2450
INDUSTRIAL AND ENGINEERING CHEMISTRY
Vol. 44, No. 10
TABLE 11. EQUILIBRIUM DISTRIBUTION OF ACETONEBETWEEN CARBON TETRACHLORIDE AND AQUEOUSPHASES Acetone-CC14 Phase, Wt. %
Acetone-Water Phase, Wt. %
30 30 30 30 30
0.80 3.95 4.60 13.90 16.50 24.20
3.35 10, a0 11.50 22.30 24.40 30,40
30 30 30 30 30
33,30 39.60 47.20 51.90 59.04
35 80 40.20 46.00 49 80 69.04 Plait point
20.8 20.8 20.8
2.10 6.20 13.80
6.90 14.30 22.50
2.21 4.90 9.00 13.80
9.50 14.90 20.60 25.30
Temp., 30
C.
4 4 4 4
WEIGHT PER CFYTACETONE IN AQUEOUS PHASE Figure 3. Distribution of Acetone between Carbon Tetrachloride and Water with Temperature
In the tie line determinations, the three components were weighed carefully into the same constant-temperature jacketed cylinder, the relative amounts being chosen t o cause the interface to be near the middle. After 5 minutes of vigorous shaking, the phases were settled and the Rpecific gravity of both determined in situ with a Westphal balance. Near the plait point, where the specific gravities of the two phases are almost identical, 3 or 4 hours were required for settling. The specific gravities of the two equilibrium phases were converted to tie line concentrations with Figure 1. The accuracy of the data is verified when the over-all, and the two equilibrium compositions on the triangular graph, fall on the same straight line. A few tie lines are shown in Figure 2. The distribution of acetone between the two
I
phases is shown in Figure 3 at varying temperatures The equilibrium data are presented in Table 11. This system is solritropic ( 1 ) since the distribution favors the aqueous phase a t low acetone concentrations, the reverse being true a t high concentrations. In Figure 3, the change in distribution between 20' and 30" c. is small and is probably negligible in comparison with the experimental error. When t h r direction of mass transfer is unimportant as in many liquidliquid extraction studies, simple operation can be achieved by extracting acetone from the carbon tetrachloride with water, recycling the raffinate, and discharging the extract t o the sewer. LITERATURE CITED
(1) Smith, A. S., IND.ENG.CHEM.,42, 1206 (1950). RECEIVED for review February 11, 1952.
ACCEPTED May 2, 1952.
Vapor- iquid Equilibria at 760 Mm. Pressure 2-PROPANOL-METHANOL, 2-PROPANOLETHYL ALCOHOL, 2-PROPANOLPROPANOL, AND 2-PROPANOL-2-BUTYL ALCOHOL SYSTEMS LOUIS H. BALLARD' AND M. VAN WINKLE University of Texas, Austin, Tex.
A
DVENT of the various processes for oxidation of hydrocarbons has posed numerous problems of separation of the components present in the process products. Fractional distillation is one of the common separation procedures and design Qf fractional distillation equipment requires a knowledge of vapor-liquid equilibrium data of the systems of multicomponent mixtures produced in the oxidation processes. The data presented here are binary vapor-liquid equilibrium data on Z-propanol (isopropanol)-alcohol binaries and are the result of one of a series of investigations concerned with the experimental study of vapor-liquid equilibria of the oxidized hydrocarbons. The materials used for vapor-liquid equilibrium determinations must be of high purity. The methanol and ethyl alcohol Present address, Monsanto Chemical Co., Texas City, Tex.
were obtained pure, but the 2-propanol, propanol, and isobutyl alcohol (isobutanol) required further purification before use. They were fractionally distilled in a &foot, 0.75-inch diameter, glass distillation column packed with glass helices. I n each case, both ends of the original charge were discarded and the middle fraction was retained until the desired purity was obtained. Purity was determined by boiling point and refractive index. The refractive index was determined a t 20' C. with a Bausch and Lomb refractometer. The boiling points were determined in a Cottrell boiling point apparatus. The total pressure was maintained a t 760 mm. of mercury by the use of air pressure. Comparison of the experimental refractive indexes and boiling points on the materials used in this investigation with values reported in the literature on pure compounds is presented in Table I.
October 1952
INDUSTRIAL AND ENGINEERING CHEMISTRY
'0
MOLE PERCENT METHANOL
30 40 50 60 70 00 MOLE PERCENT ETHANOL IN LIQUID, X
20
IO
IN LIQUID, X
Figure 1. Vapor-Liquid Equilibria Diagram for the Methanol-2-Propanol System at 760 Mm.
2451
90
IOC
Figure 2. Vapor-Liquid Equilibria Diagram for the Ethyl Alcohol-2-Propanol System at 760 Mm. in the liquid chamber of the still and filled with a high-boiling mineral oil. T h e thermocouple was calibrated in place using the boiling points of pure methanol, ethyl alcohol, benzene, toluene, and ethylbenzene. Temperatures were read to the nearest 0.1O F. The refractive indexes were det,ermined with a Bausch and
TABLEI. COMPARISON O F LITERATUREA N D E X P E R I J I E N T A L VALUESOF BOILING POINTS AND REFRACTIVE INDEXES Boiling Point, 0 F. Lit. Expt. , . . , . . 1 4 8 . 4 ( 8 ) 148.4 148.1 6 ) Ethanol U. S. InduPtrial 173.3 [ 8 ) 173.3 Chemicals 172.4 ( 6 ) Propanol Eastman Kodak Co. 206.9 (8) 207.4 205.7 6 ) 2-Propanol Eastman Kodak Co. 180.1 i 8 ) 180.2 180.1 ( 6 ) Isobutyl alcohol Eastman Kodak Co. 227.1 ( 8 ) 226.5
Compound Methanol
Source
~
Refractive Index at 200 Lit. Expt. 1.32880(4) 1.32880
c.
::%9"
{;[
1.36120
1.38543 8 ) 1.38576 1 3855616) 1: 37757 ( 8 ) 1.37716 1.37711 ( 6 ) 1.39549 (6) 1.39614
~~
~~
TABLE11. EXPERIMENTAL VAPOR-LIQUID EQUILIBRIUM DATA FOR 2-PROPANOL-METHANOL SYSTEM AT 760 M M . M E R C U R Y Figure 3. Vapor-Liquid Equilibria Diagrams for the Isobutyl Alcohol-%-Propanol and the n-Propanol-2-Propanol Systems at 760 Mm.
APPARATUS
A modified Colburn still (3) was used for the vapor-liquid equilibrium determinations. This apparatus is the same as that used in a n earlier investigation (1). Pressure was maintained a t 760 mm. of mercury by air pressure applied to the system through the isolation condenser. Air from the main line was passed through a filter and a calcium chloride drying bottle before entering the system. The filtered and dried air was connected by a tee to an air bleed, a manometer, and the isolation condenser. The air bleed allowed the proper pressure to be maintained on the manometer and the system a t all times. The manometer fluid was n-tetradecane. A Cottrell boiling point apparatus was used in the determination of the purity of the chemicals used in the investigation and in the analysis of the 2-propanol-propanol system. Equilibrium still temperatures were measured by an ironconstantan thermocouple. The thermocouple well was enclosed
Run No. 1 2 3 4 5 0 7 8
Temp,, O F 151.2 154.3 158.4 162.8 166.6 170.7 174.1 177.8
-
Mole yo Methanol Vapor Liquid 95.35 90.10 89.10 79.00 80.00 66.05 68.50 52.20 40.80 57.00 42.85 29.30 29.60 19.50 13.20 8.10
Activity Experimental Coefficients 2-Propanol Methanol 0.96 1.00 0.98 0.99 1.00 0.98 1.00 0.97 1.01 0.95 1.02 0.92 1.02 0.89 1.01 0.89 ~
~
~
~
~~~~
TABLE111. EXPERIMENTAL VAPOR-LIQUID EQUILIBRIUM DATA FOR ~-PROPANOL-ETHYL ALCOHOL SYSTEM AT 760 MM. MERCURY
Run No, 1 2 3 4 5 6 7 8 9 10 11 12 13
Temp.,
F.
Experimental Activity Coefficients Mole % E t h s n o ~ _ Ethyl Vapor Liquid 2-Propanol Alcohol
2452
INDUSTRIAL AND ENGINEERING CHEMISTRY
I
I7d
lb
20
?,
I
Vol. 44, No. 10
5L
i o o;
MOLE F€RGENT SO-PROPANOL
1i
Id0
MOLE PERCENT ETHANOL
1s-m MOLE PERCENT ISO-PROPANOL
Figure 5 . Equilibrium Boiling Point Diagrams for the Isobutyl Alcohol-2-Propanol and the n-Propanol-2-Propanol Systems at 760 Mm.
MOLE PERCENT METHANOL
Figure 4. Equilibrium Boiling Point Diagrams for the Methanol-2-Propanol and the Ethyl Aloohol-2-Propanol Systems a t 760 Mm.
Lomb precision refractometer using sodium D light. Temperatures were maintained a t 20" C. by the use of a constant temperature bath. PROCEDURE
A refractive index us. mole per cent 2-propanol curve was made for each system except the 2-propanol-propanol system. Various amounts of each component were weighted on a chainomatic balance, and thoroughly mixed, and the refractive index was taken. The results were plotted on large scale graph paper, allowing the composition t o be read to the nearest 0.05%. The refractive indexes of 2-propanol and propanol were too nearly equal to allow an accurate refractive index-composition curve to be drawn. Therefore, a boiling point-composition curve was constructed and used for the analysis of this system. The charge t o the equilibrium still was introduced through the isolation condenser. Tap water was sent through the primary condenser and ice water through the isolation condenser t o assure condensation of all vapors. The pressure was adjusted to 760 mm. of mercury and the main power switch was turned on. All heaters were turned on and the liquid chamber heater adjusted t o high heat. The vapor superheater was kept hot enough a t all times t o prevent any condensation of the vapors leaving the liquid chamber. The condensed vapor heater was kept on low heat. Only enough heat t o 13-arm the liquid %-asused, thus tending to decrease the load on the flash chamber. The flash chamber heater was kept on low heat until liquid in the liquid chamber began t o boil. The liquid chamber heater was kept on high heat until the liquid head in the condensed vapor chamber became high enough t o cause liquid flow through the three-way stopcock into the flash chamber. The heat input t o the flash chamber was increased until practically all the liquid entering it vaporized be-
fore entering the liquid chamber. The heat input to the liquid chamber was then decreased until only a few drops of liquid remained in the flash chamber liquid trap. Thus the vapors passing through the liquid trap were assumed t o be saturated as they entered the liquid chamber. The still was allowed to operate in this manner until the temperature became constant and for a t least 30 minutes thereafter. In no case was the total time of operation less than 1 hour. When it was assumed that equilibrium conditions had been reached, the main power switch was opened, the three-way stopcock was closed, isolating the vapor sample from the liquid sample as quickly as possible, and the flash chamber cooled with an air jet, causing vaporization to cease. Approximately half of both the liquid and vapor samples was discarded and the remainder of each sample was drained into cold, dry sampling bottles and immediately capped and chilled. The refractive index of each sample was determined with a Bausch and Lomb precision refractometer using sodium D light. The still was flushed with dry air and cooled before making another determination. RESULTS AND CONCLUSIONS
Vapor-liquid equilibrium data are presented for the binary systems 2-propanol-methanol, 2-propanol-ethyl alcohol, 2propanol-propanol, and 2-propanol-isobutyl alcohol. The experimental data are given in tabular form in Tables I1 through V and graphically in Figures 1 through 5. The equilibrium compositions are reported to the nearest 0.05 mole yo and the temperatures to the nearest 0.1 O F. The estimated limit of error in the experimental procedure is around 1%. Both the experimental vapor-liquid curves and the experimental temperaturecomposition curves are given. None of the systems formed azeotropes. With the exception of the 2-propanol-methanol system, the 2-propanol-alcohol systems appear to be very nearly ideal. The 2-propanolmethanol vapor-liquid curve tends to flatten out somerhat a t low concentrations of methanol. This tendency can also be seen on the temperature-composition diagram.
INDUSTRIAL A N D E N G I N E E R I N G C H E M I S T R Y
October 1952
2453
isothermal conditions, could not be considered to be strictly applicable to these data. The 2-propanol-ethyl alcohol system has a boiling range of 7“ C. and the experimental y - x plot of this system indicates it to be essentially ideal. (b) experimental data on heats of solution or variation of activity coefficient with temperature for these systems are not available. Thus suitable corrections for the effect of temperature cannot be applied. With the exception of a few experimental points the data plot as smooth curves on the temperature-composition plots and on the z - plots. Points not fulfilling this requirement were rerun. This tends to eliminate random errors in the experimental determinations. Systematic errors such as those caused by a constant presaure deviation were considered to be at a minimum because occasional random check determinations reproduced points on the curves within the estimated limit of error of the experimental procedure
TABLE V. EXPERIMENTAL VAPOR-LIQUID EQUILIBRIUM DATA FOR
2-PROPANOL-ISOBUTYL ALCOHOL SYSTEM AT 760 MM. MERCURY Mole % 2-Propanol
MOLE PERCENT-I
IN LIQUD. X
Figure 6. Activity Coefficient Diagrams for the 2-Propanol-Alcohol Systems a t 760 M m . The experimental activity coefficients were calculated for each system and are plotted against the mole per cent 2-propanol. These data are presented in tabular form in Tables I1 through V and graphically in Figure 6. The activity coefficients were calculated using vapor pressure d a t a taken from the literature (6). The calculated activity coefficients for the 2-propanol-ethyl alcohol and 2-propanol-isobutyl alcohol systems were near a value of one throughout the composition range, indicating an approach to ideality. The activity coefficients for the 2-propanol-methanol system deviated from a value of one with the methanol y x deviations greater than those of the 2-propanol. The activity coefficient-composition curves for the 2-propanol-n-propanol system indicate the system deviates from ideal solution behavior.
-
VAPOR-LIQUID EQUILIBRIUM DATA TABLE IV. EXPERIMENTAL FOR 2-PROPANOL-PROPANOL SYSTEM AT 760 MM. MERCURY Run No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Temp., ’F. 201.6 199.1 196.6 191.3 188.6 185.5 183.4 205.0 194.0 182.2 186.5 193.4 204.7 196.6
Mole % 2-Propanol Vapor Liquid 23.25 14.55 22.85 35.10 44.35 30.95 51.90 66.00 63.10 74.80 84.95 76.75 91.75 85.80 11.00 5.75 42.00 55.45 91.00 95.25 73.05 82.25 43.55 57.25 6.10 11.10 31.25 45.00
Experimental Activity Coeffioients 2-Propanol Propanol 1.03 1.03 1.02 1.03 1.03 1.02 1.08 1.01 1.05 1.00 1.07 1.00 1.01 1.02 1.17 1.03 1.05 0.99 1.01 0.94 1.06 0.99 1.04 0.99 1.12 1.04 1.03 1.02
The experimental vapor-liquid equilibrium data are presented, rather than smoothed data derived by application of the thermodynamic consistency equations. The reasons for this are as follows: (a) the boiling range of three of the four 2-propanolalcohol systems investigated ranges from 27” to 46’ F. Thus, a smoothing method based upon equations derived, assuming
Run No. 1 2 3 4 5 6 7 8 9 10 11 12
Temp., F. 223.2 218.2 211.8 204.4 200.7 195.7 191.6 188.5 185.7 181,7 210.9 202.9
Vapor 11.20 25.10 42.70 58.45 67.70 75.80 82.45 88.75 92.95 98.05 44.10 62.90
Liquid 4.65 11.55 21.85 34.55 44.10 54.55 63.80 74.50 82.75 94.85 23.05 38.70
Experimental Activity Coefficients Isobutyl 2-Propanol Alcohol 1.02 1.01 1.02 1.02 1.00 1.01 1.02 1.00 1.00 1.01 1.02 1.01
1.00 1.01 1.00 1.02 1.00 1.03 1.04
1.01 1.00
1.02 1.01 1.00
LITERATURE CITED (1)
Hill, W. D., and Van Winkle, M., IND.ENG.CAEM.,44, 205 (1952).
(2) Hodgman, J. H., Editor, “Handbook of Chemistry and Physics,” 31st ed., pp. 610-1129, Cleveland, Ohio, Chemical Rubber Publishing Co., 1949. (3) Jones, C. A., Schoenborn, E. M., and Colburn, A. P., IND. ENG. CHEM.,35, 666-72 (1943). (4) Lange, N. A., “Handbook of Chemistry,” 5th ed., pp. 314-653, 934-1026, Sandusky, Ohio, Handbook Publishers, Ino., 1944. (5) Stull, D. R., IND. ENG.CHEW,39,517-40 (1947). (6) Vogel, A. I., J . Chem. Soc., 1814-9 (1948). .4CCEPTED March 12, 1952. RECEIVED for review October 2, 1951. From the thesis by Louis H. Ballard submitted in partial fulfillment of t h e requirements for the degree of Master of Science in Chemical Engineering.