Phase Equilibria of the Ternary System Benzene + Cyclohexane + 1-Pentanol at 101.3 kPa Beatriz Orge, Gonzalo Marino, Miguel Iglesias, and Jose ´ Tojo* Departamento de Ingenierı ´a Quı ´mica, Facultad de Ciencias, Universidad de Vigo, 36200 Vigo, Spain Vapor-liquid equilibrium at 101.3 kPa has been measured for the ternary system benzene + cyclohexane + 1-pentanol in an isobaric equilibrium still with secondary recirculation of both vapor and liquid phases. Satisfactory results were obtained for the prediction of activity coefficients and the equilibrium compositions with the ASOG and UNIFAC group contribution models; low standard deviations of vapor mole fraction and temperature were computed. Azeotropic behavior was observed only in the benzene + cyclohexane mixture. The correlation parameters for the Tamir-Wisniak and UNIQUAC equations are presented. Introduction Experimental data collections of phase equilibria for ternary or higher order complexity are scarce because of the time-consuming experimental procedures to obtain a complete description of every mixture. As an extension of our earlier work concerning vapor-liquid or liquid-liquid equilibria (VLE or LLE) 1-3 and thermochemical proper- ties, 4-6 we present new phase equilibrium data concerning 1-pentanol as an alternative extractive rectification solvent for the azeotropic mixture benzene + cyclohexane. This paper presents VLE data for the mixture benzene + cyclohexane + 1-pentanol at a pressure of 101.3 kPa; no literature data are available for this system. Such ther- modynamic data can be obtained from available predictive models of functional molecular group contribution, as the well-known UNIFAC method. 7 These models require com- plete and fully updated experimental data in order to compute group interaction parameters and reproduce the behavior of systems at other mixing or operation conditions. The application of the ASOG, UNIFAC, and their modifica- tion group contribution methods leads to satisfactory predictions in terms of activity coefficients and composi- tions for this mixture. Fitting parameters corresponding to boiling temperatures and activity coeficient mole fraction dependence are presented. Experimental Section Chemicals. All chemicals were Merck chromato- graphic grade. Purification was attempted by ultrasonic degassing and molecular sieve drying (4 Å, 1 / 16 in.). The purity of materials was checked by gas chromatography and found to be better than 99.9 mass % for benzene and cyclohexane and 99.0 mass % for 1-pentanol, and the maximun water contents of the pure liquids (Metrohm 737 Coulometer) were 3.0 × 10 -2 , 4.9 × 10 -3 , and 2.3 × 10 -1 mass % for benzene, cyclohexane, and 1-pentanol, respec- tively. Their purity was also checked by determining their densities and refractive indices at 298.15 K and their normal boiling temperatures T b (Table 1), prior to the measurements. Apparatus and Procedure. VLE measurements were carried out under an atmosphere of dry argon (less than 3 ppmv in water) in a modified all-glass Othmer-type ebulliometer with secondary recirculation of both the liquid and vapor phase. 9 Thermal isolation was ensured because the whole apparatus was insulated except for the part corresponding to the vapor condenser. Boiling temperatures of mixtures were measured with a Yokogawa 7563 digital thermometer with a precision of (10 -2 K (temperature scale IPTS-75), calibrated with an Anton Paar MKT-100 digital thermometer (precision (10 -3 K and temperature scale ITS-90) over the entire range of work temperatures. Pressure was kept constant at (101.3 ( 0.1) kPa by a controller device which introduced argon to the apparatus in order to maintain the pressure difference with respect to the pressure in the laboratory. Each experiment was continued at least for 1 h after the boiling temperature had become stable. Samples of both the liquid and vapor phases were taken at low temperature by a built-in refrigeration device and sealed in ice-cooled graduated test tubes to prevent evaporation leakage. Once the sample tempera- ture became stable using a PolyScience controller bath model 9510 with a temperature stability of (10 -2 K, the samples were analyzed by measuring their refractive indices and densities at 298.15 K. Densities of the pure liquid and mixtures were measured with an Anton Paar DSA-48 densimeter (accuracy of (10 -4 g‚cm -3 ) and refractive indices with an automatic refractometer ABBEMAT-HP Dr. Kernchen (accuracy of (5 × 10 -5 ). The estimated uncertainty for mole fractions was determined as (7 × 10 -3 . Results and Discussion Equilibrium Equation and Activity Coefficients. Experimental density (F) and refractive index (n D ) values at 298.15 K for this ternary system as a function of x i have been published previously. 4 Such physical properties were applied in order to compute the mixing composition by * To whom correspondence should be addressed. Fax: +34 986 812382. E-mail: jtojo@uvigo.es. Table 1. Densities G, Refractive Indices nD, and Normal Boiling Temperatures Tb of the Pure Components F(298.15 K)/ (g‚cm -3 ) nD(298.15 K) Tb/K component exptl lit. a exptl lit. a exptl lit. a benzene 0.8736 0.87370 1.49692 1.49792 353.16 353.250 cyclohexane 0.7737 0.77389 1.42320 1.42354 353.79 353.888 1-pentanol 0.8110 0.8112 1.40782 1.4079 410.78 410.95 a TRC Thermodynamic Tables. 8 410 J. Chem. Eng. Data 2001, 46, 410-413 10.1021/je000086b CCC: $20.00 © 2001 American Chemical Society Published on Web 01/30/2001