Articulo mtbe

Computers chem. Engng Vol. 21, No. 10, pp. 1117- I 124, 1997

Pergamon
PIh S0098-1354(96)00322-5

© 1997 Elsevier Science Ltd. All rights reserved Printed in Great Britain

0098-1354/97$17.0o+0.00

Multiplicity in reactive distillation of MTBE
S. Hauan, T. Hertzberg and K. M. Lien*
Department of Chemical Engineering, The Norwegian Institute of Technology, The University of Trondheim,N-7034 Trondheim, Norway

Abstract
This paper presents an explanation of why methyl tert-butyl ether (MTBE) production by reactive distillation may yield multiple solutions. Widely different composition profiles and conversions may, as already reported by Krishna and others, result with identical column specifications, depending on the initial estimates provided and the VLE models being used. Ahypothesis yielding a qualitative understanding of this phenomenon has been developed. The inert n-butene plays a key role in the proposed explanation. As the reaction mixture is diluted with n-butene, the activity coefficient of methanol increases substantially and the temperature decreases. This dilution has a profound effect on the equilibrium conversion, enabling MTBE to escape from thereactive zone without decomposition. The minimum boiling point azeotrope between MTBE and methanol plays an important role in the internal transport mechanism for the heavy methanol, in particular when fed below or in the lower part of the reactive section. © i 997 Elsevier Science Ltd

Keywords: reactive distillation; multiplicity; MTBE-production List of symbols
Y,: mole fraction for component i,liquid phase X~: mole fraction for component i, vapour phase a~: activity for component i, liquid phase TJ :activity coefficient for component i F~ :formation rate for component i, mol/s solutions, considering several alternative hypotheses. After having discarded both crossing of non-reactive distillation boundaries and CSTR-Iike multiplicity behaviour as possible explanations, they concluded thatsimultaneous reaction and phase equilibrium could explain the observed behaviour. By tracing residue curve maps for simultaneous chemical and phase equilibria, they were able to identify two distinctly different families of curves, depending on the curves' origin in composition space. In order to reproduce Jacobs and Krishna's results, we have carried out steady state simulations using the samemodel and parameters: the model described by Venkataraman et al. (1990) and implemented in ASPEN Plus as the rigorous distillation module RADFRAC. The liquid phase activities have been calculated using UNIQUAC, with the binary interaction parameters reported by Rehfinger and Hoffman (1990). The ASPEN Plus property set Sysopl 1 defines the various thermodynamic models in use, Redlich-Kwong forequation of state calculations, and UNIQUAC for liquid activity and flash calculations. The polynomial given by Rehfinger and Hoffman (1990) has been used to represent the temperature dependence of the reaction equilibrium constant.

Introduction
Figure 1 shows a reactive distillation column that may be used for production of MTBE from methanol and isobutene. The 17 stages include a partial reboilerand a total condenser. There are three different zones in the column, each with a different process task. The rectification and stripping zones operate exactly as if in a nonreactive distillation column, purifying top and bottom products. Chemical conversion takes place in the middle, reactive section, and the liquid composition of stages 4-11 is assumed to be in chemical reaction equilibrium. Amixture of i-butene and inert n-butene is fed to the column on stage 11 while the methanol feed location is varied between stages 2 and 16. This column configuration has previously been analysed by Jacobs and Krishna, 1993 who were able to demonstrate that widely different converged solutions could be obtained for certain sets of column specification. They examined why they would get multiple *...