J. Chem. Eng. Data 1981, 26, 413-415
of measurement for NaNO, is a little narrow: &,%NO3
Ps
P
> Ps.NaN02 > P ~ , M s> > Cv,KNO, @S,k+"
CP C"
a
M
The sound velocity in NaNO, is greater than that in KNOBand conversely for P,, while in NaNO, both uand P, are greater than those of NaNO,. Since the size of the cation is smaller in NaNO, than in KNO, the coordination in the structure would make the sound velocity in NaNO, greater than in KNO,. For NaN02, the librational contribution between cation and anion becomes more important than for NaNO, from the structural difference of the anion, and this causes the sound velocity in NaN0, to be greater than that in NaN0, although Ps,NaN02 is greater than Ps,NaN03.
T AL AT
adiabatic compressibility, m2 N-' density, g cm-, isobaric specific heat, J mol-' K-' isochoric specific heat, J mol-' K-' thermal expansivity, K-' molar volume
Literature Clted Fox, F. E. phys. Rev. 1937, 52,973. Hanson, R. L. J . Acoust. SOC.Am. 1949, 21,60. Pellam, J. R.; Oak, J. K. J . Chem. fhys. 1948, 14, 606. Selfen, N. 2. fhp.1938, 108, 681. Roderick, R. L.; Trueii. R. J . Appl. fhys. 1952, 23,267. Klrst, W. E.; Nagle, W. M.; Castner, J. B. Trans. Am. Inst. Chem. En$. 1940, 36, 371. (7) Jam, G. J. "Molten Salts Handbook"; Academic Press: New York, 1967; pp 42, 200. (8) Bodtris, J. O'M.; Richards, N. E. Roc. R . Soc. London, Ser. A 1957, 241,44. (9) Higgs, R. W.; Llovltz, T. A. J . Amust. SOC.Am. 1980, 32, 1108. (IO) Cerisler, P.; Finlels, G.; Doucet, Y. J . Chim. fhys. 1974, 6 , 836. (11) Greenspan, M.; Tschiegg, C. E. J . Acoust. Soc. Am. 1959, 31,75.
(1) (2) (3) (4) (5) (6)
Glossary U
413
sound velocity, m s-' temperature, K change in distance between the conductor and the reflector, m change in time interval, s
Received for review November 21, 1960. Accepted June 12, 1981.
Isobaric Binary Vapor-Liquid Equilibria for the Systems Allyl Alcohol-Toluene and Toluene-Benzaldehyde K. Venkateswara Rao," A. Raviprasad, and C. Chiranjlvl Department of Chemical Engineering, Andhra University, Visakhapatnam 530003, India
Isobaric vapor-llquld equlllbrla for the binary systems allyl alcohol-toluene and toluene-benzaldehyde were measured at 760 f 1 mmHg pressure by uslng a modified Jones vapor recirculatory stlll. The system allyl alcohol-toluene has shown posltlve devlatlons from Raoult's law. Redllch-Klster, Wilson, NRTL, and UNIQUAC models were employed to correlate the experimental data. The system toluene-benzaldehyde was nearly Ideal In-Its behavlor, and the data were correlated by the method of Hlrata.
Isobaric vapor-liquid equilibrium (VLE) data at 760 f 1 mmHg pressure were determined for the systems allyl alcohol-toluene and toluene-benzaldehyde. The VLE data of the former system find use in the purification of allyl alcohol. Toluene-benzaldehyde mixtures occur in the manufacture of benzaldehyde from toluene. Hence, knowledge of vapor-liquid equilibrium data on these systems is of use in the design of separation equipment. The VLE data on allyl alcohol-toluene mixtures at 760-mmHg pressure and at 90 O C were reported in the literature ( 7 , 2). Experlmental Section Purlty of Chemicals. Analytical-grade allyl alcohol and toluene were distilled in a Defton column, and the middle fractions were collected and used. Benzaldehyde, also of analytical grade, was distilled under vacuum, and the heart cuts of the distillate were used in the experimental runs. The physical 002 1-9568181I1 726-0413$0 1.2510
constants of the pure components, compared with the literature values (3),are presented in Table Ialong with the source of the chemicals. Special care was taken to see that benzaldehyde did not come in contact with oxygen in the air during the distillation and use in the equilibrium still by connecting an alkaline pyrogallol tower to the system to prevent atmospheric oxidation of benzaldehyde. Apparatus, Technlque, and Analysis. The VLE data were measured by using the Jones vapor recirculatory still modified by Ward (4). The operation of the still was described previously (5).Vapor condensate and liquid-phase samples were analyzed by precision refractometry (Abbe's refractometer) at 30 f 0.1 O C with the aid of calibration curves constructed from n, vs. x data measured for synthetic binary mixtures. The equilibrium temperatures were taken with a mercury-inglass thermometer having an accuracy of f O . l O C . The estimated precision of the equilibrium mixture composition measurements was f0.006 mole fraction for allyl alcohol-toluene mixtures and fO.0 1 mole fraction for toluene-benzaldehyde mixtures. Results and Discussion
The vapor-liquid equilibrium data for the two systems studied are given in Tables I1 and 111. The liquid-phase activity coefficients for each component were calculated from the experimental data by using eq 1. At atmospheric pressure, the ratio 4,"/4: was found to be around unity. Hence, it was ignored in the calculations. This is in 0 1981 American Chemical Society
414
Journal of Chemical and Engineering Data, Vol. 26, No. 4, 1981
Table 1. Physical Constmts of the Chemicals
a
refractive index at 30 “C
bp, “C tit. (3)
chemical
source
exptl
allyl alcohol toluene benzaldehyde
BDH Indiaa BDH Indiaa BDH Indiaa
97.1 110.6 179.0
97.08 110.6 179.0
exptl
lit. (3)
vapor-pressure equation
1.4090 1.4912 1.5400
1.4090 1.4912 1.5401b
l O g P = 11.18664 - 4068.457/(392.732 + t ) l o g P = 6.95105 - 1342.310/(219.187 + t ) logP=45.625 -4536.82/T- 12.3189 log T
BDH Chemicals Division, Glaxo Laboratories (India) Limited.
Calculated by using dnldt.
Table II. Vapor-Liquid Equilibrium Data for the System Allyl Alcohol (1)-Toluene (2) at a Pressure of 760 mmHg x, 0.017 0.028 0.121 0.152 0.162 0.194 0.247
temp,’C 107.1 105.5 98.0 97.0 96.1 95.3 93.8
Y, 0.079 0.134 0.351 0.386 0.407 0.419 0.467
xl 0.452 0.530 0.603 0.756 0.909 0.922
temp,’C 92.0 91.7 91.5 91.9 94.0 94.1
Table IV. Model Parameters for AUyl Alcohol-Toluene System at 760-mmHg Pressure
A
B C
D B’ C‘ D’
Table 111. Vapor-Liquid Equilibrium Data for the System Toluene (1)-Benzaldehyde (2) at a Pressure of 760 mmHg
( A l z - A,,),
Yl
temp, “C
Xl
exptl
calcd (Raoult’s law)
166.4 160.9 153.2 146.9 141.1 138.6 134.8 131.8 126.7 122.7 114.5
0.077 0.097 0.158 0.209 0.281 0.327 0.380 0.424 0.490 0.615 0.846
0.418 0.522 0.583 0.658 0.745 0.790 0.836 0.863 0.896 0.939 0.993
0.400 0.462 0.600 0.679 0.758 0.795 0.831 0.855 0.885 0.933 0.989
a
+
- C ( 6 ~ 1 ~- 21) -
(ii) Wilson ( 70): In y, = -In (xl In Y. -, =
+ A12x2)+ x 2
[ [
x 1 ;Il2x2
x2
+ AZ1x1
+
NRTL Model ( a l z= 0) 545.8001,426.7179
0.0131
(glz-gzz),(g,, - g , J
(g12-g2J3 k2, -g1J
NRTL Model (T-x Data, c y l z = 0.4) 791.585, 498.789
0.020
UNIQUAC Model ( u I z - u z z ) ,(u,, - u l l ) -20.6189, 417.2956
0.013
Absolute average deviation in vapor composition.
(iii) NRTL with a preset value of alp( 77):
(2)
]
-In ( x 2 - A2,x1)- x1 x 1 - ! 2 1 2 x 2- x 2 A,,A 2 1 ~ 1 (5)
0.0121
0.0123
A, = (qL/VIL)exP[-(XJ - b ) / R T I
D(x2 - xl)(l - 8 ~ 1 x 2 )
Y la
Wilson Model 900.2861, 294.6341
(A21 - AII)
where
accordance with the suggestion of Hudson and Van Winkle (6). The vapor-pressureequations used in the calculationsare listed in Table I. A/@/ Akohol-Toluene System. The system exhibited positlve deviations from Raouk’s law as observed from the activity-coefficient data. It formed an azeotrope at 62 mol % allyl alcohol and at a temperature of 91.4 OC. The experimental azeotropic data are found to be in good agreement with the data of Horsiey (7). Further, the experimental areotropic composltion was verified and confirmed in a laboratory fractionation set up under total reflux conditions. The experimental T-x-ydata were found to be thermodynamically consistent by the Herington method (8). Four models describing the activity coefficient in the liquid phase were tried to fi the experimental data. (i) Redlich-Kister polynomial expansions as modified by Chao and Hougen (9): In ( Y l / Y 2 ) = A B(x2 - xi)
parameters Redlich-Kister 0.0228 0.5330 0.0470 0.0355 0.6352 0.07874 0.07 5 37
Yl
0.555 0.577 0.607 0.696 0.825 0.861
Journal of Chemical and Engineering Data, Vol. 26, No. 4, 198 1 415
Germany, for help in processing part of the data. Glossary
parameters in Redlich-Kister model
parameters in NRTL model energetic parameters in NRTL model 0
0.2
0.6
0.4
1.0
0.8
x,
as defined in eq 8 and 9 Na D line refractive index saturation pressure of pure component excess free energy function pure-component area parameter gas constant pure-component volume parameter temperature energetic parameters in UNIQUAC model
Flgure 1. Equlllbrlum curve for the system allyl alcohol-toluene: (0) experimental; (- - -) Wilson; (-) Redllch-Klster.
V I 0
I
02
I
I
I
04
I
06
I
1
08
4
I
X
1.0
Y z
x-
Flgure 2. EquHlkkwn curve for the system toluene-benzaldehyde: (0) experimental data: (-) Raoult’s law.
liquid molar volume of pure component i liquid-phase mole fraction vapor-phase mole fractlon lattice coordination number
Greek letters
The parameters of the Redlich-Kister equations were obtained by following the method of Chao ( 73),whereas the parameters of the Wilson, NRTL, and UNIQUAC equations were evaluated by using a nonlinear optimization method ( 74) with an objecthre function based on the calculated and experimental activity coefficients. Further, the model parameters for NRTL equations were determined by following the procedure described by Fields ( 75) using the experimental T-x data. All of the model parameters are listed in Table IV along with the absolute average deviations between the experimental and calculated vapor-phase mole fractions. The Redlich-Kister and Wilson models represented the experimental data better than the other models. Figure 1 shows the equilibrium curve for the allyl alcohol-toluene system denoting the calculated vapor composition values using Redlich-Kister and Wilson equations. Toluene-Benzaldehyde System. The mixture of toluenebenzaldehyde was found to be nearly ideal. The y-xdiagram is shown in Figure 2. The vapor composition of the more volatile component was calculated from the vapor-pressure data by using Raoult’s law and was compared with the experimental values. The absolute average deviation in y 1 is 1.52%. A linear relation resulted when the data were correlated by the method of Hirata ( 16). As the data conform with Raoult’s law wlthin the experimental errors, the data are thermodynamically consistent. Acknowledgment
We are grateful to Dr. Jurgen Gmehling, Lehrstuhl fur Technische Chemle, 6, Universitat Dortmund, Federal Republic of
nonrandomness constant in NRTL model liquid-phase activity coefficient area fraction in UNIQUAC model energetic parameter in Wilson model total pressure volume fraction in UNIQUAC model standard-state fugacity coefficient of pure component vapor-phase fugacity coefficient Subscripts
i , j , 1, 2 components
Llterature CHed Froiov, A. F.; Loginova, M. A.; Samarina, G. M.; Meinik, L. V.; Obedkova. L. V. Zh. Ml. Khlm. (Lenlngrad)1973, 46. 1499. Venkateswara Rao, K.; Raviprasad, A.; Chiranjivi, C. J . Chem. Eng. Date 1979, 24, 272. Welssberger, A.; Proskauer, S. E.; Riddick, A. J.; Toops, E. R., Jr. “Organic Solvents”, 2nd ed.; Interscience: New York, 1955. Ward, S. H. Ph.D. Thesis, The University of Texas, Austin, TX, 1952. Raviprasad, A.; Venkateswara Rao. K.; ChhaflHvi, C. Indian Chem. Eng; 1977, 79, 29. Hudson, J. W.; Van Winkle, M. Ind. Eng. Chem. Process Des. Dev. 1970, 9 , 467. Horsley, L. H. “Azeotropic Data”; American Chemical Society: Washinaton. DC. 1952. Herington,.€. F. G. J . Inst. Pet. 1951, 37, 457. Chao. K. C.; Hougen, 0. A. Chem. Eng. Sei. 1958. 7, 246. Wilson, G. M. J. Am. C m . Soc. 1964, 6. 127. Renon, H.; Prausntk, J. M. AIChE J. 1968, 74, 135. Abrams, D. S.; Prausnitz. J. M. A I C M J . 1875, 27. 116. Chao, K. C. Ind. Eng. Chem. 1959. 51, 93. Nelder, J.; Mead, R. Comput. J. 1962, 7 , 308. Fields, R. 6. “Prediction of Vapor-LiquM Equilibria Using the Nonrandom Twoliquld EqWtlon”, Report NO. C.D.-2184, DepNhMnt Of SCC entiflc and Industrial Research, New Zeaknd, Aprll 1976. Hirata, M. Jpn. Sei. Rev. Ser. 1952. 3. 265.
Received for review December 8, 1980. Revised Manuscript Received June 23, 1981. Accepted July 13, 1981.
