Studies of Some Properties of Hydrocarbons Citation Mendenhall, William Rees (1936) Studies of Some Properties of Hydrocarbons. Master's thesis, California Institute of Technology. doi:10.7907/g0ff-aq67. https://resolver.caltech.edu/CaltechTHESIS:11132024-183212993 Abstract No abstract. Item Type: Thesis (Master's thesis) Subject Keywords: (Chemical Engineering) Degree Grantor: California Institute of Technology Division: Chemistry and Chemical Engineering Major Option: Chemical Engineering Thesis Availability: Public (worldwide access) Research Advisor(s): Lacey, William Noble (advisor) Sage, Bruce H. (advisor) Thesis Committee: Unknown, Unknown Defense Date: 1 January 1936 Record Number: CaltechTHESIS:11132024-183212993 Persistent URL: https://resolver.caltech.edu/CaltechTHESIS:11132024-183212993 DOI: 10.7907/g0ff-aq67 Default Usage Policy: No commercial reproduction, distribution, display or performance rights in this work are provided. ID Code: 16853 Collection: CaltechTHESIS Deposited By: Benjamin Perez Deposited On: 14 Nov 2024 17:00 Last Modified: 14 Nov 2024 17:22 Thesis Files PDF - Final Version See Usage Policy. 13MB Repository Staff Only: item control page

The~is by W.R. Men:d&nhall In Partial l?-ulfillment of tbe Requirements for the Degree of lmste-r of Sci.eJ1ee, California Institute of Technology Pa-sa,d-ena. Cali forn1a 1936 The Joule 111.h<"'lmson Coaf:ricients of I so - Butanf'l 38 Pa.rt I Vi: SCOSI 'l'Y O:lf t . ~- ,Ciu1130lf SO'.Ltn! ONS J ·I VISCOSITY OF HYDROCARBON SOLUTIONS Solutions of Four Hydrocarbon Gases in a Crude Oil BY B. H. SAGE, W. R. MENDENHALL, AND W. N. LACEY OF THE CALIFORNIA INSTITUTE OF TECHNOLOGY PASADENA, CALIF. Reprinted from Production Bulletin No . 216 American Petroleum Institute November 1935 Issued by Division of Production American Petroleum Institute 1508 Gulf States Building Dallas, Tex. Price, 15 Cents Viscositv., of Hydrocarbon Solutions , Solutions of Four Hydrocarbon Gases in a Crude Oil B. H. SAGE,* W. R. MENDENHALL,* AND W. N. LACEY* In previous studies the viscosity of solutions of meth-ane, ethane, propane, and nbutane gases in a highly-refined hydrocarbon material, known as crystal oil, has been determined. It is the purpose of the present paper to report the results of similar studies when the same gases were dissolved in a crude oil. This work was a part of the investigation of the behavior of hydrocarbon mixtures being carried on as Research Project 37 of the American Petroleum Institute. Materials Used The crude oil used in this study was a blended sample produced from several zones in the Santa Fe Springs field in California. The sample, wliich has been used in other studies,'· 2 • '" was kindly furnished by the Shell Oil Company, Los Angeles, Calif., together with an anitlysis made in the following manner: Three liters of the sample was placed in an iron still (9 in. in diameter by 10 in. high) equipped with a 2-in. diameter vertical pipe rising 4~ in. before the delivery or condenser tube branched off. Ice water and ice were used to cool the condenser tube. The crude was topped with fire alone until a still-head vapor temperature of 518 deg. F. (270 deg. C.) was reached. The distillation was then continued with the injection of steam until the residue in the retort had a Saybolt Furol viscosity of 120 sec. at 122 deg. F. The combined distillate was then charged to a " Badger-type " distillation column, consisting of a 3-liter electrically-heated pyrex distillation flask discharging vapors to a· J_-in. diameter fractionating column 4 ft. long filled with glass beads nearly to the vapor outlet tube. The column was surrounded by an electrically-heated jacket. The overhead vapors passed to a glass spiral-tube 'Condenser cooled with ice water. In this column the material was separated into gasoline, kerosine, and stove-oil fractions, the latter being the residue in the flask. Each of these fractions was then subjected to ASTM Engler distillations. The results of these distillations and various determinations upon · the original oil and the several fractions are summarized in Table 1. The methane used wa s obtained from natural gas by fractional-condensation and charcoal-adsorption methods previously described • with the addition of a fractional conden sation directly as solid at liquid-air tern• Ca lifornia Institute of Tech n ology, Pasad ena, Calif . • Fig ures r efe r to bibliograph y on p . 7. perature and a pressure of 2 in. of mercury, absolute, to aid in removal of nitrogen and similar impurities. The resulting material gave an analysis in terms of mol percentage's as follows: 99. 5 per cent methane, 0.02 per cent ethane, and less than 0.5 per cent of nitrogen and similar gases. The ethane was prepared from a sample of crude ethane (approximately 90 per cent ethane) by fractional-condensation and charcoal-adsorption methods. The material used contained 0.5 per cent nitrogen and methane, 99.2 per cent ethane, and 0.3 per cent propane and higher hydrocarbons. The samples of propane and nbutane were used as obtained from the Phillips Petroleum Company. Analyses of tliese two materials were furnished by that company. The former was 100.0 per cent propane, while the nbutane sample contained 0.18 per cent isobutane, 99.14 per cent nbutane, and 0.68 per cent isopentane. Apparatus and Methods The apparatus used for the measurement of the viscosity of the saturated solutions has been described by Sage and co-workers.3· • The viscometer consisted of an inclined tube, filled with the solution to be investigated, and a closely-fitting, spherical ball which was allowed to roll down the tube from one electrical contact to another. The roll time of the ball was determined by means of a chronograph. The temperature of the viscometer and contents was kept within 0.02 deg. F. of the desired value in an oil thermostat bath. Equilibrium between gas and liquid phases was attained by mechanical agitation and re-circulation of the liquid through the apparatus. The equilibrium pressure in the apparatus was measured by means of fluid-pressure scales having sensitivities of 0.1 lb. per sq. in. up to 300 lb. per sq. in., and of 1 lb. per sq. in. for higher pressures. The sample of crude oil was placed in the apparatus, fillin g about two-thirds of the total volume. On account of the volatility of the oil, it was not feasible to remove the residual air from the apparatus by evacuation. The reported pressures represent, therefore, the increase of pr essure due to addition of gas. The gas was added in a succession of quantities, the system being brought to equilibrium for measurements after each addition. The necessary values of the density of the saturated solutions were taken from previously-published data for the propane solutions• and from as yet unpublished measurements by the authors for the other cases. 4 SOLUTIONS OF FOUR HYDROCARBON GASES IN A CRUDE OIL TABLE 1 Analysis of Crude-Oil Sample Gravity, 34.8 deg. AP! Water and sediment, trace Sulfur, 0.39 per cent by weight Viscosity, Saybolt Universal, at 100 deg. F., 42 sec.; at 130 deg. F., 39 sec. Products Gasoline Per cent of charge (dry) .............. . ............ . 37.37 Gravity, deg. API ............. . .................... . 53.2 Flash point (closed cup), deg. F ........ . . . ......... :. Viscosity, Saybolt Universal, at 100 deg. F., sec ....... . Viscosity, Saybolt Furol, at 122 deg. F., sec ........... . Absolute viscosity (at 68 deg. F.) centipoises . ......... . Sulfur, per cent by weight .. . ... . . .. ..... .... ....... . 0.047 Octane No., ASTM ................................. . 59 Pour point, deg. F ................................. . Distillation Summary, in Deg. C. Initial boiling point. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-per-cent point . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20-per-cent point . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30-per-cent point . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40-per-cent point . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50-per-cent point . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60-per-cent point .................................. . 70-per-cent point . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80-per-cent point . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90-per-cent point . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . End point . : . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Per cent over at 100 deg. C. . . . . . . . . . . . . . . . . . . . . . . . . . Per cent over at 300 deg. C . ... . .. .................. . Per cent recovered................................... Kerosine Stove Oil Residuum Loss 6.83 38.6 182 18.20 33.3 230+ 42 36.20 17.3 230+ 1.40 120 2.16 0.125 0.33 0.79 85 52 88 99 110 120 130 143 136 170 185 205 220 223 224 225 226 227 228 229 230 233 242 258 272 275 278 285 290 297 307 323 345 387 98.0 64.0 98.0 294 20.5 98.0 1.0 Experimental Results The viscosity of saturated solutions of methane in the crude oil was measured at five temperatures from 100 deg. F. to 220 deg. F. and for equilibrium pressures up to 3,000 lb. per sq. in. At the higher pressures, appreciable quantities of the more volatile constituents of the crude oil were transferred to the gas phase as a result of proximity to the critical state of the mixtures under these conditions. In order to render the results for different temperatures comparable to each other, the same amount of crude oil was used in each set of measurements. The experimental results for the viscosity of the methane solutions, as a function of equilibrium pressure for the series of different temperatures, are shown in Fig. 1. The same data are re-plotted in Fig. 2 to show viscosity as a function of temperature for a series of different equilibrium pressures. Table 2 presents interpolated values of viscosity for convenient pressures at each of the temperatures at which measurements were made. In the solutions of the three other gases-ethane, propane, and nbutane-the maximum concentrations studied were limited by the appearance of a second V) w V) 0 Q. :; .J ~ ...>- iii 0 u ~ > ...:::> w .J 0 V) ID < 5 500 1000 1500 PARTIAL PRESSURE 2000 2500 LBS . PER SQ. IN . Experimental Results for Viscosity of the Methane Solutions. FIG. 1 I, 5 SOLUTIONS OF FOUR HYDROCARBON GASES IN A CRUDE OIL liquid phase. This new phase, probably asphaltic in character, was semi-plastic, and prevented further measurements at higher concentrations. For these three TABLE 2 Absolute Viscosities of Solutions in Milli poises Q) ;., . ;::s ..0 ~ t:,-:~ Q) ::-i Q) 1--( µ_; µ_; µ_; µ_; ~ oil Cl) i::: oil Cl) i::: OD ~ oil Cl) i::: oil Cl) i::: 0 0 0 M 0 0 rl rl O> r-i c-., c-., 13.3 10.6 12.1 11.2 10.4 9.2 8.1 7.1 (6.3) (5.7) 10.0 9.4 8.8 7.9 6.9 ( 6.1) (5.4) (4.9) 11.3 9.7 8.5 7.5 6.6 (5.8) (4.8) (4.0) 9.4 8.4 7.5 6.8 6.2 ( 5.7) (5.0) (4.5) (3.8} 12.2 11.1 10.1 9.1 7.4 (6.1) (3.8) 10.0 9.4 8.8 8.3 7.3 (6.5) (5.1) 12.1 11.1 9.1 7.3 (5.7) ( 4.1) 10.0 9.5 8.4 7.4 (6.4) (5.4) ( 4.3) p... ~ • -oi - o en O' VI w ~ Ul VI ;., 6 a.. :i Cl) '1l 'H p... p... 0 r-i ...J 0 35.3 NO SOLUTE 23.7 17.3 200 400 600 1,000 1,500 2,000 2,500 3,000 29.0 25.3 22.6 18.8 15.6 13.4 11.6 10.4 20.3 18.3 16.8 14.7 12.5 10.8 9.4 8.3 100 200 300 400 500 600 800 1,000 1,250 23.5 16.7 12.3 8.8 (5.7) 17.9 13.9 11.0 8.8 6.8 (5.0) ~ >- ~ Cl) i::: 0 IVI METHANE 0 u VI > w I:, ...J 0 VI (lJ <l: 15.4 14.2 13.1 11.5 10.0 8.7 (7.7) (6.7) 175 TEMPERATURE "F ETHANE Viscosity of the Methane Solutions as Related to Ten1perature. FIG. 2 14.1 11.5 9.5 8.1 7.0 6.1 ( 4.5) PROPANE 25 50 75 100 150 200 300 26.5 18.3 12.3 8.1 19.7 16.3 13.2 10.4 (6.3) (3.5) 15.3 13.5 11.8 10.2 7.5 (5.5) nBUTANE 250 PARTIAL 500 750 PRESSURE 1000 1250 LBS. PER SQ. IN . Experimental Results for Viscosity of the Ethane Solutions. 10 20 40 60 80 100 125 19.3 15.1 7.9 15.4 13.6 10.1 (6.9) (3.9) FIG. 3 gases the large increase in volume of the liquid phase with increasing saturation pressures necessitated the withdrawal of portions of the liquid phase in the course of a set of measurements. The experimental viscosity values for the ethane solutions, at three of the five temperatures studied, are shown in Fig. 3. Table 2 presents values for even pressures. The viscosity of these solutions is plotted as a function of temperature for a series of different pres- 1 6 SOLUTIONS OF FOUR HYDROCARBON GASES IN A CRUDE OIL sures in Fig. 4. The highest equilibrium pressure in the case of these ethane solutions was 1,600 lb. per sq. in. at 220 deg. F. the study, the vapor pressure of nbutane at 100 deg. F. being too low to permit study at that temperature. The maximum pressure was 160 lb. per sq. in. at 220 deg. F. V) w 25 V) 6 II) w II) ~ ..J ..J 6 0. ::::; ~ ..J i >':: V) ,_>II) 0 0 u V) > u 15 II) > ,_w w I- :> ::J ..J ...J 0 0 II) V) (I) <D <l: ..: 150 175 200 125 TEMPERATURE 175 150 TEMPERATURE Viscosity of the Ethane Solutions as Related to T e1nperature. 200 "F Viscosity of the Propane Solutions as Related to Te1nperature. FIG. 4 FIG. 6 II) w II) 6 0. ::::; ..J i ,_>- >- in ':: 0 II) u 0 II) u > II) > ,_ ,_ w w :> ..J :> 0 ..J II) 0 < (I) m 5 II) < 25 PARTIAL PRESSURE LBS. PER SQ. IN . PARTIAL 50 75 PRESSURE 100 125 LBS. PER SQ. IN. Experimental Results for Viscosity of the Propane Solutions. Experimental R esults for Viscosity of the nButane Solutions. FIG. 5 FIG. 7 The corresponding values for propane solutions are given in Fig. 5 and 6 and i~ Table 2. The maximum pressure in this case was 400 lb. per sq. in. at 220 deg. F. In the case of nbutane solutions, only four temperatures from 130 deg. F. to 220 deg. F. were included in Plots and a tabulation similar to those for the other cases will be found in Fig. 7 and 8 and in Table 2. A comparison of the effect upon viscosity of dissolving separately each of the four gases in the crude oil is shown in Fig. 9, for equal equilibrium pressures at I ' 7 SOLUTIONS OF FOUR HYDROCARBON GASES IN A CRUDE OIL V) w V) 6 25 Q, .J .J ~ 20 >- t-- ...>iii iii 0 u 15 V) 0 u V) > > w ...w:, t-::, 10 .J 0 _J 0 V) V) (D <( cD <t 5 125 150 TEMPERATURE 175 200 "F PARTIAL PRESSURE LBS PER SQ. IN . Viscosity of the nButane Solutions as Related to Temperature. Comparison of the Viscosity of Solutions of Different Gases in Santa Fe Springs Crude Oil at 130 Deg. F. on the Basis of Saturation Pressures. FIG. 8 FIG. 9 130 deg. F. The smaller decreases in viscosity for the gases of lower molecular weight are partly due to their smaller solubilities in the crude oil on the basis of equal saturation pressure. A corresponding comparison on the basis of equal concentrations of dissolved gas was not attempted on account of uncertainties in the compositions of the various solutions when as volatile a solvent as this crude oil is used in the presence of a gas phase. assistance of D. C. Webster in the preparation of the figures is gratefully acknowledged. Acknowledgments The aid and encouragement of the American Petroleum Institute is much appreciated by the authors. The BIBLIOGRAPHY W . N. Lacey, A .P.I. Production Bull. No. 210, 16 (1932). W. N. Lacey, B. H. Sage, and C. W. Kircher, Jr., Ind. Eng. Ohcm., 26, 652 (1934). •B. H. Sage, Ind. Eng. Chem. (Anal. Ed.), 5,261 (1933). • B. H. Sage, H. S. Backus, and W. N. Lacey, Ind. Eng. Chem., 27, 686 (1935). • B. H. Sage, W. N. Lacey, and J. G. Schaafsma, Ind. Eng. Chmn., 26, 874 (1934). 6 B. H. Sage, J. E. Sherborne, and W. N. Lacey, Ind. Eng. Chem., 27, 954 (1935). 1 2 Part II TRE ~«.'iDYi'.ifA!tr'.l C l~Ro.FERTI ES O)1 N- ~ur.m /0 Introdu-etion In recent developments of petroleum production and refining~ the applieation of the thermodynamic properties a.:f the oils and gases ,. and. t hose of their components, to design problems is constantly inereasill€ . !'.ne pr operties which are the most useful in such ealeulations a.re, be - sides the pressure•volume-"'tempera.ture rel s.tioxu;, the heat content , entropy, and fuga.e1 ty. The P--V-T rel ations have been studied experiment- all;y and the other quantities calcule.t ed fo:r normal butane. .Also values of the internal energy have be-en calculated, although tb.e-y are not as use:f'ttl as the heat content i.n eng ineering celeula.tions. The range covered was from 20 to 3000 pounds per square inch absolute pressure ., and from 70 to 220 degrees Fabrenhe1 t. ln oi-der to be able t0- calculate tbe therm.odynsmie properties of a s"nbste...nce. the necess~.ry and sufficient data a.re the press-ure-volune-tempera.ture relations and one mea..su.rement of the change of internal. energy with tempera ture throughout the rm'lge to be covered.. Two sets 0f data were available from which the latter could be ealculated, namely the specific heats at constant volume. and the Jeule Thomf'ion coeffidents, both of which have been previously published (1 , 2) . T'ne press,u-e-volume- tempera.tv.re relations have been stut:l.ied a.--id &re here reported.. Ap:p-aratu$ and Procedure !he work was done as pert of the Ame-riean Petroleum Institute Re,ear-ch Jrrojec.t Number 37 an,d. all of the apparatus and procedures have " been fully described in p reviously published pape rs (3 9 4,5). Thsy will therefore only be bri efly tre a ted here . The pressure-vol ume- tempera ture measurements were made in a varia.'ble vol ume cell. The pressures were measured by 1.1.se of fluid pressure seal.es,, the values below 300 pounds per square ineh being good. to 0 . 1 pound, thos:e from 300 to 3000 pounds being good to 1 potmd per square inch . The t empera.ture w~LS held constant a t the desired value during a nm by mean~ o.f an oil bath, whose temper a ture was automatically controlled to Vii t hin 0 . 0-1 degrees F. 2ro~ F. ffhe tempera t u res chosen were ?O,, 100,. 130, 160., 190,. and The volume of the s.a.mple u.ncter eonsid.-eration was eon.trolled by the add.ition or retr.oval of me:rcury from the cell . The measureme'Rt of t he volume was accomplished by loca t ing accurately the top of the mereury witli an electric eeatact.. 'l'.1le eont aet was moved verti~ally bf a care- fully maeh.ined ge-a:r JJ. 1&chanhm t-o whi eh a counter was atte.ched. The v Gltm(:l oemllpi&'d by the st11.mple could be then measu:re,d to within 0 . 0012 cubic- inches after S'lli t able eali'br~tion .• mich ,;.a.,v-e an accu.:raey of o.19t A constant voliame. adiabatic calorime.te-r w-8--s ,aged for d-etermining t lle speeifie beats . The a.pparat1.t s consisted of a 'bomb {vm.-ose volume was an evacuated space fl-mu a ring by thr-ee fine s t-eel wires-. The pre-s sure -1 in the va.enum ehB;l.ll.ber was maintained a t erpproximately 10 - milli meters o-f The v ~ spa.ee was SllI"l'ounded. by an oil batho. the temperature of' wii ch was kept 'within 0 . 02'0 of tb.at of" the surface of the bomb by a hea ter whi eh was controlled automatically by a phot o- e l ectric circ:ni t. This circuit consisted first of a t her mocouple" one junction of which v.ras on the surface of the bomb , t he other i n the oil 'bath. The electro.mo tive fore.es produced in the thermocouple de fl ected. a galvanome ter from whieh a light beam was reflecte d to a photo el e ctri c cell . t he beam .fall1.ng on t he cell v.hen the surface of the bomb and the oil ba th were a t the same temper ature , and falling to one side when the tempera tures were different . The cell .. t hen, actuated relays which controll ed the current to the bath heat er. The on in the bath was circvlated so that temper ature eqm.li- brium in the bath was r apidl y a ttained. Around the oil bath there was an ai r thermostat. whose temperature was k ep t within 2~F of that of t he oil bath manually, using electric h ea ter-s and a t h ermometer . In order to assure equUibrium wi th.hi the bomb . i t was agitat ed by r ocking an arm, a ttached ri gidly to the ring support:i.ng the bomb, by elect romagnets external to the va.cmmi chamber. '11l.1e ter-.1perature of the bomb , after equilibrium had been r eached, wa.s measured t o an accurncy of 0 . 002q,F . by means of a thermocouple~ t he cold ,junction of m ich wa.s kept in a eon• s tantly @€i t a ted ice and water rnixt1lre 1 and the hot j1:mction of which was in a. well in the bomb . The electromotive forces produced were :re·a d by means of a potentiometer •ose scale was divided. into 10,000 dividons for a ran-ge of -5 millivolts. eaeh reading being estimated to 0.10 of a. division . \I/he pr .oeednre was to weigh a sample into the bomb, measure the te-m- pe;rat,u,e after eqnilibrliim w:a.s established (aa determined. by no furthex-change in the tempe:rature for a.t lea.st five minutes), add heat electrically for a period of t1me 0 agi t ate until equilibt•i-uin ,...ar; again reached. and measure t he new tempert;..ture. This proces s was then continued through• out the temperature range to be covered . The &mount of h eat added. was measured by determining the current and voltage in the heater circ~~t and the length of the heating perlod. In this manner Cv e.s a function of tempera:t1.tte could. be de termined for a. given sample throughout the whole te114?e-rature range without removing the sample :from the apparatus. The Joule Thomson coefficients; were determined, using the a.pparatu,s descrj_bed by Sage and co-workers (5) and briefly treated eleewhere in this thesis. The butane used in all this work we~s furnl shed by the l?hilge.s Campany, and was reported. as being 99 .21 pel.'eent n-butane, O. l ~ ip- buta.ne-.. and 0.61% iso-pentane . In the esl.culation of values of heat content, specific heats at constant pressure a.re more 'V.sef-ul than at constant vohnne, consequently. the values obtained from the experimental. measurements were conver-ted to the constant pressure basis. The amount of gas phase present in the bomb was small in coinparis~n to the amount of liquid phase.,. an.d therefore. the heat absorbed by it was a very small :pert ion of the whole a nd did not need to be corrected for. The equa.tions 'ti.sea fo:r the other corrections we-re C - rx ~Xs.- r, X, + s-~TM6T T (1) an<i C - C + /~V\,dP.T ,, - .) ""\dT.l.P, d T (2) Q (See list of s;rmbols used for definitions tJf symbols) Equation (1) was used to correct for the amou..."lt of heat used. in the vaporiza ti on e:f:' a -poi"tion of the liqtud upon rl se in temper a ture, after which equation ( 2) corrects the values from those of satura tion to ones of constant :pressure . These equations can be de rived by consideTation of appropriate cycles (6). Values o,f Op calculated in this way have been published (1) and were used in the c.alculation1:J. clescri bed below. Iso--ba.ric changes of heat content can be c alculated by integration of the e q:uation 3 and those of entropy by dS= CtedT T In these calculations . since it would be necessary to make measurements of the properties near absolute zero to be a.b le to evaluate absol ute values of heat content and entropy, the reference state was t.'.!1ken as 3000 po'Ulld@ per square inch absolute pre ssure end 60° F. a.t which state the heat content and. entropy were set equal to zero . The above inte- gra tion was ea,rried on along the 400 pound pet square inch line. The values 0f bea t content were then d.e termined throughout the condensed by integra ti on of the equat:1:on ~ ~ dH =Lv-T(:-f4Jd P 5 along t he 60(,J F . isotherm from 3000 pounds to 400 pounds par square inch., adding this value to that obtained by the inte,:i:ration of equa.tiou (3),. as above, arii subsequent integr ation of equation (5) al.on~ the various isotherms from the 400 pounc p oint. A similar calculation_. using the reS"J.l. ts of the above 1ntegra.tion of equation ( 4.) and of the equation d s- - (~'dP - aTh 6 gave the values ~f entropy for the condensed region . Jq the addi tion of the heat of vaporization. or the entropy of vap orization_, :f'rom the corresponding value at saturation obtained above, and ecmtinuing the same processes. one could complete . the calculations t h,r:Ollgbout the superheated regi on also . Bowever, by choosing a new reference p oint and using different methods of' cal cul ations for the vaJ:ues in the superhee.ted region, a new set of saturati'.;n values could be obtained 'll'lhieh sho1-ud be different from ea.ch other by a constant value after the proper a;pplic~tion of the values for the vaµorization process ha.d been used.. This would render a check on the accuracy of the ealcw.ations or values of the heat o:f ,raporil!:a.tion . Such .-oro.ced.ure ,. wzas followed . The new reference point i;e,s chosen as atmospheric pressure and 60Q F., again the val~es being assumed to be ~ero at the reference point . Instead of u.sing equati0n (5) for the hea.t content in this :region. the d H = - C ,.( egrw:..tion !!)d P 7 was us:ed,, since the values of the Joule Thom.son coef:ficient 1 {! ~) H were known . (! ~ ,,,µ. = This was especially valu.a ble since the values of )P for the su-parheated region were not considered to be as accurate as those of C and p since the former were obt.ain.ed by graphical differ- entiation of the isobars on the volurne-temperatu.re charts. v reeso», us• was mad.e of the (~~)~on · { ! ! for values t>f For the smie h (!i)..to be used in the integrat:J.on ef equation (5) in the supe,rhea teri region. For the qua.li ty, volmne. and_ hea t content lines on the t~er:.;.ture- entropy diagram. in the two phase region. the rel a ti ons _ V...; V' 'I - \j'' - \/' t,.,. s~ ~ s , -t-=;= were uBed. ~ definition. g.; lil i-PV • conse quently the calculation of the values of internal energy entails only the subtraction of the PV p roduct f:rom the valuet;i of heat content as obtained above . iJ:h.e fugacities were calculated by t,he proper integration of the expression lt\f""Ji-1.dfn~ 0 in which In t his c aleu.l a t1on. j: was c onsider e o. to be c ons t Rn t bel ow 10 :p®nds per square i nch absolute pr essure, and. a t zero pr e~sure the f'ugaeity \\'as assumed equal to the pre ssure . Resulh 1'a.ble I is the complete t tilmlation of the physical and t lle:rmod.yn81n1 c properties o:f n - butane ~-e obtained. from the pre sent, work , excep t that the i nternal energ ies are not i n cluded . since t hey can be ea sily cal culated. from the t abul a ted values . Table II g ives the properti es of saturated li qui d and gas f or even pressures . The press'!lre-volume-temper a,ture rel a ti ons are p resente d in the f irs t t hree f i gures . gion . Fi gure 1 is & plot of the isobar s in t he super heated re- The ne xt figure shows speci f ic gr avi ty a s a function of presS'\ll'e for the series of t em-per 2.tures. The devi a tions from a per f e ct gas are shown by the l?V / RT chart vs . P chart i n figure 3. since a perfect gas behavior would be represented. by a str ~.igh t . horizontal Hne . The triangle in the upper left hand corner r epresents a value taken f rom an at mospheric densi t y- measurement a t 104°F . This furnishes a point obtained by a dif- ferent me thod f rom the rest of the dat a and is a good check . The deter- mination wa.s made by weighing a sample in a bomb, wi thdr a.wing a de:fini te volwne , and then rede t ermini ng the weight. On Figure 3 t he values for 70 cle 1!;rees a re not shewn since they are not fe asible to present on the same s cale as t he other vi:;1.lues. due to the pre ssures being so low. Values of heat content a s a :func:tion of temperc:.ture at a series of 2res~ures in the superheated regi on are shown as Figure 4 . diagram for t he liqui d res;ion h A similar not shown si nce t he lines lie very cl ose together and cross e ach other at the hi,?:her te!!:rpera ture s . On the temper- B,ture-entronv di Mram. Fi~e 5. t he va lues of vres su.re. vol ume 9 and 17 tempera ture are shown, as well a s those of qu&lity in the two phase reglon . A he.at conten.t-entropy dh,eram, or Mollier char-t~ was not included since interpolation on it woul cl 1, e very diffict,1 t,, due to t h e position a:nd slope of the lines . ! It should be noted., however, tha t the entropy of satu.raterl g as is almost i n dependent of hea t content, whi ch wo'l1ld. make the Mollier chart very d.i:fferent fr om tha t of water . The whole range of: :pressures was not 1.nc).-b1ded in the internal energypressm·e diagra.•11, Figure s. since beyond the range shown the values continue nea rly linearly . As for the fuga eity val11es, in the superheated :region fuga e-i ty divided by pressure is shown plotted. ~a:inst 1)ressure in li"'igure 7 while :figure 8 d.e-pi ct s fugE,c5. ty of t he condensed Uqu.:i.d as a funct;1on ~f pre ssure for the series of temperi?,tures . 11'1.gure 9 is a representa.ti on of e, re la.ti on fol lo·~ed. by hy<l.roca.rborrn, namely that i f the reciprocal of the absolute tempera ture is 13lotted against the logarithm of the c orresponding vapor pressure, the result is a line that is almost straight . Ordinarily thh woul d. :Indica te tha t the gas follo ws the p erfect g a.e law~ a nd. that there is no c.~ange in the hea t of "/J"aporization with tempera ture . :Butane follo ws neither of these s ti p -- ulat'ions, but the deviations from the perfe ct ga s l aws eo1:1:pensa.te for the change in the heat of vapori M ti on, . eouse quen tly the relation r ernains nearly linear . This rel ation has been used for hydroef>.roons in general to extrapolate va.por pressures ·beyond the eritieal tempera ture f~l'" the ur,ie of Ra oul ts law in tha t region . TJ!.BL1C I 700 (31.67) Satd. vapor V f/p h 2. 707 o.9453 152.01 f/p s 0.3165 1.6988 0.9183 Satd. 1i quid O. 02777 0. 9453 20 4.7059 0.970 153.59 Superheated, Gas a.nd Condensed Li quid Regions 130° 160° 100° (52.81) 0.3341 5.0173 h y s (82~59) f/p h 190° s v f/p h 161~3.2 0.3175 1ao79 o~aea:a l 70.M 0~3194 o. 7419 o.as1s 170~91 14.14 0.0538 0HJ2962 0.8882 32.71 0~0858 0.8618 0~03080 o i i222 ·o. B05l 0.9755 165.20 0.35!54 2~·5649 40 0.9545 175~31 0.3504 2. 7386 0~9645 220° (249.2) (178.5) (123~7) 187~86 0~3113 \2.9023 f/p h s V f/p h s o.a30a iss.97 o.3249 o,3444 0.0010 194. 41 0.3273 ?3.3:8 0.),506 0.03443 0.8010 95.61 0.1833 o.9aso 202.19 o • ..t1ss a.2202 o •.9925 215.31 . o.4363 0.9743 200.84 '0.3890 3.0681 0.9840 214.19 D.4li 7 0.9675 199 ..34 0.3764 2.0059 0.9736 212.95 0.3964 0.-9386 197.67 0.3646 l.4646 0.9584 211.00 0.3851 0.9181 195.85 0.3549 1.1435 0.9423 210.ll 0.3759 0.8912 193.M 0.344-4 0.8802 0.9208 208.10 0.3661 o. 3.'350 o. 7006 0.8973 205.86 0.3575 . ~ 60 lt6~$ 0.9250 173.16 0.3338 1.7689 0~94:;i 186.0-1 80 1a~ Ol8925 170.58 0.3209 l.2706 0.9196 184~07 Oi ~ - 0.9766 0~8945 181~85 ... 100 . O~:J $3 ll.8875 ; .; ~·.3764 , ;· •'. · . •; 4 125 o••l •.4•.2··•.-:r: .:.·•.·.·9. •·.·.•.·•.•.•.·.•.·.·f·•:··l ··r.0125 Oo." 2., 150 ••r,-· . • 6428 0.86.'33 190.55 115 ' • . • :0.5194 0.8348 187.44 0,.3261 0.5768 0.8'731 203.40 0 .. 3495 0.7459 73..35 0.4799 0.8485 200.67 0.3419 0.4043 0.8?,.'31 197.65 0.-3345 200 •• 3.·.·•.•.•.. :'<: ' · , ... 0 .. 02?68 0.1573 32~89 -2.92 0.0211 0.02851 0.2525 0.0850 0.03066 0~5445 0 .;. 225 •tft = - ,.J 00 14~90 0.0519 C>.02938 0.1999 33~23 0.0838 0.03037 0~28'11 0.0940 15~37 0.0509 0.02,911 0~1406 33.60 0,.082(; 0.03012 0~2019 52~82 0.1144 0.03142 0,02740 0.0470 -1.27 0.0185 0.02810 0.0744 15."88 0.0499 o.~Q,~~J> 0.1112 34.01 0.0014 0.02992 0~1594 53.09 o.uso( o.o:n10 1000 0.02?30 0.0394 -0.70 0.0176 0.02799 0.0629 16.40 0.;0490 0.028~ 0;.0939 34.45 0.0808 0.02972 0.1344 1250 0.02718 0.0338 0.03 0.0166 0.02784 0.•0538 17.07 0.0479 0.02851 0.0819 35.04 0.0791 0~02948 1600 0 .. 02?07 0.0302 0.769 0.0157 0.•02772 0.•0480 17.76 o.,0468 0~02845 0.0714 35.€:" 1750 0.02698 0.0277 1.51 0.0147 0.02760 0~0440 18.47 0.0458 0.©2829 0.0653 2000 0 .,02687 0 .0260 2.27 0.0138 O.O~t748 19 .20 0.0448 o.oia13 2250 0.02679 0.0248 3.03 0.0129 0.02738 0.0391 19.:94 0.0439 2500 0.02670 0.0239 3.81 0.0121 0.02728 0.0376 2750 0.02661 0.0232 4.59 0.0111 0.02718 0 .•0365 0.02709 0.0357 400 0.02759 0.0833 GOO 0.02750 0.0590 800 3000 £1.p 0.02655 0.0228 •l.84 0.0193 -0.0282!3 5.38 0.0103 pressure lb. per $q. in. abs.; v 0.0411 0.1504 73.12 0 .. 1482 0.03382 0.5197 94.,98 0.1809 0.2764 73.01 0.1463 0.03328 0.3656 94.35 0.,1782 o.21a4 73.oo 0.1445 0.03278 o. zs90 93.,91 0 •.1757 53~40 O~lll.8 0.03084 0.1841 '73.08 0.1428 0.03238 0.2436 93 ..60 0.1735 0~1147 53.84 0.1103 0.03052 0.1569 73.27 0.1409 0.03191 0.2077 93.36 0.,1710 o.om 0.02928 0.1020 54.33 o.toaa o.ozo2e 0.1393 ,73.55 0.1392 0.03152 0.1844 93.25 0.1687 !36.31 0.0768 0.02906 0.0931 54.86 0,.1075. 0.02999 0.1272 73.89 0.1375 0.03ll8 0.1682 93.2.'3 0.1665 0.0610 36.98 o~075? o.ozssa o.oss9 55.42 o.tos3 o.02976 o.11a4 74.29 o.13so 0.03009 0.1566 93.?9 0.1545 0.02800 0.0578 37.67 0.0747 0.02871 0.0822 56.02 0.1051 0.02954 0.1118 74.73 0.1346 0.03060 0.1480 9:3 . .42 0.1626 20.69 0.0429 0.02788 0.055f. 38.37 0.0737 0.02856 0.0788 56.65 0~.1 040 0.02934 0.1072 75.20 0.1332 0.03034 0.1416 93.59 0.1608 :U.45 o.•0421 o.om7 o.oroa 39.10 0.072? 0.02840 o.o7s2 57.31 o·.1029 0.02913 0.103s 75.70 0.1319 0.03009 0.1367 93.ao o.1sso 22.21 0.0412 0~02'166 0.0525 39.84 0.0718 0.02827 0.0743 57.98 0.1019 ~ -02898 0.:1.009 76.22 0.1306 0 .02%9 0.1~~ f f~aeity, lb. per sq. in.; h heat content. E.t.u. per lb.; ' s entropy. 13.t.u. per lb. per ° F . abs. 94.05 0.1573 I I specific voh11ne, eu.• ft. per lb 1 ' PEYSI CA.L MW TH.'E1RH.il . FROPElETI 1".'S OF N- BUT.l\NE Se,turated Liquid P.,n d Satura ted Gas Sa turate d Ga.s Sa turted. Liquid h e p ta ro 44. 30 40 82 . 85 0 . 9338 2 . 2069 1 i:-.s . 1 0 . 3158 0 . 0 ~,812 4.2 0 . 0353 00 108 . 25 0 . 9100 1. 50.'51 163 . 8 0 . 3.175 0 . 0288'? 19 .1 0 . 0623 80 12?. 77 0 . 8910 l.i427 1 69 . 6 0 . 3190 0 . 02953 31 . 2 0 . 0835 100 143 . 67 f/p h V s V 0.02715 •· 0 ~8759 0 . 9183 174 . 2 0 . 3207 0 .03013 41.6 0 . 1008 125 160 . 83 0 ~8602 0 . 7340 179 . 1 0.3224 0 . 03.0 84 fi2 . 9 0 . J.192 l '75 . 44 0 . 8469 0 . 6081 183 .l 0 . 3 ?.36 0 . 03151 62. 9 0 . 134,3 175 188 . 33 0 ~8323 0 . 5161 1B6. 5 0 . 3 247 0.03220 72. 2 0 . 14$7. 200 199 . 93 0 . 8205 0 . 4'45!5 189 . 5 o . ::~256 0 . 0 ~~29?. 80 . 5 0 . 1612 •. 100 225 210 . 53 0 . 8101 0 . 3894 19 2.l 0 . 3265 0 . 0 ~1366 88 ~5 0 .1729 2f,O 220 .28 0 . 8008 0 •.3433 194 . 4 0 . 3274 0.0.'.344:3 95. 8 0 .1838 a t . temperature iR g'W ., other inits e,s in Tabl e I . ;;; ~ JOULE-THOMSON CO!i.!J'F!C!'l!!h".'11$ OF ISO-.W'.M.NE Introduction Tlae Joule-Thomson. ooefftcient ,_p. . of a eu.bsta.nce is de.f ined &s the rate of change of temper&ture with pres sure when a change of sta te of the substance t e.kes place under conditions of cone.t ant heat c ontent . The. coeffici ent is of va.lue in obta.ini~ the cllange in internal anerey with pressure, which is one of the necessary pieces of data needed to ~nable one to ce.lculate the t hermod.ynaraic pro:p&rties of a subs t ence. as W8.S mentioned in part two of t his t hesis. The coeff icients of iso-b,it ~ne were me-asttred a.t temper a ture s r angi ng f r om 70 t o 220 f;F . at 30 degree intervals, the pr e ssures being va-.ried from atmo s, heric to vn t!-i.1-n ten 'fH;>"ltnns per sqtlare i.nch of the va-p.or pr essure a t the tenrperatu!'e i n qu.estion . 1'he h o- butane used i n the~e deter minati ons was furnished by the }'hilgas Company . Their analysis shom; 0 . 00 per cent pr opane , 100. 00 percent i so- bu.tt1ne . ud an u.i:mieasm·abl e t re..ce of n-bu.t ane. Apparatus and fl."ocedu.re Since i t h ~£' been do 1ae el sewhere.( 5) . t he e,p:p.aratus will only 'be deseri bed briefl y here . It consis t ed of &-. porou s alundtm1 t himble. t hr cru,'-;h which the gas w1::u,~ :forced to flow. A c m-i1 pump wa.~ used to maint(f,'.i,n the flo w, by- pase valves s,nd pi p e s beinifc i nclt1.ded t o f acilitat e the fila.intenance of co ns tant flo w. Afte.:r t he gas had passed thro~h the thi mo le~ it was r ecirculated through the sys tem, vihi ch effec t ed a grea t sav1.ng in t he amount of ga s necessary fo r a de te rm:i.nation. A l!'.B.f''tie tic valve,. 01>er- a.ted through relays by a me:rcmry manometer, ke pt tbe presnm-e d.i ffe rential a cross the thi mble consta;nt to wi t hin 0 . 5 percent . In order that the t emperature of t he manometer would not ehange wit h the tempera ture of the bath or room, and that t he ges in the apparatus would not condense in the .3 f me.nome ter . the latter was enclosed in e, s team ,j acke t ., s·o tha t its pera ture was &1 ways 212 "'F. tem- T.he p r essu..re di fferential whieh the stemrt "."" Ja cketed m~ ometer· maintained was ca libr a t ed. by the u se of another manometer, parall~l to the first. whose dif:f'erent1e..,J. cou_ld. be mea sured by an e speci ally designed verti~a l component ca thet ometer . wru; dete rmined by the hei.ght The differen tial w- ich one arm of the ste~ - j a c -:-et ed manometer reached before it touched an electric e on t ac t., closing a. ei.rev.i t which. , i n t".rn. closed t..-"1e magnetic v~,lve. The ptlmP..- magnetic v-nlve. ::md. eell cont a ining the thimble were i m- mersed in an eil ba th, mose t emperature was held nt the de$31.r ed. valn.e... with an a bsolute e;ccu.raey of 0 . 02 ~,. ,. by a control mose drift was le-ss th..,-m 0 ~-002Q:!r. per h our . Be tween the mEr,;netie val ve and the thimble was a conditi onir-.g coil, (alM i rl.lmersed in the "il ba th) Vlhich w-ae about twelve feet l ong. The t ernper ;:1.ture d.iffere~ce across t he t h i mble was measured b;r a. t h ree junction differenti al t hermocouple, there being three j'lmotions on ea.eh side of the t h i..mble. A thMe junction thermocouple, the t he .. o- coupl e bei nf.t i n ser1.es, wa.s 11sea_ so th at the electromotive forces -produ.ced. would be measurt.ble with a. grea t er d.egree of acm:tra..cy t han would be pos i Me with the eruall electromo tive forc e s wh ich 'l'lOuld "be prod.uc d by one t heri.:ro- cou.ple . The electromotive forces were mea sure d. by a potentiometer ,those SC'.&le wgs di vided into 10, 000 divisions for a r ange of 0 . 5 millivolts, being r ead to the neare!.lt division . T"ne total pre ei sure on the ,s ystem was a.eterm1.ned by e fluid pressure s c ;;iJ .e im.oS\e e.cci..,ra.cy was 0.1 pound. per sg:u.are 2.neh in 300 pounds pe-r e.q:u.a re lnch . Before any f,'as wa s added. to the system i. t ws.s ev~tcue,ted to an abso 1ute pressure of approximately 4 mi llimeters of mercury? then pu r ged tllliee with the gas. re -evacua ti ng e&c . time . With the oil ba th regula,ti ng at the desi r e d t em.p era tu:re. g z.s wa s a.do.ea. un til the pres e-ore reached the value chosen. the cam pump s t arte d~ and then the by - pass valve ~ set so tha t .;-ust enough gas wa.s flowi ng thro~h the m~·netic valve to :keep it rhythmically Gpening and. closing, ?..s could be d.eten.nined b;r watching a. loo1p in ae rie$ with the exciting e lectric circuit of the val ve . As nn.ieh as half ~-n hour was sometimes needed befoTe t h e tempe r ature differen.tial wo1.;1ld become const ant. It was found th&t equilibri um was r eaclled nm.e..11 sooner when t he f<Jnount of f:B,$ in the a ppara tus w~,s incre ,.sed f"rom de tel"miB~tion to determina tion t han \1i.hen it was d.ecrea.~ed. . Also. as soon a s satm·ation was approa ehed, the va.lues dropped be low tha t which w011ld. be expected ( probably due to tme,11 dropl ets of liquid e.p pearing in t h e ga.s ), and , consequently, no values cmlld be meamu-erl at pressures elo ser than 10 -pou.nds per square inch to t he ve,por p:remsure a t the tem:perf.t tu,re \Uader condclera ti o:n . 'l.Taces of L'llpl;l.l'i t;r., such a s oil f rom leaks into the system :f'rm!l. its $t.!l"roundin-,g bath. or from t he grea.se used on the c&m pump. car1sed mu.eh tro1:.ble , :neeessi tating frequent 11 thoroue...h cleaning of the apparatus . Such i mpu.ri. ties would often cond-enee in the thi rrJ;;le and t h en would h,;-,.ve to he removed, either by burning t hem out or by re>phwing the t hiinble entirely. :Besi. o.e s the e,rrors due to i rnpu.rHies,. the Sll'lell drift in hath tempe-r r1ture vra.s e no11,gh to be a fru:i tful Mu.ree of eeror. With a three j¥1net- ion thermoeouplec, the t.eroperatv.re measu.rements we re :not hard to obtain within satis fa.ct or y aecn.1:r.a cy . ential wa s more di:ffieul t. The me,inta ins,nce of the riresrnure differ- A di f ferenti~l of eligh tly under one pound per sq:a.are inch was u.sed thro1.,ghout t11e determ:i.na.tions , t he value ch~ing slightly f r om time to t:l me . an error in thi s v al11e. Fre quent eali bratio:ru.1 were made t0 eliminate However. the by-,pass valve adjus t ments were very delicate ~d, had to be contirro.:ally watch ed. Mechaniea l diffi culties were encountered in the magnetic valve, due to the clo se fit between the moiling pa.rt and the side \'l'&ll s. In spite of these d.ifficulties , the a ccuracy of the values as a whole is believed to be abou t one and one half percent . Calculations and Results The scale divisions on the potentiometer per degree Fahrenheit were were determined by calibr a tion of the wire~ used in the t hermocouple circul ts. These values , when divided into t h e pote ntiome ter r eading . yielded the- tem.per at'li.re difference between the insi de and outside of the thimble. The temper~.ture differential was then divided by the pres sure d.if:feren tia.l, giving the Joule - Thomson coeff icient . The coefficients have been ca.lcnlated by the above meth od. e:..nd are t ~hulated. in Table I. They a.re shown plotted. a.gainst pre ssure in ~'igure! • sh¢wirig the experi mental points . Figure 2 shows them internola.ted a.ga.1.nst temper ature. In ~nerru, the curvature of the Hnes in Fi gure 1 is somewhat gre ater than tha.t ,.,f similar curves for the other hydrocarbon gases which have been studied. and re-p$rted. lt is of the same general shape as woul d be expeeted for ~ases in the region tn which this work was done, namely that of low lfeduee cl pressure and temperl¼tu.re. 'llhere a. reduced value c onsists of 'the actual. vti>.l u:e divided by the critical value. At higher temper,&- turei(. the slope of the line would be expe cte d. to become less until finally. not far 'below critical temperature it woul d change in sign. Above cri tienl tempera.tu.re, the slope woltl d. he negative, and the lines become ne&rly linesr gt still higher tempera tures . ga.s . The above remf'. rks perta in only to the For t he corresponding U quid val11es a.t the low tempera tures; the lines on a similf.a- }.A vs. P plot woµld be straight and h orizontal. However• as t he tempel"a tur-e rises, they show a.never increasing nega tive slo-pe a nd eurva.ture, u..ntil t he critical temperature is rea.ched. A.bes . 70QF Lb . pe r Sq. In. • ]'. per 130<>F 100"1' 190°F 160°1 2Z0°ll' 0 . 3544 0 . 3090 0 . 2748 0 . 2488 o. 23?,l 0 . 2215 0 0.1503 0.1312 0 . 1118 0 . 0928 0 . 07:~8 0 . 0544 20 0 . 2298 0 .1830 0 .1468 0. 118? 0 . 0972 0 . 0791 40 0. 3260 0 . 2330 O.l 799 0 . 1450 0 .1216 0 . 1008 60 ... 0 . .2790 o. 2110 0 .1727 0.14111 0 .1185 80 - o. 2!385 0.1942 0 .1600 0 . 1325 ' 0 .2115 0 .1760 0 . 1469 0 . 2290 0 .1919 0 .1612 0 . 2439 o. r),()55 0 . 1?35 0 . 2160 0 .1 830 0 . 2242 0 .1925 Pressurer Lb ~/ Sqr In ~ Gas Satd. f.l 0 . 2635 100 125 150 175 200 250 - - - 0 . 2076 Symbols Cp Spe<?ific Heat a t constant -pressure C8 Specific Reat at s a tura.tion conditions Cv Speeifie Heat a t constant 't"olume f • Fugaeity H - Heat Content J Mechanical equivalent of heat M Molecular wei ght P Pressure ,, Q, Heat absorbed ;a Ga.s constant r Beat of vaporiza tion S Entropy ! Temperature Joul e Thoms on coefficient Aeknowledgeme:nts Thanks are due fo-r the a ssistanee of J. E. Sherborne, H. S. l3aekns. and E. R. Kennedy 1n carrying out the experimental work of the butane study. D. C. -Webster assisted in the drafting. The work was made ;poss~ble . through the financial assistance of the .American Petroleum Institute. The advice and criticism of Dre. W. N. ta.eey and '.B. H. Sage. as well as their patient and untiring efforts while leadil\g the author thro1:1gh the theoretical background and experimental di.:fficulties., was invaluable,.. Literature Cited (1) Sage, B. H., Backus, H. S. • and Vermeulen, 'f •• Ind. Eng. Chem•• ~- 489 (1936) (2) Kennedy, E. R.• C. L T. Thesis, 1936 also to be published in June, 1936, Ind. Eng. Chem.,· £!!, 0000 (3) Sage, 13. H., Schaafsma, J. G., and Lacey. W. N., Ind. Eng. Chem.• &i• 1218 (1934) ( 4) Sage, :B . H., and Lacey, W. }J., Ind. 1Jng. Chem., l!• 1484, ( 1935) (5) Sage,:B. H., Kenned7 1 m. :A. ... e,nd Lacey, W. N., Ind. Eng. Chem •• ~. 601 (1936) (6) Osborne, N. s .. and Van Dusen. M, S . ., J3ur. Stand,a.rds. :Bull. li• 39? ~1918)