This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Isothermal vapour–liquid equilibrium for cyclicethers with 1-chloropentane
Beatriz Giner, Ana Villares, Santiago Martın, Carlos Lafuente, Felix M. Royo ∗Departamento de Quımica Organica-Quımica Fısica, Facultad de Ciencias, Universidad de Zaragoza,
Ciudad Universitaria, Zaragoza 50009, Spain
Received 26 September 2006; received in revised form 25 October 2006; accepted 30 October 2006Available online 7 November 2006
bstract
Isothermal vapour–liquid equilibrium measurements for mixtures containing cyclic ethers: tetrahydrofuran, tetrahydropyran, 1,3-dioxolane or,4-dioxane and 1-chloropentane at the temperatures of 298.15, 313.15 and 328.15 K are reported. The thermodynamic consistency of the VLE
easurements was satisfactorily checked with the van Ness method. Activity coefficients were correlated with Wilson, NRTL, and UNIQUAC
quations. The calculated excess Gibbs functions for tetrahydrofuran and tetrahydropyran are negative over the whole composition range while for,3-dioxolane and 1,4-dioxane the excess Gibbs functions are positive.
We have previously reported measurements on isobaricapour–liquid equilibrium of mixtures containing a cyclic ethernd normal and branched chloroalkanes [1–5]. Now we starthe determination of vapour–liquid equilibrium of these kind of
ixtures at isothermal conditions that provides more interestingnformation from a theoretical point of view.
Here we present isothermal vapour–liquid equilibriumeasurements for the mixtures formed by a cyclic ether: tetrahy-
rofuran, tetrahydropyran, 1,3-dioxolane or 1,4-dioxane with-chloropentane at the temperatures of 298.15, 313.15 and28.15 K. The VLE experimental results have been checked forhermodynamic consistency and the corresponding activity coef-cients and excess Gibbs functions have been correlated with
he following equations: Wilson [6], NRTL [7] and UNIQUAC8].
To our knowledge, the VLE data for these mixtures have noteen reported before.
The liquids used were: tetrahydrofuran (better than9.5 mol%), 1,3-dioxolane, 1,4-dioxane and 1-chloropentanebetter than 99 mol%) obtained from Aldrich and tetrahydropy-an (better than 99 mol%) provided by Acros. No additionalurification has been carried out. A comparison between exper-mental densities and vapour pressures and literature values9–14] at 298.15 K is reported in Table 1.
.2. Methods
The vapour–liquid equilibrium was studied using an all-glassynamic recirculating type still that was equipped with a Cottrellump. It is a commercial unit (Labodest model) built in Ger-any by Fischer. The equilibrium temperature were measured
o an accuracy of ±0.01 K by means of a thermometer (model25 with a PT100 probe) from Automatic Systems Laborato-ies, and the pressure in the still was measured with a Digiquartz
35-215A-102 pressure transducer from Paroscientific equippedith a Digiquartz 735 display unit. The uncertainty of the pres-
ure measurements is ±0.005 kPa. Equipment and experimentalrocedure has been previously described [15,16]. The vapour
B. Giner et al. / Fluid Phase Equilibria 251 (2007) 8–16 9
Table 1Physical properties of the pure compounds, comparison of their densities and vapour pressures with literature values at 298.15 K and cross second virial coefficients
and UNIQUAC equations, the mixture nonrandomness parame-ter α12 in the NRTL equation was fixed at 0.3. Estimation of theadjustable parameters of the equations was based on minimiza-tion using the Simplex method [19] of the following objective
ressures of the pure compounds at work temperatures are gath-red in Table 1.
The composition of both phases has been analyzed by mea-uring simultaneously the density and the speed of sound of theample, the procedure has been described in a previous paper5]. Densities and speed of sound of the samples were measuredith an Anton Paar DA-48 densimeter, which was calibratedsing deionized twice distilled water and dry air. The error inhe determination of the mole fraction composition of liquid andapour phases is estimated to be ±0.0004.
The proper operation of the different devices was periodicallyhecked and rearranged if necessary.
. Results and discussion
The vapour–liquid equilibrium data, P, x1, y1 are given inable 2 and the pressure-composition diagrams are shown inigs. 1–4.
The system 1,4-dioxane with 1-chloropentane shows max-mum pressure azeotropes at the temperatures of 298.15 Kx1az = 0.706, Paz = 5250 Pa), 313.15 K (x1az = 0.706, Paz =0,840 Pa) and 328.15 K (x1az = 0.730, Paz = 20,600 Pa). Theocation of the azeotropic points were made using y1 − x1 ver-us x1 diagrams to determine x1az at y1 − x1 = 0 together with
versus x1 diagrams, where P should be a maximum at x1azFig. 5).
Thermodynamic consistency of the experimental results haseen checked using van Ness method [17] described by Fre-enslund et al. [18] using a third order Legendre polynomial forhe excess Gibbs functions. This method considers that experi-
ental data are thermodynamically consistent if mean absolute
eviation between calculated and measured vapour phase com-ositions, �y, is lower than 0.01. All the studied mixtures arehermodynamically consistent and values of �P and �y areathered in Table 3.
Fee
123.729 −1497125.950 −1393
Activity coefficients were correlated with the Wilson, NRTL
10 B. Giner et al. / Fluid Phase Equilibria 251 (2007) 8–16
Table 2Isothermal VLE data of the binary mixtures: experimental pressure, P, liquid-phase, x1, and vapour-phase, y1 mole fractions and correlated activity coefficients γ i,and excess Gibbs function, GE
here xi and yi are the liquid and vapour phase composi-ions, γ i are the activity coefficients, P is the total pressure,oi are the vapour-pressures of the pure compounds, Bii are
he second virial coefficients, Bij is the cross second virialoefficient and V o
i are the molar volumes of the saturatediquids.
The second virial coefficients were taken from TRC tables21] for tetrahydrofuran or calculated using the PRSV-EoS22,23], the cross second virial coefficients were estimatedsing a suitable mixing rule and the molar volumes of sat-rated pure liquids were obtained from our own densityeasurements. The values of all these properties are given inable 1.
The adjustable parameters, A12 and A21 of the equations alongith the average deviation in P, �P, and the average deviation
n y, �y, are listed in Table 4. The correlated parameters doot show any temperature dependence. As one can see in this
B. Giner et al. / Fluid Phase Equilibria 251 (2007) 8–16 13
able all the equations correlated the activity coefficients quiteell.The activity coefficients together with the corresponding
xcess Gibbs function calculated using the Wilson equation areathered in Table 2, the results obtained with the rest of the equa-ions are very similar. The calculated excess Gibbs functions at= 298.15 K are plotted in Fig. 5.The excess Gibbs functions for the mixtures containing
etrahydrofuran and tetrahydropyran are negative over the wholeomposition range being the GE values for the mixture withetrahydrofuran appreciably higher in absolute value. The tem-erature behaviour of excess Gibbs functions is quite complex.or tetrahydrofuran the minimum GE values are shown at= 298.15 K and at the temperatures of 313.15 and 328.5 K
he excess Gibbs functions decreases in absolute value and areearly the same while for tetrahydropyran the minimum GE val-es are reached at 328.15 K and the lowest ones in absolute valuere shown at T = 313.15 K.
The behaviour of the mixtures containing 1,3-dioxolane and,4-dioxane are very similar showing positive excess Gibbsunctions, although for the mixture with 1,3-dioxolane the GE
alues are slightly higher. For both cyclic diethers the GE valuest the temperatures 298.15 and 313.15 K are nearly equal while
t T = 328.15 K the excess Gibbs function decrease.
It can be pointed out that the excess Gibbs functions strongly influenced by the structure of cyclic ether, forhe cyclic monoethers GE is negative while for the cyclic
iethers GE is positive. The excess Gibbs functions observedan be explained taking into account three factors: (i)
positive contribution coming from the weakness of thehloroalkane–chloroalkane interactions, this contribution isimilar for all the systems; (ii) another positive contribution dueo the disruption of the ether–ether interactions, these interac-ions in the pure ethers become progressively stronger in theequence [11,24–26]: tetrahydropyran < tetrahydrofuran < 1,4-ioxane < 1,3-dioxolane, so the magnitude of the positiveontribution increases following the same sequence. That is,he GE values for the mixtures containing 1,4-dioxane and 1,3-ioxolane must be higher than those for the mixtures containingetrahydropyran and tetrahydrofuran as the results show; (iii)nally, a negative contribution due to the Cl–O specific inter-ction, the heteroassociation has major importance for cycliconoethers than for cyclic diethers [26,27], and between the
yclic monoethers it is well known that the donor ability ofhe oxygen atom to interact with the chlorine atom that acts ascceptor is bigger for tetrahydrofuran than for tetrahydropyran16,28,29]. This negative contribution for the mixtures con-aining tetrahydrofuran and tetrahydropyran is large enough toxceed the above mentioned positive contributions and lead toegative excess Gibbs functions, being the absolute GE val-
es for tetrahydropyran lower than for tetrahydrofuran. Thexistence of this specific interaction is also supported by theegative HE values showed for the mixture tetrahydrofuran + 1-hloropentane [30].
14 B. Giner et al. / Fluid Phase Equilibria 251 (2007) 8–16
Table 4Correlation parameters and average deviation in pressure, �T, and average deviation in vapour phase composition, �y
ist of symbols12, A21 adjustable parameters for VLE correlation equations
(J mol−1)ii second virial coefficient of component i (m3 mol−1)ij cross second virial coefficient cm3 mol−1)ij − gii parameters for NRTL equation (J mol−1)E excess Gibbs function (J mol−1)oi vapour pressure of component i (Pa)
total pressure (Pa)molar gas constant (= 8.31447 J mol−1 K−1)temperature (K)
ij − uii parameters for UNIQUAC equation (J mol−1)oi molar volume of component i (m3 mol−1)i mole fraction of component i in the liquid phasei mole fraction of component i in the vapour phase
reek letters12 nonrandomness parameter in the NRTL equationi activity coefficient of component i
average deviationij − λii parameters for Wilson equation (J mol−1)
Authors thank for financial assistance from Diputacioneneral de Aragon. B. Giner wishes to thank Ministerio de Edu-
acion y Ciencia for the F.P.I. grant and A. Villares wishes tohank Ministerio de Educacion y Ciencia for the F.P.U. grant.
eferences
[1] A. Reyes, M. Haro, I. Gascon, H. Artigas, C. Lafuente, J. Chem. Eng. Data48 (2003) 887–891.
[2] B. Giner, A. Reyes, M. Haro, S. Martın, F.M. Royo, Phys. Chem. Liq. 42(2004) 173–183.
[3] B. Giner, H. Artigas, I. Gascon, P. Cea, F.M. Royo, Phys. Chem. Liq. 43(2005) 299–307.
[4] B. Giner, M.C. Lopez, P. Cea, C. Lafuente, F.M. Royo, Fluid Phase Equi-libria 232 (2005) 50–56.
[5] B. Giner, S. Martın, M. Haro, H. Artigas, C. Lafuente, Ind. Eng. Chem.Res. 44 (2005) 6981–6988.
[6] G.M. Wilson, J. Am. Chem. Soc. 86 (1964) 127–130.[7] H. Renon, J.M. Prausnitz, AIChE J. 14 (1968) 135–144.[8] D.S. Abrams, J.M. Prausnitz, AIChE J. 21 (1975) 116–128.[9] O. Kiyohara, P.J. D’Arcy, G. Benson, Can. J. Chem. 57 (1979) 1006–1010.10] A. Inglese, J.P.E. Grolier, E. Wilhelm, J. Chem. Eng. Data 28 (1983)
124–127.11] P. Brocos, E. Calvo, A. Amigo, R. Bravo, M. Pintos, A.H. Roux, G. Roux-
Desgranges, J. Chem. Eng. Data 43 (1998) 112–116.12] J.A. Riddick, W.B. Bunger, T.K. Sakano, Organic Solvents. Physical Prop-
erties and Methods of Purification (Techniques of Chemistry, vol. II), 3rded., Wiley-Interscience, New York, 1986.
13] S. Cabani, G. Conti, L. Lepori, Trans. Faraday Soc. 67 (1971) 1943–1950.14] H.I. Paul, J. Krug, B. Gutsche, H. Knapp, J. Chem. Eng. Data 31 (1986)
448–456.
1 se Eq
[
[
[[
[[
[
[[[
[
[
6 B. Giner et al. / Fluid Pha
15] H. Artigas, C. Lafuente, M.C. Lopez, F.M. Royo, J.S. Urieta, J. Chem. Eng.Data 39 (1994) 729–732.
16] B. Giner, S. Martın, H. Artigas, M.C. Lopez, C. Lafuente, J. Phys. Chem.B 110 (2006) 17683–17690.
17] H.C. Van Ness, S.M. Byer, R.E. Gibbs, AIChE J. 19 (1973) 238–244.18] A. Fredenslund, J. Gmehling, P. Rasmussen, Vapor–Liquid Equilibria
Using UNIFAC, Elsevier, Amsterdam, 1977.19] J.A. Nelder, R.A. Mead, Comput. J. 7 (1965) 308–313.
20] N. Silverman, D.P. Tassios, Ind. Eng. Chem. Process Des. Dev. 23 (1984)
586–589.21] TRC Thermodynamic Tables-Non-Hydrocarbons, Selected Values of Prop-
erties of Chemical Compounds; Thermodynamic Research Center, TexasA&M University College Station, TX, 1973.
[[[[
uilibria 251 (2007) 8–16
22] D.Y. Peng, D.B. Robinson, Ind. Eng. Chem. Fund. 15 (1976) 59–64.23] R. Stryjek, J.H. Vera, Can. J. Chem. Eng. 64 (1986) 323–333.24] M.D. Guillen, C. Gutierrez Losa, J. Chem. Thermodyn. 10 (1978) 567–
576.25] I. Gascon, H. Artigas, S. Martın, P. Cea, C. Lafuente, J. Chem. Thermodyn.
34 (2002) 1351–1361.26] J.A. Gonzalez, I. Mozo, I. Garcıa de la Fuente, J.C. Cobos, V.A. Durov,
Fluid Phase Equilibria 245 (2006) 168–184.
27] T.M. Letcher, P.U. Govender, J. Chem. Eng. Data 40 (1995) 1097–1100.28] S. Searles, M. Tamres, J. Am. Chem. Soc. 73 (1951) 3704–3706.29] E.M. Arnett, C.Y. Wu, J. Am. Chem. Soc. 84 (1962) 1684–1688.30] J. Biros, A. Zivny, J. Pouchly, J. Chim. Phys. Phys.-Chim. Biol. 72 (1975)