Analysis and multiple steady states of an industrial heterogeneous azeotropic distillation

Citation
Ma. Gaubert et al., Analysis and multiple steady states of an industrial heterogeneous azeotropic distillation, IND ENG RES, 40(13), 2001, pp. 2914-2924
Citations number
27
Categorie Soggetti
Chemical Engineering
Journal title
INDUSTRIAL & ENGINEERING CHEMISTRY RESEARCH
ISSN journal
08885885 → ACNP
Volume
40
Issue
13
Year of publication
2001
Pages
2914 - 2924
Database
ISI
SICI code
0888-5885(20010627)40:13<2914:AAMSSO>2.0.ZU;2-Q
Abstract
A study of an industrial multicomponent heterogeneous azeotropic distillati on is presented. The process concerns an organic acid dehydration using an immiscible entrainer. First, a validation of the MESH and thermodynamic mod els through a comparison between pilot-plant experimental reconciled data a nd simulation results is conducted. A four-component mixture is considered for the simulation. Case studies of the boiler heat duty are automatically generated by an operating path tool. An infinity/infinity analysis is perfo rmed for the heterogeneous azeotropic pilot column and an industrial column with a decanter. Because of practical constraints, the pilot and the indus trial columns do not have the same reflux policies. This leads the infinity /infinity analysis to predict multiple steady states for the industrial uni t but not for the pilot column. However, multiple steady states are found b y simulation both for the pilot and for the industrial unit. Multiple stead y states are confirmed by simulation and experimental data for the industri al unit. Because of the positive infinity/infinity analysis, they are attri buted to the phase equilibrium properties of the quaternary system. For the pilot column, multiple steady states are found by the simulation and linke d to experimental observations. The multiplicity is not caused by the phase equilibrium properties; rather, it is attributed to interactions between t he material and energy balances. An analysis of the simulation results help s explain the behavior of the industrial unit: the temperature of the sensi tive tray gives rise to a peak in heat. This peak is located very close to the industrial temperature set point and is correlated with an impurity con tent minimum in the main product stream. An impurity minimum is also eviden ced by the simulation for the pilot column. This complex behavior can expla in observed difficulties in controlling the process at the industrial set p oint.