Document Sample
					                          CRITICAL PHENOMENA IN THE SYSTEM
                           ACETIC ACID - CHLOROFORM - WATER

                                A. N. CAMPBELL E. M. KBRTZMARK
                     Department of Chemistry, University of Manitoba, Winnipeg, A!lanitoba
                                           Received November 30. 1962

          This paper is an experimental study of ( a ) the solid-liquid equilibria a t low temperatures;
        (b) equilibrium compositions a t temperatures other than 25" C ; (c) the critical phenomena,
        L1-L2, L1-V, and L2-V, in the acetic acid - chloroform -water system. No previous study
        has been made of critical phenomena in a system exhibiting partial liquid miscibility.

   The system acetic acid - chloroform - water is one of the classical examples of partial
miscibility in a ternary system, first studied by Wright, Thompson, and Leon in 1891
(1). Thanks t o the subsequent workers Brancker, Hunter, and Nash ( 2 ) , the equilibrium
relations a t 25" are very well known, as far as the compositions of equilibrium liquid
layers are concerned. Nothing, however, is known about:
   (a) the solid-liquid equilibria a t low temperatures;
   (b) equilibrium liquid compositions a t temperatures other than 25";
   (c) the critical phenomena, L1-V (L1 = chloroform layer) and L2-V (Lz = aqueous
  So far no study has been made of critical phenomena in any system exhibiting partial
miscibility in the liquid state and the results arrived a t in this paper have general appli-
cability to all systems of this kind.
  A previous paper (3) has dealt with the thermodynamics of this system, as exemplified
by the vapor pressures and densities. Further work is proceeding on the isobaric boiling
points, dielectric constants, and heats of mixing.

                                         RESULTS AND DISCUSSION
   Because of the chemical natures of the substances named in the title, formation of
compounds or of solid solutions is inherently improbable and therefore the study of the
freezing point curves was expected t o yield little of interest. The study was, however,
carried out for the binary systems chloroform - acetic acid and chloroform - water, and
for the ternary system, using a double-junction copper-constantan thermocouple, a Brown
Elektronik recorder, and liquid nitrogen as a coolant. The system acetic acid -water was
previously studied by various workers (4) and the eutectic found t o lie a t 58.1% acetic
acid and -28.5 t o -27.0'. The eutectic of the chloroform - acetic acid system was found
t o lie a t 91.8% CHCl3 by weight, and -67.5" (Fig. 1). In the system chloroform-water,
the eutectic lies a t O.lyo Hz0 and -64.0". The ternary eutectic was found to lie a t -70"
and 90.7% chloroform, 8.3% acetic acid, and 1.0% water. The eutectic trough leading
from the acetic acid -water eutectic to the acetic acid -chloroform eutectic was deter-
mined experimentally. The results are given in Fig. 2, on which has been superimposed
the partial miscibility gap for 0'.
   Two types of temperatures of homogeneity are t o be distinguished in this work, namely,
the temperature a t which two partially miscible liquids become completely miscible
     +                                                               +
(L1 L2 --t L) and the true critical temperatures (L1 V -+ V and/or L2 V + V) ;                  +
Canadian Journal of Chemistry. Volume 41 (1963)

 FIG. 1. Freezing point diagram of system chloroform -acetic acid. Eutectic: -67.5" C, 91.8% CHC13.

       FIG. 2 . 0' isotherm superimposed on a projection of the ternary freezing point diagram.

certain of the mixtures, though not all, will exhibit all three temperatures. For the deter-
mination of all three types of temperatures, the simple apparatus shown in the photograph
(Fig. 3) worked very well. I t consists essentially of a cylindrical block of copper, weighing
about 75 lb and having a radial boring of 1-cm diameter. Slots parallel t o the boring
permit light to pass through. A sealed capillary tube placed in the central boring can be
viewed in a telescope a t a safe distance. Two heating circuits in parallel, wound around
the copper cylinder, make i t possible t o raise the temperature to 350' in about 3 hours.
An ordinary mercury-in-glass thermometer placed in a separate boring of the copper
block gives the temperature. T h e temperature was corrected for exposed stem but with
1090                    CANADIAN JOURNAL O F   CHEMISTRY. VOL.   41, 1963

                                     FIG. 3. Apparatus.

.this relatively crude arrangement we do not believe the results to be more accurate
 than +lo.We saw no reason to refine the temperature measurement, although it would
 have been easy to do so. Temperature was easily controlled and occasional explosions
 constituted no hazard to the operators.
    A possible criticism of this method, as applied to L-V critical temperatures, is that,
 unless the amount of the mixture and the volume of the container are such as to represent
 the critical density, the temperature of disappearance of t h e meniscus is not the true
 critical temperature. Nevertheless, the phenomena as observed usually presented the
 fluctuating striae said to be characteristic of critical phenomena. I t is true that the
 temperature of reappearance of the meniscus was frequently 5-10' lower than that of
 disappearance but this may be a characteristic of mixtures as distinct from pure liquids:
 we reproduce only the temperature of disappearance. The results were also reproducible
 in different experiments and the curve of critical temperature vs. composition was smooth.
 T h e explanation is, no doubt, t h a t in the four-dimensional representation of pressure,
 volume, temperature, and concentration, all curves are very flat in the neighborhood of
 the critical point, or, in plain words, the critical temperature is not very sensitive to
 small changes in volume of the system. At all events, we repeat that in all cases the disap-
 pearance of the meniscus appeared to be a critical phenomenon: cases where the 111eniscus
disappeared by exhaustion of the one or other phase were always discarded. If our figures
 are not true critical temperatures they are sufficiently close for the present purpose. We
note that the critical temperatures of the pure liquids as determined by us are in good
agreement with accepted literature values.
   T h e critical temperatures of the homogeneous binary systems acetic acid - water and
acetic acid - chloroform are reproduced graphically in Figs. 4 and 5. The critical tem-
peratures of homogeneous ternary mixtures are given in Table I, since they cannot be
expressed graphically on a plane diagram to any purpose.

                          W t % CHCI,

       FIG. 4. Acetic acid - water critical temperatures.
     FIG. 5. Chloroform - acetic acid critical temperatures.

                           TABLE I

    Composition of mixture in wt%
                                             Critical temperature
Acetic acid     Water          Chloroform            in OC
1092                    CANADIAN JOURNAL O F               O.
                                               CHEMISTRY. V L 41.   1963

   Within the region of heterogeneity, as mentioned in the introduction, two, and some-
times three, temperatures of homogeneity (2 phase) have t o be determined. The
L1 - L2 -+ L temperature was easily determined in the apparatus described as the tem-
perature a t which one or the other liquid layer vanished by exhaustion (method of
Alexejew). In order to do this accurately, it is necessary to heat very slowly in the neigh-
borhood of the temperature of homogeneity, t o remove the tube and shalile vigorously
t o produce a fine suspension, and then to determine if the suspension persists a t the
temperature of the furnace.
   As a preliminary to the above work, however, the isothermal area of heterogeneity
was determined a t such temperatures as were readily accessible, viz. 0°, 48.5', and 91.5".
Such quantities of the three components as would give approximately equal volumes of
the two layers were stirred in thermostat for 24 hours and then each layer sampled for
analysis. The method of analysis was that of determining acetic acid by titration. A
weighed amount of acetic acid was then added to a separate sample and the density
determined. A previous calibration had been made, a t 25', of the densities of mixtures
lying without the area of heterogeneity (3). These data yielded a complete analysis. We
have subsequently found that the determination of refractive index gives a more sensitive
determination. The results are represented graphically in Figs. 6, 7, and 8. The plait
point compositions are given in Table 11.

                                         T A B L E I1

                                                   Composition in wt%
               Temperature in O C        Chloroform      Water       Acetic acid
                0                           45.0          14.0             41 .O
               25.0 (Brancker. Hunter
                a n d Nash (2))

    The results of the Alexejew method for the heterogeneous region are represented
graphically in Figs. 9, 10, 11, 12, and 13, each of which represents a pseudobinary section
of the solid model having a fixed amount of acetic acid. Some comment is necessary. In
systems containing less than 20y0 acetic acid, the surface of heterogeneity (L1-L2)
intersects the critical surface L1-V, where L1 represents the chloroform layer. This
means that the chloroform layer has a constant composition when passing through the
critical change Ll-V, and this change therefore takes place a t a constant temperature,
independent of the total composition, so long as the composition lies within the miscibility
gap a t the temperature of the L1-V transition. The L1-V transition exhibits experimentally
all the characteristics of a true critical point, except that a (separate) liquid layer, the
aqueous layer, remains behind and this in turn goes through the L2-V critical transition
a t , however, a much higher temperature. Since, however, there is only one liquid, its
composition varies with the total composition of the mixture and therefore the second
critical temperature (Lz-V) is not constant but is a function of total composition.
    As the concen-tration of acetic acid in the pseudobinary mixtures is increased, the Ll-L2
surface separates from the L-V surface and the temperatures of L-V transition now also
become a function of total composition. This is clearly shown in the pseudobinary
sections Figs. 9, 10, 11, 12, and 13.

                  Acetic Acid

          FIG.6. 0' C isotherm.
          FIG.7. 48.5' C isotherm.
          FIG.8. 91.5" C isotherm.

                                WtSb CHCI,

FIG.9. Critical d a t a in system chloroform-water.
FIG. 10. Critical data in pseudobinary system containing   lOy0 acetic acid.

      400-                                                        7
         I                                     20% A c e t i c A c i d

                                Wt% CHCI,

                                               30% A c e t i c A c i d

                               W t % CHCI,

Fig. 11. Critical data in pseudobinary system containing 20% acetic acid.
FIG.12. Critical data in pseudobinary system containing 30y0 acetic acid.
                        CANADIAN JOURNAL O F                 O.
                                                 CHEMISTRY. V L 41,    1963

                                                          40% Acetic Acid

                 300                  O
                                      -      - --- - - - - - -

              FIG. 13. Critical data in pseudobinary system containing 40% acetic acid.

    From the data given it is readily possible to construct the solid (composition-tempera-
 ture) model, on an equilateral triangular base, and we have done this as an aid to the
 teaching of advanced phase rule. The experimental data of this paper are available on

1. C. R.A. \\'RIGHT,C. THOMPSOK, J. T. LEON. Proc. Roy. Soc. (London), 49, 174 (1891).
2 . A. V. BRANCKER, G. HUNTER, A. mT. NASH. J . Phys. Chem. 44,683 (1940).
                     T.         and
3. A. N. CAMPBELL, KARTZMARK, J. M. T. M. GIESKES. Can. J. Chem. 41, 407 (1963).
4. A. FAUCON.Ann. Chim. Phys. 19, 70 (1910). M. HIRATA, HIROSE, and K. KOBAYASHI.      Kagaku
        Kagaku, 23, 403 (1959).