Elsevier

Fluid Phase Equilibria

Volume 309, Issue 1, 15 October 2011, Pages 15-19
Fluid Phase Equilibria

“Vapor–liquid” equilibrium measurements and modeling for the cyclohexane + n-hexanoic acid binary system

https://doi.org/10.1016/j.fluid.2011.06.027Get rights and content

Abstract

To simulate cyclohexane to cyclohexanol oxidation reactors, the acquisition and modeling of vapor–liquid equilibria of the key components, under the process conditions, are essential. n-Hexanoic acid is a co-product of the reaction. Vapor–liquid equilibrium data are reported for the cyclohexane + n-hexanoic acid binary system at four temperatures: 413, 423, 464 and 484 K. All measurements have been carried out using an apparatus based on the “static-analytic” method, with two ROLSI™ pneumatic capillary samplers. The generated data are successfully correlated using two equations of state, the Peng–Robinson (PR) and the Perturbed Chain Statistical Association Fluid Theory (PC-SAFT). Both models are capable of representing the experimental data, but the PC-SAFT EoS uses less binary interaction parameters.

Highlights

• We determine new experimental data concerning the cyclohexane + n-hexanoic acid binary system. • An equipment based on “static analytic” method with phase sampling is used. • Two models are compared for the data treatment. • Data and models are used for the design of phase separation units on the Rhodia site.

Introduction

Selective oxidation of hydrocarbons to give oxygen-containing compounds is an extremely important and useful reaction in the chemical industry [1]. Normally, more than one oxygenated product is formed from a given starting material. The oxidation of cyclohexane is important for the commercial production of nylon. Cyclohexanol and cyclohexanone are produced by oxidizing cyclohexane with air.

n-Hexanoic acid (caproic acid) is a by-product obtained from the oxidation of cyclohexane (primary reaction). The selectivity of the primary reaction is strongly downgraded by the formation of the carboxylic acid, which is present as a secondary reaction. In order to control the secondary reaction, it is necessary to control the two main parameters of the primary reaction, i.e. temperature and pressure. The combined knowledge of the cyclohexane + n-hexanoic acid vapor–liquid equilibria (VLE), and reaction kinetics of the secondary reaction, allows optimization of the reaction conditions for the oxidation of cyclohexane.

In this study, vapor–liquid equilibrium (VLE) data for the cyclohexane + n-hexanoic acid binary system are presented at 423, 443, 464 and 484 K. Presently, there are no data available in literature concerning this binary system. In this presented work, we have tested modeling of our new measured data using cubic and non-cubic equations of state (EoS). The former is based on the Peng–Robinson (PR) cubic equation [2], using the Wong–Sandler (WS) mixing rule [3], and the NRTL Gibbs free energy model [4]. The latter model is the Perturbed-Chain modification of the SAFT equation [5], [6], we have enhanced for this application through an additional dipolar contribution proposed by Gross and Vrabec [7]. For brevity, we refer to this EoS as PCP-SAFT.

Section snippets

Materials

Cyclohexane (C6H12, CAS No.: 110-82-7) and n-hexanoic acid (C6H12O2, CAS No.: 142-62-1) were purchased from Fluka and Acros, respectively. Their GC certified purities are at least 99.8% and 99%, respectively. Both compounds were used after careful degassing.

Apparatus

The apparatus used in this work is based on a “static-analytic” method with liquid and vapor phase samplings. This apparatus and the experimental procedure are identical to those described by Valtz and coworkers [8], [9]. In brief, the

Modeling approach

For this work, a symmetric (ψ  ψ) approach is followed to calculate both liquid and vapor phase fugacities.

Results and discussion

The experimental isothermal Pxy measurements at 413.46, 423.48, 463.63 and 483.69 K are reported in Table 2, and presented in Fig. 1.

In Fig. 1, each isotherm is represented via their corresponding BIPs for each model, with the parameters given in Table 3. Using PCP-SAFT EoS, we have obtained k12 = 0.0091. The deviations resulting from the BIP adjustments are listed in Table 4.

The quality of fit between the two EoS is similar, with the PR-WS-NRTL EoS being slightly better than the PCP-SAFT. This

Conclusions

In the context of reactor simulation of cyclohexane oxidation, we have measured vapor–liquid equilibrium data for the cyclohexane + n-hexanoic acid binary system at four temperatures, using a “static-analytic” method. The experimental results are of high accuracy, with the following uncertainties: ±0.03 K and ±0.001 MPa for temperatures and pressures, respectively, and a maximum of 0.007 for vapor and liquid mole fractions. The data are very well represented by the PR-WS-NRTL EoS, albeit not

Acknowledgments

The authors would like to thank Prof. Joachim Gross for useful discussions regarding the general modeling approach and the bonding scheme presented in this work.

References (15)

  • C. Coquelet et al.

    Fluid Phase Equilib.

    (2010)
  • P.M. Mathias et al.

    Fluid Phase Equilib.

    (1983)
  • P. Coutsikos et al.

    Fluid Phase Equilibr.

    (1995)
  • A. Kumar et al.

    Trans. Met. Chem.

    (2003)
  • D.Y. Peng et al.

    Ind. Eng. Chem. Fundam.

    (1976)
  • D.S.H. Wong et al.

    AIChE J.

    (1992)
  • H. Renon et al.

    AIChE J.

    (1968)
There are more references available in the full text version of this article.

Cited by (2)

View full text