Vapor-Liquid Equilibria for

Commercially Important Systems

of Organic Solvents
The Binary Systems Ethanol-n-Butanol, Acetone-Water and
Isopropanol-Water
AUSTIN S . BRUNJES AND MARCEL J. P. BOGART
The Lummus Company, New York, N. Y.

Vapor-liquid equilibrium data are pre- systems which comprise the quaternary system n-butanol-
sented for the three binary systems acetone-ethanol-water, all of the equilibrium data have been
published in the literature except those for mixtures of
ethanol-n-butanol, acetone-water, and iso- ethanol and n-butanol. While it might be expected that
propanol-water. Activity coefficients for mixtures of the two monohydric alcohols would closely follow
the acetone-water and isopropanol-water the ideal solution laws, it was considered desirable actually
systems are also given. These data, to- to determine the experimental data for comparison with
gether with observations on the operation theoretically calculated values.
I n the production of high-purity acetone from aqueous
of commercial distillation equipment, show mixtures by distillation, it has been found that fractionating
that the original published values for these column performance is at variance with that predicted from
two aqueous systems were not sufficiently existing vapor-liquid equilibrium data. This was also noted
accurate for engineering purposes. by Luaces ( I $ ) , who cited large discrepancies in equipment
sizes for acetone-water distillation units by various designers,
although he was not able to ascribe these differences to their
true source. Hence the original data of Bergstrom (6)were

D ATA are available on the vapor-liquid equilibria for

most of the binary syktems involved in the distillation
of the solvents obtained from the acetone-butanol
fermentation of molasses or grain. Of the six possible binary
carefully redetermined, the region of the high acetone con-
centration receiving particular attention. Data on this sys-
tem were also recently determined by York and Holmes ($3).

120- , 1 I I I 1

0 20 40 60 80 too 0 20 40 60 80 100

FIGURE1. REFRACTIVE
INDEXOF MIXTURESOF ETHANOL FIQURE2. TEMPERATURE-COMPOSITION
DIAQRAM
FOR SYS-
AND %-BUTANOL AT 25" c. TEM ETHANOL-+BUTANOL
AT 1 ATMOSPHERE

255
256 INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 35, No. 2

TABLEI. VAPOR-LIQUID EQUILIBRIUM DATAFOR THF) SYSTEM

ETHANOL%-BUTANOL
Cor. Temp. a t
R ~ , ~ 760 Mm. H g Ethanol in Liquid Ethanol in Vapor
No. C. ' F. Wt. r0 Mole Yo Wt. r0 Mole
1 79.1 174.4 90.3 93.8 99.6 99.7
13 80.3 176.5 83.0 88.7 98.0 98.7
2 81.0 177.8 77.2 84.5 96.9 98.2
3 83.3 181.0 63.8 73.8 96.0 97.6
14 83.9 183.0 64.3 74.3 91.4 94.6
16 85.7 186.1 55.5 66.6 88.6 92.7
4 86.4 187.5 48.4 60.1. 84.9 90.0
15 88.3 190.9 45.7 57.4 82.7 88.5
5 88.1 190.5 43.7 55.5 82.1 88.1
6 90.6 195.0 36.9 48.8 75.0 82.8
7 93.5 200.3 31.7 42.7 69.3 78.4
8 96.4 205.6 22.9 32.3 60.3 70.8
9 101.3 215.3 17.3 25.2 48.1 59.7
10 105.9 222.5 12.0 17.8 38.0 49.6
11 111.1 232.0 7.3 11.3 22.3 31.6
12 114.3 237.9 4.0 6.2 12.2 18.1

pressure of 759.4 mm. of mercury. Density and refractive

index determinatioiis of two samples gave the following results:
GZ 0.80816 0.80815
0 20 40 60 80 100 diS 0.80567 0.80567
ng I . 3970 1,3970
3. VAPOR-LIQUID
FIGURE EQUILIBRIA
FOR SYSTEM
ETHANOL-
?+BUTANOL AT 1 ATMOSPHERE
These values are in good agreement with those of Brunell,
Crenshaw, and Tobin (4); Clarke, Robinson, and Smith (Y).
ETHANOL.The raw material used was 190-proof ethanol.
The original vapor-liquid equilibrium data This mas dried by refluxing for 4 hours each with two batches
on the system isopropanol-water were de-
termined by Lebo ( I O ) who used simple
batch distillation with an averaged liquid
c o m p o s i t i o n . Equipment designed from
these data again did not operate a t the
reflux ratio and the overhead composition
predicted from Lebo's data. It was thus
decided to redetermine these data accord-
ing to more recent technique. Miller and
Bliss ( I S ) determined a few points for this
system while studying isopropanol-isopropyl
ether mixtures, Schuniacher and Hunt (18)
also recently determined points for this
system.
Purification of -Miaterials
n - h ~ a x o ~The
. c. r. grade was used in
these determinations. The A. S.T. hI. dis-
tillation ( I ) of the unpurifiecl sample
follows:
First drop C. 238' F. 114.4'
10% over 239 116 0
20 % 240 115 5
30%. 241 116 1
40 % 241.8 116.5
5OV" 242 116.6
::g
End point
242
242
242
116.6
116 G
116 6
Recovery, 97F7,; corrected pressure, 7 8 0 . 4 mm. I l g .

The n-butanol was purified by fractioimtion

in a glass column of 2.5-cn1. diameter, hav-
ing a 73.7-em. packing of jack chain. The
portion distilling belov 116.7" C. was dis-
carded. The rest of the sample boiled
within 0.5" C. The sample was then shaken
17-ith Drierite and allowed to stand in con-
tact with the reagent with frequent shak-
ing for 3 days. It was refractionated through
the column equipped wit'h a calibrated
Anschutz thermometer. The entire fraction FIGURE DIAGRAM
4. TLMPXRATURE-COMPOSITION FOR SYSTElf A4CET0.UI!.-
distilled off a t 117.8" C. under a corrected WATER AT 1 A4TR10SPHCRE
February, 1943 INDUSTRIAL AND ENGINEERING CHEMISTRY 257

Experimental Procedure
The well established method of Othmer (14,
16) was used for the determination of the
equilibrium data in each of the systems
studied. The original apparatus was modi-
fied to permit the use of external gas heating,
and an auxiliary vent condenser was perma-
nently sealed onto the vent. When'running
the ethanol-butanol system, a large drying
tube filled with soda lime and anhydrous
calcium chloride was attached to the vent
line beyond the condenser. When running
the acetone-water system, an additional long
jacketed condenser was attached directly to
the end of the small vent condenser on the still.
Refractive indices were determined on a
Bausch & Lomb Abbe-type refractometer
equipped with a microvernier and held a t
constant temperature by circulating water
from the thermostat. Sodium flame illumi-
nation was provided by a shielded Bunsen
burner and a mixture of sodium carbonate
and borax. A metal shield covered the top
and three sides of the instrument, and ex-
cluded stray white light from the room.
Specific gravities were determined b y 25-
ml. pycnometers equipped with thermometers
and caps having no vent holes a t the extremi-
FIGURE5. VAPOR-LIQUID
EQUILIBRIA
FOR SYSTEMACETONE-WATER
AT
ties. They were immersed in the constant-
1 ATMOSPHERE temperature bath for a sufficient time before
each weighing to come to equilibrium tem-
perature.
of quicklime, according to Fisher (9). The alcohol was The System Ethanol-n-Butanol
distilled off from the lime into a previously dried receiver and A series of mixtures of the two alcohols were weighed out
was found to contain 0.37 per cent water by weight. Severe on the basis of an approximately 10-gram total sample and
humidity conditions prevailing a t the time prevented further
dehydration. This alcohol was considered satisfactory for
the purpose of this experiment and was stored in bottles
equipped with tin-foil covered stoppers. The density and
refractive index of the alcohol are tabulated below:

dBE 0.78863
di6 0.78623
ny 1.3589

The density conversions were obtained from the Bureau of

Standards alcohol tables (.E?).
ACETONE.Reagent-grade acetone was repeatedly agitated
with anhydrous calcium chloride, and the mixture allowed to
stand overnight. The acetone was decanted off and distilled
through a column 2.5 cm. in diameter and 137 cm. long,
packed with glass Raschig rings, under a reflux ratio of 2
to 1; the first portion distilling over was discarded. The
remainder of the sample distilled a t a constant temperature
of 56.6" C. and had a specific gravity (df:) of 0.78623. This
value is in agreement with the experimental value of Perkin
(16) and the computed value from the equation of Timmer-
mans (21) and Squibb (19).
ISOPROPANOL. The c. P. grade as received had a boiling
point of 81.72' C. at 760 mm. of mercury, and a specific
gravity (d::) of 0.78888, corresponding to 98.86 weight per
cent alcohol. This checked the analysis furnished with the
sample and, since the remainder was reported as water, this
sample was not further purified before use. The gravities
were determined a t 20/15O C. in order to use the tables
furnished by the Standard Alcohol Company (20). 92 94 96 98 IO0
WATER. The water used in the equilibrium and analytical FIGURE6. MAGNIFIEDSECTION
OF VAPOR-LIQUID
EQUILIBRIA
runs was double-distilled and freshly boiled. FOR HIGHACETONE
CONCENTRATIONS
258 INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 35, No. 2

of the high volatility of the acetone. The

EQC~LIBRIUM
TABLE11. VAPOR-LIQUID DATAFOR THE ACETONE-
SYSTEM boiling rate in the still was such that at the
WATER high acetone concentrations the receiver filled
Cor. Temp. at Acetone in Acetone in Activity in 3 t o 4 minutes. At the low concentration
R ~ , ~ 760 Mm. H g Liquid Vapor Coefficient
No. C. E'. Wt. yo Mole yo Wt. yo Mole Acetone Water end, because of occasional bumping, the rate
19 56.5 133.7 99.50 98.40 99.65 98.44 0.996 4.212 of distillation was reduced so that the re-
20 56.6 133.9 99.20 97.47 99.35 97.93 0.992 4.900 ceiver would fill in 6 to 7 minutes. Runs
21 56.8 134.2 98.85 96.38 99.10 97.15 0.989 4.665
1 Determination lost were made from 40 to 75 minutes, depending
22 57.0 134.6 98.00 93.83 98.60 95.62 0.990 4.278
upon the concentration range.
10 57.7 135.9 97.50 92.37 98.30 94.72 0,973 4.034
2 57.6 135.7 96.70 90.04 97.90 93.53 0.988 3.777 At the conclusion of each run, after the
3 57.7 135.9 95.40 86.55 97.40 92.08 1.011 3.349 vapor temperature and the barometer were
4 58.1 136.7 93.65 82.08 96.80 90.37 1.031 2.998
5 58.35 136.9 91.20 76.29 96.20 88.71 1.083 2.632 read, the still was shut off and allowed t o
6 58.7- 137.8 89.50 72.87 95.90 87.90 1.110 2.393 cool.' Liquid and condensed vapor samples
7 58.8" 137.9 87.40 68.29 98.05 87.23 1.168 2.174
8 59.4 138.9 84.20 62.32 95.20 86.03 1.237 1.946 were then taken in bottles surrounded by ice
9
11
59.8
59.8
139.8
139.8
78.60
77.00
52.28
50.98
94.86
94.90 85.12
85.24 1.434
1.477 11.551
.607 baths. Pycnometers were filled and placed
12 60.3 140.5 72.00 44.39 94.50 84.21 1.653 1.429 in the constant-temperature bath. When they
13 60.9 141.6 08.20 30.85 94.00 83.09 2.297 1.197 reached equilibrium temperature, they were
14 62.7 144.9 40.20 17.27 93.10 80.73 3.744 1,050
15 65.3 149.4 29.75 11.64 91.85 77.77 4.930 1.008 capped, carefully wiped dry, and rapidly
,

16 72.1 161.8 18.00 6.38 87.30 68.09 6.362 1.014 weighed on a Chainomatic balance. The
17
18
85.2
92.0
185.3.
197.6
6.80
3.60
2.21
1.15
72.40
55.50
44.88
27.91
8.253
8.216
0.981
0,978
weights were left on the balance, and the
pycnometers again returned to the bath and
refilled with sample. The second weighings
were accomplished rapidly by merely adjust-
the refractive indices of the mixtures were determined at ing the chain. All determinations in the high acetone
25" C. When a sufficient number of mixtures had been made region were checked three times in this manner to make
to define the refractive index-composition curve, the points sure that there n-as no loss of weight due to evaporation.
were plotted and a smooth curve drawn through the data. At lower .concentrations of acetone, two weighings were
Figure 1 is a plot of the following data: found sufficient for a check. A second series of runs on the
high acetone concentrations were made to ensure a sufficient
Weight % number of points to define clearly the equilibrium curve in
Ethanol%
Weight ny Weight
Ethanol70 ' a
;. Ethal,ol
nY the region where previous data have been questionable.
0.00 1.3970 23.19 1.3875 62.45 1.3731 The data for the system are reported in Table 11. The
5.96
9.60
10.14
1.3930
1.3940
1.3929
35.06
29.57
43.06
1.3830
1.3853
1.3800
!!:!:
88.78
::;::$
1.3626
boiling point-composition curves are plotted in Figure 4
together with the data of Carveth (6),Pettit ( I ? ) , and the
15.51 1.3909 51.91 1,3770 100.00 1.3689
recent data of York and Holmes ($3) for comparison. Figure
5 is a plot of the equilibrium data showing a comparison with
Four hundred milliliters of a mixture of ethanol and buta- the data of Bergstrom ( 2 ) and those of York and Holmes.
no1 were introduced into the equilibrium still; the rate of Figure 6 is an enlarged section of the upper portion of the
boiling was regulated so that the contents of the receiver equilibrium diagram, the dotted line showing the previously
would change in 3 t o 4 minutes, and the runs were carried on accepted curve of Rergstrom and the solid line, the data of
from 45 to 60 minutes before samples were taken. Samples this paper. This difference in the position of the curve
of high ethanol concentration reached equilibrium, judged by accounts for the discrepancy between the older data and
constancy of vapor temperature, more quickly
than samples of high butanol concentration.
Samples of liquid and condensed vapor mere
withdrawn simultaneously in small bottles. The TABLE 111. VAPOR-LIQEID EQUILIBRIUM DATAFOR THE SYSTEM
ISOPHOPASOL-TYATER
bottle on the liquid side was immersed in a Cor. Temp. a t Isopropanol Isopropanol Activity
beaker of crushed ice. R~~ 760 Mm. Hg in Liquid in Vapor Coefficient
The data obtained from the equilibrium runs Xo. C. O E'. Wt. 9% Mole % Wt. Mole % Alcohol Water
are plotted in Figure 2 and listed in Table I. 1
2
81.21
81.01
178.2
177.8
97.85
97.29
93.19
91.53
96.82
96.07
90.11
88.01
1.012
1.014
2.963
2.912
The x-y values are shown in Figure 3 together 3 80.75 177.4 96.33 88.72 94.97 85.00 1.022 2.741
4 80.51 176.9 95.05 85.20 93.54 81.26 1.027 2.657
with the values computed from Raoult's ideal 5 80.30 176.5 93.42 81.00 91.77 76.98 1.032 2.564
law. These theoretical values for vapor-liquid 6 80.14 176.3 91.77 77.02 90.47 74.01 1.050 2.409
equilibrium at 760 mm. of mercury follow: 7
8
80.07
80.05
176.1
176.1
90.15
88.48
73.33
69.71
89.28
88.27
71.42
69.31
1.067
1.123
2.289
2.166
9 80.03 176.0 88.15 69.05 88.02 68.79 1.093 2.157
27 80.04 176.0 87.90 68.57 87.74 68.24 1.091 2.163
Mole % Ethanol 80.08 176.1 87.82 68.38 87.86 68.46 1.098 2.132
10
Liquid Vapor Liquid Vapor 28 80.16 176.3 87.70 68.13 87.70 68.13 1.093 2.134
94.7 98.8 25.2 59.6 11 80.04 176.1 86.08 $4.96 86.91 66.69 1.122 2.041
71.2 92.2 15.1 43.0 12 80.14 176.3 82.39 08.38 85.34 63.58 1.190 1.864
62.8 83.9 6.7 22.7 13 80.44 176.7 73.95 48.97 82.98 69.39 1.395 1.582
37.9 73.3 2.84 10.5 14 80.77 177.4 82.30 33.14 81.27 56.54 1.820 1.350
15 Determination lost
26 81.11 178.0 51.15 23.87 80.67 55.59 2.448 1.195
25 81.19 178.2 45.24 19.86 79.93 54.44 2.873 1,160
The System Acetone-Water 24 81.39 178.4 39.36 16.29 78.99 52.98 3.380 1.137
22 81.41 178.5 31.90 12.32 79.50 53.78 4.531 1.067
A curve of specific gravities was computed 21 82.63 180.9 23.50 8.43 77.10 50.24 5.873 1.049
and drawn on a large-scale plot according to :! 8";: 1g:E7 :$1: i$:fE $3":; 8.926 1.033
1,018
the equation of Squibb (19) for analytical pur- 19 90.80 195.4 6.50 2.04 57.96 23.08 8.000 1.102
23 93.19 199.7 4.38 1.36 49.00 22.44 10.54 1.008
poses. The technique o f making the equilib- 17 95.30 203, 5 2.72 0.83 36.55 14.73 10.54 1.02"
rium runs and handling the samples was 16 98.87 210.0 0.52 0.16 11.16 3.64 11.68 1.007
modified from the standard procedure because
February, 1943 IN D U S T R I A L A N D E N G I N E E R I N G4-C H E M I S T R Y 259

rium still. The temperature remained con-

stant a t 80.18" C. under 760 mm. absolute
pressure. The equilibrium run gave a
liquid and vapor at the above tempera-
ture and pressure, both having a compo-
sition of 87.7 weight per cent alcohol.
Various values reported in the literature
are :

Boiling
Point, Wt. % Mole %
Author O C. Alcohol Alcohol
Lecat (if) 80.37 87.85 68.54
Young (94) 80.4 87.90 68.56
Lebo (10) 80.4 87.7 68.13
Miller and Bliss (13) 80.1 87.9 68.60
Schumaoher and H u n t (18) 80.3 87.4 67.8
Authors 80.16 87.7 68.13

The temperature-composition plot of the

data are shown in Figure 7 and compared
with the data of Doroshewski and Polian-
ski (8)and Lebo. The vapor-liquid equilib-
rium curve is plotted in Figure 8. The
points of Miller and Bliss, Lebo, and Schu-
macher and Hunt were added for com-
parison.

Activity Coefficients
The activity coefficients, y, were com-
puted from the equation

7. TEMPERATURE-COMPOSITION
FIGURE DIAQRAMFOR SYSTEM
ISOPROPANOL-
WATERAT 1 ATMOSPHERE y = PyIPox

actual distillation column performance a t IO(

high acetone product compositions.

The S y s t e m Isopropanol-Water
Equilibrium runs for this system were 8C
made in the same manner as for the
ethanol-butanol system, starting with
alcohol-rich mixtures. Gravities of the
vapor and liquid samples were measured
in pycnometers a t 20/15O C . , and the per-
centage composition interpolated from the 6C
isopropyl alcohol-water gravity tables pub-
lished by the Standard Alcohol Com-
pany (20). Small pieces of copper wire
were introduced into the still during the
runs t o destroy any peroxides that might 40
be formed from the presence of minute
amounts of isopropyl ether. No tarnish-
ing effects on the wires were noted during
the entire series of runs. The data are
tabulated in Table 111.
20
* MILLER
SCHUMACHER
4 BLISS
t
Still residues were fractionated through
a 100-plate Bragg column (3) at nearly
total reflux. The distillate was collected I I I I I I I I I
I I I I I 1 I
a t a constant temperature of 80.3" C.
under 760 mm. absolute pressure. This I MOL
1
% 150PROPANOL IN LIQUID
I I I 1 1 I
1
fraction was again run through the column C 20 40 60 80 It 0
at nearly total reflux, and the distillate
collected until a sufficient sample was 8. VAPOR-LIQUID
FIGURE EQUILIBRIA
FOR SYSTEM
ISOPROPANOL-WATER
AT
obtained to make a run in the equilib- 1 ATMOSPHERE
260 INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 35, No. 2

IO showed only a slight deviation from the theoretical equilib-

9 rium, it was decided not to compute the activity coefficients.
6 The activity coefficients for the acetone-water system are
I plotted against the mole fraction of acetone in the liquid in
Figure 9. The activity coefficients for the system isopro-
0 panol-water are plotted against mole fraction of alcohol in
the liquid in Figure 10.
Conclusions
Carlson and Colburn (6),in their recent article on the
application of theoretical methods to vapor-liquid equilibria
in nonideal solutions, stated that the activity coefficient
provides a sensitive method for checking the accuracy of
vapor-liquid equilibria. The plots in Figures 9 and 10 for
the acetone-water and isopropanol-water systems show that
the activity coefficients lie on a well defined line except for one
point on each system which is a marked deviation.
The boiling point-composition and vapor-liquid equilibrium
curves also show that the data are consistent within them-
selves. The lowered enrichment for the more volatile com-
ponent in the acetone-water and isopropanol-water systems
relative to the original data may be considered a satisfactory
explanation of the subspecification quality of the overhead
0 20 40 60 80 100 product obtained from continuous stills designed according
to the older data. Recently designed continuous stills have
[ FIQURE 9. ACTIVITYCOEFFICIEXTSFOR BINARYSYSTEM been operating in good agreement with the data presented
ACETONE-WATERAT 1 ATMOSPHERE
in this paper.
The vapor-liquid equilibrium data determined for the sys-
where P = total pressure on system tem ethanol-n-butanol show that this system closely follows
PO = vapor pressure of given component the ideal solution laws. The equilibrium data determined
y = mole fraction of component in vapor for the systems acetone-water and isopropanol-water are in
x = mole fraction of component in liquid good agreement with those of recent investigators and in-
for the acetone-water and the isopropanol-water systems, dicate that the original data on these two systems did not
and are listed in Tables I1 and 111, respectively. The vapor represent the true values.
pressures for the pure components were taken from Inter- Acknowledgment
national Critical Tables. Since the ethanol-butanol system
The authors wish to acknowledge the assistance of Beatrice
G. Brunjes in the preparation and correction of the manu-
IS script.
Literature Cited
IC (1) Am. SOC.Testing Materials, Standard D86-38 (1939).
(2) Bergstrom, H., “Aftryk ur Bitrang til Jem-Kontoret’s Annaler”,
9
Stockholm, 1912.
3 (3) Bragg, L. B., Trans. Am. Inst. Chem. Engrs., 37, 19 (1941).
7 (4) Brunell, Crenshaw, and Tobin, J . Am. Chem. Soc., 43, 561
(1921).
6 (5) Carlson, H . C,,and Colburn, A. P., IND.ENG.CHEM..34, 581
(1942).
5 (6) Carveth, H. R., J . Phys. Chem., 3, 193 (1899).
(7) Clarke, Robinson, and Smith, J . Chem. SOC.,129, 2647 (1927).
(8) Doroshewski and Polianski, J . Buss. Phys.-Chem. Soc., 42, 1448
4
(19 10).
(9) Fisher, H. L., “Laboratory Manual of Organic Chemistrq”, 2nd
ed., p. 33, New York, John Wiley & Sons, 1925.
3 (10) Lebo, R , B., J . Am. Chem. SOC., 43, 1005 (1921).
(11) Lecat, “La tension de vapeur des mblanges de liquides”, Brus-
sels, Henri Lamertin, 1918.
(12) Luaces, E. L., Trans. Am. Inst. Chem. Engrs., 36, 171 (1940).
2 (13) Miller, H. C..and Bliss, R. H., private communication.
(14) Othmer, D. F., IND. ENG.CHEM.,20, 734 (1928).
(15) Othmer, D. F., IND. ENQ.CHEM.,ANAL.ED., 1 , 4 6 (1929).
(16) Perkin, TV. H., Trans. Chem. SOC.,45,478 (1884).
(17) Pettit, J. H., J. Phys. Chem., 3, 349 (1899).
(18) Schumacher, J. F., and Hunt, H., IND.ENG. CHEM.,34, 701
(1942).
(19) Squibb, E. R., J . Am. Chem. Soc., 17, 187 (1895).
1.0 (20) Standard Alcohol Co., “Petrohol”, p. 56, Table 1 (1939).
MOL ye ISOPROPANOL IN LIQUID (21) Timmermans, J., Sci. Proc. Roy. Dublin SOC.,13, 310 (1912);
I I I I I I International Critical Tables, Vol. 111, p. 28 (1928).
20 40 60 80 100 (22) E. S.Bur. Standards, Circ. 19, 6th ed. (1924).
(23) York, R., Jr., and Holmes, R. C., IND.ENQ. CHEM.,34, 348
10. ACTIVITYCOEFFICIENTS
FIGURE. FOR BINARYSYSTEM (1942).
ISOPROPANOL-WATER AT 1 ATMOSPHERE (24) Young, S.,J . Chem. Soc., 81, 728 (1902).