68.  Begised  by  EXTRACT/ON  /APPARATUS.  - 6CQHSTRUCTICEI Following  OF EXTRACTION APPARATUS  the relatively  special solvents proprionic and butyric extraction  on  a  with two parallel circuits of heavily lagged  i?ith 1/8 The  on the  frame  acid,  i t was  decided  larger scale. For this purpose three  5 centimeters in diameter and 100  using the  successful extraction  to attempt  columns  centimeters i n length, vjero  nichroma  resistance  or stores, each t?rapped  x?ire and the outside  inch asbestos paper.  accompanying  work by means of  shov/s how the stoves are placed  diagram  adjustable copper  bands, with  glass  delivery  tubes leaving from the bottom of one stove to the top of the next. The stoves are staggered so that they are continually over half f u l l of solvent. Storage flasks rrith stop-cocks are provided at the top of the first stove and at the top of the heavily lagged 1000 centimeter distilling flask. back  She  distilling flask i s provided tilth  peep-hole  and  lighting and has a stop-cock in the bottom for the removal of the  extract to vacuum distillation. gas ringo The  distilled  The flask i s heated on a sand-bath by a )  acid can  be  either removed  from  the system  or sent bask to the top by an a i r - l i f t . At the other end of the plant fresh acid i s sent to the starting point from the The nected  xfiring diagram shone  in parallel but are  provided with  supply, by air-lift<,  that the three stoves are  con-  individual knife switches so  that they can bo heated independently. A variable resistance Is included which'can be cut out of the circuit when desired. The  stoves are provided vdth  alurael-chroinel  thermocouples  DISTILLING-  AND  RECEIVING-  FLASKS,  ana  a direct reading  pyrometer.  OPERATION  OF THE EXTRACTION  APPARATUS  After a t r i a l run, followed by minor changes and adjustmento i n the apparatus, the main extraction was made  0  Glass-wool was placed at the bottom of the stoves and a charge of 800 grams of shale added to each, with the following dimensions: Bottom layer -  10 mesh,  Hext layer  »  20 - 40 mesh,  Sezt layer  -  40 «=> 60 mesh,  Top layer  -  SO mesh.  The charges h a l f - f i l l e d the stoves and rare entirely covered by solvent during the process. The stoves could only be heated under supervision to prevent solvent boiling and escaping from the top; to regulate the flow of acid? to v;atoh the distillation,* and to return  acid v;hen necessary.  The  storage f l a s k above the d i s t i l l a t i o n f l a s k ras very important i n the  flow,  regulation of  for when gas i n the stoves expanded quickly solution  came over at a faster rate than the d i s t i l l a t i o n could handle.  It re-  quired nice adjustment to obtain regularity i n the steps of the process.  Whereas or  1080  hours,  the length of time of the extraction was 45 days  the actual  f/as only 150 hours.  time of  for  nith  distillation  D i s t i l l a t i o n of 350 c c . batches could be run at  approximately hour intervals. through a day  supervised heating  the  On the average 1800 cc. of acid were run  SO working  days,  or a total of  54,000 c c .  • H e  For the f i r s t few days the extract came over black, gradu a l l y lightening i n colour through various stages o f broraa to a l i g h t red tinge.  The temperature of the stoves was maintained as f a r as pos-  sible at 150°C., but the times when they did overheat the extract cam© out  darkero After every f i v e batches of d i s t i l l a t i o n , the heavy eztrac  was run. o f f and sent to vacuum d i s t i l l a t i o n which was being operated simultaneouslyo The extraction v?as continued u n t i l when for a vieek the weight of extract was reduced from 5 to 5 to 4 to 4 to 3 to 3 to 3 grams, for each 1800  cc. of acid put through.. 551 grams of product v/ere obtained nhich i s a y i e l d of  23.4£v  This vss a hard brorm solid with a jet black b r i t t l e coating of  very high l u s t r e .  On standing a feu hours exposed broim surface changes  to shiny black.  TRBA3BMT  OF THE SgERAOT  The extract was tested f o r inorganic matter by calcining samples i n platinum crucibles.  I t was found that 20.3$ was inorganic,  which means that the extraction removed 18.7$ organic material from th© shale.  . 10 gram samples of the extract were dissolved i n carbon  bisulphide  8  carbon tetrachloride, benzene, sad petroleum ether.  The s o l -  utions uere f i l t e r e d , v;ashed with solvent u n t i l clear, the solvent d i s t i l l e d o f f , and the extracts dried on water baths.  In oaeh case the extracted portion, and the unestractod por<= tion was tested f o r inorganic material by calcining in platinum crucibles® Evidence of paraffin precipitating out could be seen, on the flasks and funnels.  In the ease of the petroleum other solution there  t;as quite a heavy red precipitate remaining in the flask. A 5 0 gram sample of extract vjas taken and extracted ivith petroleum ether and this extract reduced to about 7 5 cc. in a tube inside f  a Detvatf flask.  The flask v;as cooled to minus  by solid carbon di-  45°C.  oxide in methyl alcohol and propane passed into solution to precipitate 8  the high molecular ivaight hydrocarbons. A volume of liquid propane equal to the volmns of extract solution was added, but no precipitate resulted. The procedure viaa repeated, using a carbon bisulphide er° tract solution, but again no precipitate resulted. The following table of results was obtained from experiments on the butyric acid erfcract:-  YIELD HELD £ OEGAHIG AS $ OF AS fo OF BOTIRIC  EXTRACT Butyric Acid Petrol Sther  Cssrhoa  Bisulphide  Oarbc  {' Tetrachloride | Benzene  SSfiLE  TRACTS)  PORTION  1 0 0 ..0  fo 0RGAHG  TRACTED  P0&T2CSI 7 9 . 7  HELD OF ORGMIC  MATffllAL SOLUBLE m THE SOLYMT AS p OF SHALE. c  1 8 . 7  2 0 c 6  4.84  86 oS  4 . 1 7  29»8  6 , 9 7  80.95  5 . 6 4  4 3 . 5  10 c. 2  7 2 . 0  78  8 00  4 5 . 4  1 0 . 6  7 1 . 6  8 0 . 8 5  o  4  o  8 . 5 5  - 10 A numerical eraraplo ©ill aero to clarify the above. Suppose two 10 gram simples of butyric extract are taken. Carbon bisulphide will  remove 2.98 grams and leave 7.02 grams; of that 2.98 grams, 2.S9 grams  are organic and 0.59 grams-are inorganics of the 7.02 grams, 505 grams are organic and 1.97 grams are inorg&nie. Benzene will remove 4.54 grams leaving 5.46 grams? of the 4.54 grams, 3.64 grams are organic and 0.90 grams are inorganic; of the 5.46 grams, 3.92 grams are organic and 1.54 grams are inorganic  e  STTERP3ETATI0K OF DATA OBTAMED On the large scale extraction the total yield of product was 25.4$, or 18.7$ organic yield, while on the small scale it was 20.7$ total yield or 16.5$ organic. Yet despite this on the large scale tho yield soluble in benzene nas 10.6$ or 8.55$ organic, and on the Eaall scale was 14.7$ or 11.8$ organic. Ihy is it that in our trial extraction we got a smaller yield but a 25$ increase is. the amount soluble in benzene? It would appear that the large scale extraction was carried on for too long a period •» 1030 hours against 150 hours i n the Sozhlet extraction <=> and that the acid had too great an opportunity to act on the inorganic material i n the shale. The formation of calcium and magnesium butyratos would account for the large percentage of organic matter end yet the relatively aaall part of this which is soluble in ordinary solvents.  It  is probable that the bituminous material wa3 removed very early in the extraction, and that the remainder of the extraction merely succeeded i s attacking other constituents.  -  11  -  The. .portion, of the extract insoluble in the various solvent which was-over 70$ organic, would be largely butyratos. It i s rather surpri sing that benzene should remove more organic material than carbon bisulphide. One might suspect  -  (be****)  that not a l l  the heavier solvent^ was evaporated fro® the extract., but this i s not the ease. The extract was heated tor several hours on the water-bath and continually pricked with a pin, until the extract was solido Benzene added to the substance insoluble in petrol ether removes 16.4$ of the original extract, bringing the ccsbirBd extraction to •37$ yieldo  .  .  The fact that liquid propane did not bring down a precipitate indicates the absence of high molecular weight hydrocarbon in the solvent extracts examined„  , CQMCLPSIQg  A yield of 8.55$ of the shale organic matter soluble in benzene would only be 190 pounds per ton of shale,  ?3hereas  a yield of 40  gallons of o i l per ton, of average specific gravity .92 would be E® pounds On the fac«S of i t , then, the results of the extraction are not very satisfactory,, However i t i s probable that on extraction apparatus built on the plan of a large Soxhlet extractor, ?Mch would allow refl«xing  s  would carry out the extraction in a relatively short time, would give the greater yield that i s required, and would largely avoid the formation of the butyratss which appeared in th*Sextraction.  To test out this theory, a large Soxhlet extractor M i l he constructedt and extraction tests made on Scottish oil-shale supplied by the Fumpherston. Oil Company.  .APmroix_i  EggtAPTIOg TSSORY H3TR0DU0TI0K; Until recently no quantitative expression existed for ex= traction systems. Quantitative data on the rate of extraction of different materials,-, tMch might form the basis of calculations are practically non-existent, and in the various industries employing extraction processes the oi.. of apparatus and tims of contact are determined by empirical experience.  In the past few years, however, several discussions have  appeared in the Journals bearing on the subject of quantitative relationsships. "Badger and McCabi discuss the theory of extraction under trio heeds: first under the assumption that timo enough has been provided for equilibrium conditions to be reached in the process; and, second„ the rate of extraction before equilibrium i s reached. If the first assumption i s mo.de and i f certain processes such an selective absorptions are neglected, i t i s possible by means of material balances to carry out calculatioas on extraction processes whether they be simple batch processes or continuous counter-current treatment of solids by liquids*, 1.  Badger and McCabe: "Elements of Chemical Engineering". McGran-Hill 1 9 3 1 .  - IS OAlCTOAaiOWS BASED 01 ATTAIHMMT OF B&TOJBKRMr Mien, certain assumptions are made regarding equilibrium in Xtdoce. io rtr*i?J>f wff'il  the extraction and washing apparatus,, the calculations/\arQ:  b«/a»rrs.  (1) The  Th* cttrunpT/^ w a s h  liquid i s completely and uniformly mixed with the solution adhering to the surface of the solid; and (2) the solid exerts no selective absorbing action, on t h e solute but i s completely inert. Hawley^ discusses such a case la the treatment of a definite quantity of solid with successive fresh batches of extracting liquid. I f "a" i s the .solvent" ratio, o r t h e voloae o f t h e .solvent drained o f f i n o n e wash divided by the volume of the solvent retained by the solid, the fraction of the original amount ©f soluble material that remains unextracted i s ; for 1 wash  I  1 for 2 washes 1 (a 4> 1)2 . a  *  for 3 washes  .1  for n washes  1  TTTTF Utilizing these formulae a table can be constructed for the percent extracted with different solvent ratios and number of treatmentc  0  Such a table makes i t very apparent that more washings with the same volume of solutions afs more efficient than using the volume a l l at once.  2. J. C. Hawley, J. Ind. Eng. Chem. 12, 493, (1920).  If the trashing i s done in a eounterourrent system the amount remaining unextraoted after the washing i s :  1  RATE OF  *a*a *a H  a  a**  ESIRACTICE?;  The general principles of extraction rate correspond closely to those of other diffusion processes. Consider a particle of solid containing some extractcblo material in contact with liquid sjhich i s being used for extracting. At any instant the solute in the extract i s "C".  At the same instant the concentration of the material at the  interface may be represented by "Co".  This value of "Co" may or may not  bo equal to the concentration of the material in the solid itself, depending upon the structure and nature of the solid. In any case the rate of diffusion from the interface into the liquid i s proportional to Co - C". r  A general calculation of the rate of extraction would require a knot/ledge of the rat© of diffusion through the solid itself, as nell as the pro-= portlonality constant or film coefficient for the diffusion from the inter<= face into the bulk of the extract. Certain cases can be distinguished. First, consider a solid entii'ely impervious and inert to the action of the solvent, with a film of strong solution on i t s surfece. For such a caso the process would involve simply the equalization of concentrations in the bulk of the extract and in the adhering film. Such : a process i s rapid and any reasonable time of contact isill bring about substantial equilibrium. This i s especially true If there i s agitation and i f the. system i s nsrm so that viscosity i s lowered and the diffusion  coefficient increased.  A  second  ease i s one where  soluble  the  material  distributed through the,interior of a permeable insoluble examples'  black  ash  or calic2>e.  In  is  uniformly  for  solid »  this case the rate of diffusion of the  solute to the interface may be the controlling factor i n the rate of  For such cases  extraction.  of  thickness  help i s  little  the l i q u i d f i l m , but  obtained  by  decreasing the  tho process is greatly hastened i f the  solid .is finely subdivided. A third leached  it  in  is  important  ease i s  impermeable and contains the  very  in which the solid  that  soluble  f i n e p a r t i c l e s , plates, or needles.  being  portion distributed through  In such eases tho actual  surfaces of the p a r t i c l e s of soluto may be an extremely small portion of  the total  surface  pondingly small.  of the  s o l i d and the  rate  In such cases the best  of extraction w i l l  method  be  is to grind the  corr©s=  solid to  such a point that the individual p a r t i c l e s of the soluble material are releasedo S t i l l another case i s that t y p i f i e d by such substances as sugar beet chips and  tanbark,  where the soluble material i s contained i n  c e l l s and tho process involves the diffusion of the solute through the  cell  walls,  which act as  semipermeable membranes.  Such a process  volves the phenomenon of osmosis and the diffusion of colloids.  In the case of such materials the  soluble  e f f e c t , very f i n e l y subdivided throughout the  i7i<=  crystalloids through,  constituent i s , i n  mass of inert  solid. I t  would seam, therefore, that the l o g i c a l procedure would be to grind the material very f i n e . that  each  This  individual c e l l  would necessitate  would be  ruptured.  grinding  the material so fine  XIh@n i t  is  recalled  that  •» 16 •» to pass a 200 mesh screes represents almost th© limit of  material ground  fin© grinding and that a particle hundreds of individual  contain  highly  impractical. Further,  of beet  traction solution.  sugar,  i t i s seen  in some cases, especially that of the  ex-  th© cell  contains  other substances than sugar i n  the  would  pass into solution;  i f extraction be carried out by direct diffusion, some of these high molecular  weight  are held back.  sort the degree of subdivision should a  will  is  that  the  screen  If the material were so f i n e l y subdivided that a l l the c e l l s  impurities of  form  200 mesh  suggestion  cells,  were broken undesirable soluble impurities  whereas  pass a  that w i l l  mass which cannot  chips  be  should have the  not be so  penetrated by  the  For material of t h i s  great but the  solute.  At  the same time  largest possible surface to increase  of contact between solvent and s o l i d .  trill  chips  the  area  The diffusion processes are acoel«=-  orated by carrying out the operation at as high a temperature as possible  and any surfaeo film  resistance i s decreased  the solvent high, with  correspondingly  by  making  the v e l o c i t y of  great turbulence and  decrease  in the thickness of the stagnant f i l m . ' I f the v e l o c i t y of the solvent past  the  chips i s high, the necessary time of contact f o r producing  high"  concentration solutions can be made s u f f i c i e n t l y great only by having  the path of  This i s  the solvent correspondingly long.  accomplished i n  diffusion batteries by placing a considerable number of c e l l s i n series. The extraction of tanbark proceeds rather rapidly, the v e l o c i t y of the solvent need not  In  be  the extraction  velocity of  the  of  co  high, and 4  beet sugar the  to 6 c e l l s  are usually  diffusions!  processes are slow, the  solvent i s high, end from 10 to 19  in ..series to give  sufficient  time of contact®  sufficient.  cells  must be  placed,  T.  an  invincible  17. EtranG*  5  In the  states that  solute such that tho partition,  extraction  law  of a solute by  holds, the degree of  solute removal i s conditioned mainly by the relative volumes o f the solvents." TJhlle uso of several small p o r t i o n , excessive  portions i o  s u b d i v i s i o n i s u n j u s t i f i e d , s i n c e a l i m i t i n g percentage  removal e x i s t s f o r every r a t i o o f solvent  division  while to  volumes. At  of which a given volume of solvent  maximum removal  by  preferable to one combined  into five  divide tho  94$ of this  capable i s achieved  so that in practise it i s scarcely  portions,  extracting  is  least  solvent  In another article  iTorth  into more than five portions.  Evans"  h a s developed e a s i l y a c c e s s i b l e  g r a p h i c a l methods f o r the c a l c u l a t i o n o f t h e o r e t i c a l e f f i c i e n c i e s i n both m u l t i p l e and counter-current  extraction.  The methods v a r y , depends  i n g on whether o r not the s o l v e n t s may bo considered i n v i n c i b l e . 5 Thiele  c l a i m s t h a t there i s  t i l l a t i o n and counter-current  dis-  l i q u i d e x t r a c t i o n , and by s u i t a b l e m o d i f i -  c a t i o n he has a p p l i e d Ponchon's g r a p h i c a l d i s t i l l a t i o n columns, t o t h i s  a c l o s e analogy between  method  for  the  analysis of  operation.  R e c e n t l y v a r t e r e s s l a n and  Fenske  have developed exact  q u a n t i t a t i v e r e l a t i o n s for l i q u i d » l i q u i d e x t r a c t i o n .  6.  Y a r t e r e s s i a n , E.A., and Fenske, M.R. - J . Ind. Chem. 28, 1553 (1836).  3.  Evans, T.W.„  J . Ind. Eng. Chem. 26, 439 (1934).  4.  Erans, T.W. ,  «T. Ind. Eng.  5.  T h i e l e , E.17.  J. Ind. Eng. Chem. 27, 392 (1935).  Chem. 26 , 860 (1934).  » 18 The various methods of extraction are arranged as follows: 1. Single-stage. 2. Cocurrent contact. (a) Multiple stage. (b) Infinite stage. 3. Counter-current contact. (a) Multiple stage. (b) Infinite stage. "Cocurrent" means multiple contact using fresh solvent in each stage. Although in extraction practice there are other methods or combination of methods, the mathematical treatment of the methods above bring out the essential principles applicable to a l l extraction processes. The use of reflux in counter-current extraction w i l l reduce the number of stages necessary for a definite separation, yet  villi  not introduce  any new variables for i t s mathematical treatment. Apart from operating data, phase equilibrium composition relations are the only data necessary for the mathematicl treatment of extraction methods. The compositions of the phases at equilibrium at contact temperature and pressure are usually plotted graphical J.y. For their derivation of formulae, Varteressian and Fenske have taken the three component system chloroform - water - acetic acid at 18° C and 1 atmosphere. They had a given quantity of a mixture of chloroform and acetic acid of a definite composition, and desired to reduce the amount of acid in the chloroform to some other definite composition by extracting the acid with water. TJith each of the methods of extraction they had to find out the amount of T?ater necessary for  extraction and the amount and composition of the tno liquid layers pro= duced. The formulae are derived on tho basis of equilibrium relations and of material balances so that no assumptions are introduced to limit their use only to cases where such assumptions w i l l bo justified. Moreover, the number of given factors is kept at a minimum and e l l variables are expressed in terms of the amount of original charge and the terminal compositions of the various stages  c  The two layers, before the removal of the solvent are distinguished by the names of "extract layer" and "tJaffinete layer". The extract layer contains a large proportion of solvent and a small proportion of the liquid to be extracted, whereas the ^affinate layer w i l l contain a large proportion of the liquid to be extracted and a small proportion of the solvent. METHOD 1, SIBGLE  STAGE %  Let Ao, Co, and Bo represent the weight of chloroform, acetic aeid and water„ respectively, in ths original mixture of weight 8  Ho; let So, xo, and so bs the corresponding compositions in weight fractions. Let this mixture be treated with a weight, Sf of rater in order to reduce the acid concentration in the heavy layer to xf, the corresponding composition of chloroform and water being Sf and Sf as determined from the saturation pl(J.t, At this stage the concentrations of the three, components in the light layer w i l l be Yf * yf, and yf, for chloroform, acetic acid and water, determined from a graph on which aro  - 30 plotted the xTsight per eeat of each of  th©  layers vs. those i n th© heavy layers at designates  It Ef  three components in the light  equilibrium.  th© weight  of tho resulting  heavy layer,  and Lf that of the light layer formed, and material balances are made on chloroform and acetic acid, we have the equation. HfXf  4  LfY  HfXf  & Lfyf  (1)  HQXO  f s  e  (2) .  HQSO  vjhich solved simultaneously for  and. Lf give:  %  s EL, Ipyf - X p %  Lf  » Ho %f*o , Xy - xY f  f  f  | | S  (A) f  From a total material/balance; %  f %  s  %  *• %  (5)  .  Substituting in equation (5), the values of Bj and Lf and and solving for 8f . Ho f Q r * f > X  3^_T£jDjD Sa:  ( y  » ^(*fof>. .  J  161  CQgUERMT KPLTIPLS STAGE:  ITith' the two stage method i t i s possible t© have any desired concentration, there results a  ations  In the  of acetic  concentration  concentration desired. enough solvent  x^,  in  Xg,  first  stage, provided  i n the second stags equal to the f i n a l  In practice, x^ could be fixed either by using  .each stage so that  saffinate  acid i n the  the decrements  of the  acid  concentr-  layer tram one.stag© to the other are equal, or .  by  using'vQQUal  amounts of solvent i n each stago. In general, neither  values of  nor the amounts of solvent  ures,  the same.  w i l l be  used, determines  "the  by the proced-  By material balances around the first and second stags® employing a procedure similar to Method  1,  values of H^,  obtained. By extending the formulations to eases stages,  the values of H,  L,  ii  T T  "' tit  X^'-i V-<" ~  "Hi /w-i  -I  F o r the ease of equal z decrements:  . ind for. the^case of equal, amount of ^solvent. used': ;  w  S  f  and  are  three, four or more  and S, for any stage, n, in a  multiple stags system of H stages, are given by:H  of  Lj»,  cocurrent  - 22 MB!EaQP,-,ab;  OOCURBEHT BfOTHEEE STAGE:  .  In this method the extract layer would continually he  removed from contact H  he.the  formed,  of  weight  with  the  layer.  saffinate  At a  given instant let  L that of the total extract layer  saffinate layer,  and S that of the total solvent used. % material  t i t i e s , for acetic • '  acid,  balances>  for chloroform, and for total  - Bdk - ac^a. -  - Sfe  S  using differential and actual quanmaterials:(12)  y(fc  *» l £  (13)  - H «• L - Ho  (14)  Combining (12) and (13) and integrating:  A - f- ^ For tho  numerical  calculation  '  d x  of Eg  the right  hand side  of equation (15) 1  may be evaluated by graphical integration plotting  _x) vs. x for - y • ' for the second member. The relations  • •  -  for the f i r s t member and  for-Zj  y  y, X and X*ar©  1  f •y • obtained from the equilibrium  f  diagrams*  ..The integration of .(12) -^Ids, for the'"  amotmt  of  the..  extract layer,  it ~  j  Tho maserieal v a l u e of the i n t e g r a l may be obtained by g r a p h i c a l integ-= ration, p l o t t i n g ~ J from (15).  vs. Hx* the value o f H f o r each x being obtained •  - 23 « Finally % may be obtained from. (14) by substituting i n i t the given value of ' METHOD  H,  COIM'ESRCURBMT MOXTJFLE  Sa,  Consider  food  a  column  of  value  at the top and withdrawn  at  of Hp and  L^.  STAGE;  a  total  of I? theoretical plates  ,..1 from top to bottom,  numbered.1„ 2 duced  and the calculated  A  heavy feed Eg i s  intro-  the bottom as heavy product i£J a light 3  i s introduced at the bottom and withdrawn at the top as l i g h t  produce L^. Making material balances around th® entire column f O F acetic a c i d  B  f o r chloroform and f o r t o t a l material, e  W^hAi - VK + VI HQ  *. X ^ !  %  s  (18)  * Li  (19)  leaking similar material balances around any plate n, *a-l * ^ 1 y * l n  V i W  2  *  & L  n *n  < > 20  W  *  i n addition X^, y , and Y n  n  are functions o f  in-, the equilibrium diagrams.  OS plotted  "  Under the conditions of the problem, the unknoxm i n equations (17), (18), and (19) are: a reasonable  y^  equation  (17) ana  (18)  % , 1^,  may ba  M i l be determined by y, by means of the  and T^.  solved f o r E T  Assuming  and I*,, since  phase equilibrium plot.  Substituting  th© values of  and Lj in (19), the value of  ic  obtained,. tfith  the values of L; and y-j, and consequently of T^,  x-^  and X-^ through the equilibrium diagram the equation for plats 1 may then be taken into consideration. n a  1 together  suffice to  with  solve  equations (20), (21), and (22). where  Thus  the saturation relation between Vg  for the four unknoms E ^  s  Xg, yg,  and Tp  will  and Y^.  This process i s continued until the equations for the last plate are taken into consideration, of  the assumed value f o r  the (18)  values will  of  y,,  and  and  so the test for  I f the assumed  the  value has been  corrections correct  g  originally calculated by equations (17) and  check those calculated by material balaneies around the last  plate e METHOD 5b; COMTJSRCURBMT HWIEITE In  an extraction  STAGS:  tower  of infinite height, the difference  between the compositions of the safflnate layers, or of the extracted layers,, from plat© to plate become exceedingly small at that end where the material to be extracted i s fed Into the tower. This means that the outgoing extract layer may be considered in equilibrium xrf.th the incoming safflnate layer. Using this concept and making material balances around the entire column:-  Bo^o f %  t  l  Y  H & 1  s  % % f I^Y Q  (24)  % f  s  %  * %  (25)  From the given values of x and 351, those of Z Q , y , T , 0  and 2^ are read from the equilibrium graphs.  D  0  Inserting these values as  well as the given values of H Q , yjp.i» and Ygj^ in equation (24, (24) and .  (25$-values  of % ,  L\,  and % f i a » obtained, and three values of Hf»  Lp, and S^. ESIATIvE ISTgODS OF EXTRACTION METHODS; It i s a mistake to ova3.uate extraction methods in thes«= selves, without considering the particular problem to the solution of which they must be applied, Th© general belief that eountercurrent ex= traction i s in a l l circumstances superior to cocurrent extraction does not have a sound basis. It i s a result of the failure to see that an increase in the concentration of a substance In one liquid phase Is not always accompanied by an increase in the other5 actually i t Is sometimes accompanied by a decrease. In certain cases reflux may be used to great advantage in an extraction column. In this respect there are two practical points of interesto  (1) Through the proper use of reflux i t i s possible, with a reasonable number of stages to make separations which would otherwise be impossible, even by the use of an infinite number of stages. (2) In order to separate a constituent from a mixture, the solvent used does not necessarily have to be selective towards that constituent, - I f the right type of -reflux Is-msed.  However these advantages may he obtained only at the expense of mors solvent or energy per unit changing stock and less through put of product.  SKETCH OF A COrlTIEUOHS COWTiSR-OTRREMT SSTRACTICw PL/SfT FOR OXL-SH&LE In the accompanying diagram i s a idea for a continuous - &krdra*-~t shale plant.  Shale from the mine would run along on trucks and duniped A  into tho crushers and r o l l e r s which would drop the required mesh through scrocno into a v e r t i c a l h e l i c a l scrow conveyor.  The ribbon s p i r a l would  take i t along the heated acid-proof passage, e l l along meeting solvent coming i n tho opposite direction, so that tho spent shale meets the freshest acid, end tho aoid which  i s nearly saturated with extract^meeto  fresh shale. Above the solvent entry l i n e i s a drainage portion so that the spent shale w i l l not. be too saturated.  The spent shale i s sent  to a solvent recovery p l a n t where the acid would be removed with water 8  or d i s t i l l e d . After leaving the shale the solvent passes up a v e r t i c a l tube to allow sediment to drop out, and be carried back by a conveyor Tho solvent i s then sent to the s t i l l where the extract i s removed.  Tho  l e v e l of the solvent entry pipe would naturally have to be above the  point  t"hero the  solvent  1©  removed  at  the other end.  The speed of the c onvoyor would be regulated f o r optinum  - 8? condition of as complete e x t r a c t i o n as possible i n a reasonable time* '  Tli© greatest problem would bs an e f f i c i e n t ssparation o f  the butyric-water mixture from the spent shale.  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items