Isobaric vapor-liquid equilibrium for methyldichlorosilane-methyltrichlorosilane-dimethyldichlorosilane system

Isobaric vapor-liquid equilibrium for methyldichlorosilane-methyltrichlorosilane-dimethyldichlorosilane system 1 2 first-author †E-mail:qiuzm@ncu.edu.cn †E-mail:qiuzm@ncu.edu.cn qiuzm@ncu.edu.cn 1 1 1 School of Environmental Science and Engineering, Nanchang University, Nanchang 330029, China School of Environmental Science and Engineering, Nanchang University, Nanchang 330029, China Key Laboratory of Poyang Lake Ecology and Bio-resource Utilization, Nanchang University, Nanchang 330029, China Key Laboratory of Poyang Lake Ecology and Bio-resource Utilization, Nanchang University, Nanchang 330029, China This paper describes measurement of vapor-liquid equilibrium (VLE) data of methyldichlorosilane-methyltrichloro silane-dimethyldichlorosilane system and that of the three binary systems at 101.325 kPa with a new pump-ebulliometer. The equilibrium composition of the vapor phase was calculated from pTx by indirect method. The model parameters of the liquid activity coefficient of the Wilson, NRTL, Margules and van Laar equations was corrected by the least square method. The ternary system VLE data were predicted by the Wilson equation, with the calculated boiling points showing good agreement with the experimental data. Ebulliometer Vapor-liquid equilibrium Methyldichlorosilane Methyltrichlorosilane Dimethyldichlorosilane INTRODUCTION 1 Vapor-liquid equilibrium (VLE) data are very important in the design and operation of separation processes in chemical industry. Such information can be obtained experimentally or estimated by using generalized methods for calculation of the properties of mixtures. For ideal system, it is relatively easy to estimate vapor-liquid equilibrium. However, most systems of industrial interest show deviations from the ideal behavior. The methyldichlorosilane-methyltrichlorosilane-dimethyldichlorosilane and constituent binary system are of importance in the production of silicon, while VLE data of these systems are very scarce in the literature. Only in the past several years did some researchers (Ana et al., 1995 1995 ; Liviu et al., 1995 1995 ; Wu and Cai, 1997 1997 ; Xie et al., 1998 1998 ) begin to study isobaric vapor-liquid equilibrium for multicomponent systems containing halogenated silane. In order to provide some necessary basic data on silicon derivatives, these systems are selected for study. The VLE data are measured by the total pressure-temperature-liquid composition ( pTx ) method. In this study, the VLE data are determined at 101.325 kPa for methyldichloroslane-methyltrichlorosilane-dimethyldichlorosilane and constituent binary system. The results calculated by the activity coefficient models were shown to be in good agreement with the experimental data. EXPERIMENTAL DETAILS 1 Materials 2 Methyldichlorosilane, methyltrichlorosilane and dimethyldichlorosilane were obtained from Xinhuo Chemical Factory, Jiangxi. All the chemicals were purified by distillation in a laboratory column to a purity of more than 99.5% by gas chromatography (GC) analysis. Methyldichlorosilane, methyltrichlorosilane and dimethyldichlorosilane should be purified just before measurement because they may easily react with water when exposed to the air. Comparison between some of the measured data and literature data is given in Table 1 1 . 2 Component 2 Boiling point (K) Lit. Exp. Methyldichlorosilane 314.15 (Dean, 2003 2003 ) 314.41 Methyltrichlorosilane 339.55 (Dean, 2003 2003 ) 339.50 Dimethyldichlorosilane 343.15 (Dean, 2003 2003 ) 343.46 Apparatus and procedure 2 A new type of magnetical pump-ebulliometer, described in detail by Qiu et al.( 1995 1995 ) was used for measuring the boiling points with different liquid phase compositions (Qiu et al., 1997 1997 ). The apparatus was an all-glass dynamic recirculation still with total volume of about 1.00×10 −4 m 3 . During the run, the still was submerged in a constant temperature bath about 3 °C below the equilibrium boiling point. The atmospheric pressure, P , was measured by a Fortin-type mercury barometer. Since the barometric pressure changed slightly, the experimental temperatures were automatically corrected to that at 101.325 kPa. The equilibrium temperature, T , was measured using a standard mercury thermometer with accuracy of 0.1 K. In each experiment, a known mass of the material was introduced into the still from the injector and was heated at a fixed pressure of 101.325 kPa by an automatic regulation system. The accuracy of the liquid and vapor phase was constant. The liquid compositions component was added into the still. The vapor compositions, y i , were calculated from the experimental pTx data by an indirect method (Qiu et al., 1994 1994 ). This pTx method was based on the Q function (The function of molar excess Gibbs energy) and had rigorous thermodynamic consistency (Hu, 1982 1982 ). RESULTS AND DISCUSSION 1 At low and moderate pressures, the pointing factor is negligible, so the equilibrium relations is as follows $$Py_i \hat \varphi _i = P_i^s \varphi _i^s x_i \gamma _i$$ where y i is the mole fraction of component i in the vapor phase, $$\hat \varphi _i$$ is the fugacity coefficient of component i in the vapor phase, P is the total pressure, γ i is the activity coefficient of component i in the vapor phase, x i is the mole fraction of component i in the liquid phase, φ i s is the pure component fugacity coefficient at saturation, and P i s is the vapor pressure of pure component i . The coefficients of $$\hat \varphi _i$$ and φ i s were evaluated using the second virial coefficients from the critical data, respectively. The vapor pressure of pure component, P i s is calculated by the Antoine equation. The critical data and Antoine constants of the pure component are given in Table 2 2 . A, B, C: Antoine constants; T c : Critical temperature; P c : Critical pressure; V c : Critical volume Component A B C T c (K) P c (kPa) V c (10 −6 m 3 /mol) Methyldichlorosilane (Rugina and Sacarescu, 1992 1992 ) 7.0484 1179 242 475 3901.1 290 Methyltrichlorosilane (Rugina and Sacarescu, 1992 1992 ) 6.8721 1167 226 518 5562.7 339 Dimethyldichlorosilane (Rugina and Sacarescu, 1992 1992 ) 7.1435 1328 241 520 3485.6 350 The experimental data were correlated using the Wilson, NRTL, Margules and van Laar equations f . The optimal parameters were obtained by minimization of the objective function F (Qiu et al., 1994 1994 ). $$F = \sum |\ln (r_1 )_{{\text{cal}}} - f_1 (\overline {a_i } )|_k^2 + \sum |\ln (r_2 )_{{\text{cal}}} - f_2 (\overline {a_i } )|_k^2$$ where γ 1 , γ 2 are the activity coefficient of component 1and component 2 in the liquid phase, $$\overline {a_i }$$ is model parameter. k is the point of difference. The binary systems VLE data were used to plot the three graphs in Fig. 1 1 . • Calculated vapor phase points; − Calculated temperature curve; ◦ Experimental temperature data The model parameters of liquid activity coefficient and the absolute average deviations are listed in Table 3 3 showing that the deviations of Wilson, NRTL, Margules equations are reasonably small, especially that of the Wilson equation, while the deviations of the van Laar equation is a little bigger. So the activity coefficient models are suitable for representing the binary experimental data. The binary parameters of the Wilson model given in Table 3 3 were used to predict the VLE data of the ternary system. The results were satisfactory. The VLE data of the ternary system and the average deviation in boiling temperatures are listed in Table 4 4 . d t =∑| T − T mod |/ N d y =∑| y cal − y mod |/ N N =Number of data points; T mod : Calculated boiling point from model (K); y cal : Calculated vapor phase mole fraction from pTx ; y mol : Calculated vapor phase mole fraction from mode Equation Parameters and deviations Methyldichlorosilane (1)+methyltrichlorosilane (2) Methyldichlorosilane (1)+dimethyldichlorosilane (2) Methyltrichloroslane (1)+dimethyldichloroslane (2) Wilson A −38.1 −120.1 −28.9 B 0.05 0.1 0.11 d t * * 0.17 0.14 0.07 d y ** ** 0.004 0.0089 0.006 NRTL A −66.5 −0.1 −142.2 B 0.01 −148.7 −1.4 A ij 0.05 0.3 0.08 d t 0.14 0.15 0.09 d y 0.0038 0.0081 0.0039 Margules A −0.125 −0.41 −0.34 B −0.11 −0.12 −0.13 d t 0.15 0.5 0.28 d y 0.0039 0.0042 0.007 van Laar A 0.001 0.0001 0.0001 B 0.003 0.05 0.03 d t 0.61 1.34 1.15 d y 0.0074 0.0157 0.0155 d t =∑| T exp − T mod |/ N ; T mod : Calculated boiling point from model ; T exp : Experimental boiling point; x 1 , x 2 , x 3 : Liquid-phase mole fraction of (1) (2) (3); y 1,cal , y 2,cal , y 3,cal : Calculated vapor-phase mole fraction of (1) (2) (3) from pTx ; N : Number of data point No. T mod (K) T exp (K) x 1 x 2 x 3 y 1,cal y 2,cal y 3,cal 1 318.36 318.67 0.7488 0.2253 0.0259 0.8638 0.1255 0.0107 2 319.01 319.22 0.7226 0.2174 0.0600 0.8505 0.1238 0.0257 3 319.73 320.03 0.6948 0.2091 0.0961 0.8353 0.1219 0.0428 4 320.41 320.61 0.6689 0.2012 0.1299 0.8202 0.1199 0.0599 5 320.90 321.10 0.6509 0.1959 0.1532 0.8092 0.1185 0.0723 6 321.56 321.78 0.6269 0.1886 0.1845 0.7938 0.1165 0.0898 7 321.98 322.40 0.6118 0.1841 0.2041 0.7837 0.1152 0.1012 8 322.57 323.37 0.5910 0.1778 0.2312 0.7692 0.1132 0.1176 9 323.12 323.81 0.5717 0.1720 0.2563 0.7553 0.1113 0.1334 10 323.74 324.18 0.5504 0.1656 0.2840 0.7393 0.1092 0.1515 11 325.04 325.60 0.5327 0.0268 0.4405 0.7326 0.0186 0.2488 12 325.61 326.02 0.5000 0.0865 0.4135 0.7018 0.0606 0.2376 13 326.03 326.34 0.4774 0.1278 0.3948 0.6802 0.0902 0.2296 14 326.26 327.03 0.4651 0.1503 0.3846 0.6682 0.1065 0.2253 15 326.52 327.12 0.4518 0.1746 0.3736 0.6552 0.1243 0.2205 16 326.93 327.38 0.4314 0.2118 0.3568 0.6348 0.1520 0.2132 17 327.20 327.50 0.4181 0.2361 0.3458 0.6213 0.1704 0.2083 18 327.55 328.02 0.4017 0.2661 0.3322 0.6044 0.1935 0.2021 19 327.80 328.18 0.3901 0.2872 0.3227 0.5922 0.2099 0.1978 20 328.07 329.03 0.3780 0.3095 0.3125 0.5794 0.2275 0.1931 21 328.19 329.08 0.3725 0.3195 0.3080 0.5734 0.2355 0.1910 22 339.34 340.78 0.0242 0.7459 0.2299 0.0544 0.7391 0.2065 23 337.44 338.58 0.0711 0.7101 0.2188 0.1498 0.6665 0.1838 24 335.40 336.04 0.1248 0.6690 0.2062 0.2457 0.5931 0.1612 25 334.06 334.86 0.1621 0.6405 0.1974 0.3055 0.5472 0.1473 26 332.90 333.80 0.1956 0.6149 0.1895 0.3552 0.5090 0.1358 27 331.88 332.98 0.2264 0.5913 0.1823 0.3979 0.4761 0.1260 28 330.44 331.61 0.2713 0.5570 0.1717 0.4558 0.4313 0.1129 29 329.76 330.53 0.2933 0.5402 0.1665 0.4824 0.4107 0.1069 30 328.84 329.87 0.3239 0.5180 0.1593 0.5177 0.3834 0.0990 Deviations: d t =0.63; d t (max)=1.44 CONCLUSION 1 Vapor-liquid equilibrium for a ternary system of methyldichloroslane-methyltrichlorosilane-dimethyldichlorosilane and constituent binary systems were measured at 101.325 kPa. The equilibrium composition of the vapor phase was calculated from pTx by the indirect method. The experimental data were correlated using the Wilson, NRTL, Margules and van Laar equations. It was shown that the deviations of the Wilson, NRTL, Margules equations were reasonably small, while the deviation of the van Laar equation was a little bigger. Ternary system VLE data were predicted by the Wilson equation; the calculated boiling points accorded well with experimental data. The results showed that the calculated boiling point is fitted by the models which satisfied the need for the design and operation of separation processes. References D Ana , M Rosa , S Margarita . Isobaric vapor-liquid equilibria of tetrachloroethylene with 1-butanol and 2-butanol at 6 and 20 kPa . J Chem Eng Data , 1995 . 40 ( 1 ): 290 - 292 . . Isobaric vapor-liquid equilibria of tetrachloroethylene with 1-butanol and 2-butanol at 6 and 20 kPa JA Dean . Lange’s Handbook of Chemistry , 2003 . 2nd Ed : Beijing Science Press , (in Chinese) . Lange’s Handbook of Chemistry Y Hu . Fluid Molecule Thermodynamic , 1982 . : Beijing Higher Education Press , (in Chinese) . Fluid Molecule Thermodynamic S Liviu , M Mihai , L Nicolae , S Gabriela . Vapor-liquid equilibria for dichlorodimethylsilane plus heptane . J Chem Eng Data , 1995 . 40 ( 1 ): 71 - 73 . . Vapor-liquid equilibria for dichlorodimethylsilane plus heptane ZM Qiu , ZC Luo , Y Hu . Vapor-liquid equilibria of two binary systems containing methylal . Journal of Chemical Engineering of Chinese Universities , 1994 . 8 ( 2 ): 106 - 110 . (in Chinese) . Vapor-liquid equilibria of two binary systems containing methylal ZM Qiu , ZC Luo , Y Hu . The new motive cottrell pump ebulliometer . Journal of Nanchang University (Engineering & Technology) , 1995 . 17 ( 4 ): 6 - 11 . (in Chinese) . The new motive cottrell pump ebulliometer ZM Qiu , ZC Luo , Y Hu . The pump ebulliometer . Journal of Chemical Engineering of Chinese Universities , 1997 . 11 ( 1 ): 74 - 77 . (in Chinese) . The pump ebulliometer T Rugina , L Sacarescu . Isobaric vapor-liquid equilibric for the binary systems CH 3 SiHCl 2 with (CH 3 ) 3 SiCl, (CH 3 ) 2 SiCl 2 , CH 3 SiCl 3 , or SiCl 4 . J Chem Eng Data , 1992 . 37 143 - 145 . . Isobaric vapor-liquid equilibric for the binary systems CH3SiHCl2 with (CH3)3SiCl, (CH3)2SiCl2, CH3SiCl3, or SiCl4 Q Wu , ML Cai . Relative volatility and separation for halogenosilanes . Chemical Engineering , 1997 . 25 ( 2 ): 57 - 62 . (in Chinese) . Relative volatility and separation for halogenosilanes JJ Xie , QM Pan , ZR Pan . Study of vapor-liquid equilibria for polydimethylsiloxane/solvents by using gas-liquid chromatography . Chemical Engineering , 1998 . 26 ( 3 ): 49 - 54 . (in Chinese) . Study of vapor-liquid equilibria for polydimethylsiloxane/solvents by using gas-liquid chromatography 10.1631/jzus.2005.B0559 Project supported by the Foundation of Science and Technology, the Ministry of Education (No. 03071), and The Natural Science Foundation of Jiangxi Province (No. 0320013), China 2004-10-26 2005-03-01 2005-05-22 2022-01-05 Copyright © 2005, Journal of Zhejiang University Science 2005 Journal of Zhejiang University-SCIENCE B (Biomedicine and Biotechnology) 06 6B