Pilot plant synthesis of n-propyl propionate via reactive distillation with
decanter separator for reactant recovery. Experimental model validation and
simulation studies
ALTMAN Ernestoa, KREIS Peterb, VAN GERVEN Toma, STANKIEWICZ Andrzeja, GORAK
Andrzejb
a
Process and Energy Department, Delft University of Technology, Leeghwaterstraat 44, 2628
CA Delft, The Netherlands
b
Department of Biochemical and Chemical Engineering, TU Dortmund University,
Emil-Figge-Str. 70, 44227 Dortmund, Germany
e.altmanrestrepo@tudelft.nl
Abstract
In today’s high demanding energy world, the study and advancement in chemical processing
operations is a challenge for both academia and industry. The future of many chemical operations
depends to a large extent on the innovation capacity of engineers in order to reduce energy
consumption, emissions, waste and risk. Process intensification (PI) is a good example of how
innovation is driving development in chemical engineering. A successful business story of PI
applied broadly is the reactive distillation (RD) process used in esterification syntheses. One
exemplifying application is illustrated in this research with the heterogeneously catalyzed
reaction of n-propyl propionate (ProPro) from 1-propanol (ProOH) and propionic acid (ProAc).
Conventional RD of ProPro was further improved. ProPro is the main reaction product and is
removed in the bottom stream of the column, together with unreacted ProAc and ProOH. In order
to have the maximum possible conversion of acid, an excess of alcohol is fed into the column. As
a result the distillate is a mixture containing mostly water, unreacted ProOH and a small fraction
of ProPro. With the objective to recover the product and excess reactant, the column set-up was
equipped with a decanter on top enabling to separate the distillate into two main streams (aqueous
and organic) without the necessity to use entrainers. The aqueous phase was discharged and part
of the organic phase (rich in ProPro and ProOH) was refluxed back to the column.
Experimental results including column profiles (temp + conc) obtained in a pilot-scale column
(DN-50), equipped with two types of structured packings (Sulzer BX for stripping and rectifying
sections, and Katapak-SP 11 with the ion exchange resin Amberlyst 46TM immobilized for the
reactive part), showed a maximum ProPro weight fraction purity of xProPro,bottom = 68.5% and
maximum ProAc conversions of 81.6% at the operating conditions investigated. In addition, a
non-equilibrium stage model (NEQ model) for the column including a model for liquid-liquid
phase equilibrium prediction for the decanter was implemented at the TU Dortmund University in
the simulation environment ASPEN Custom Modeler (ACM). The use of accurate experimental
data was of paramount importance for the model validation. Special attention was given to steady
state process verification tests and data reconciliation during experiments. Model predictions are
compared to experimental results showing good accuracy. Theoretical investigations of the most
important operating parameters (total feed, molar feed ratio, reflux ratio and heat duty) and their
effect on the overall process performance are presented.