Process analysis of microwave assisted reactive distillation

Abstract

A promising approach to intensify reactive distillation processes is the application of microwave fields. Recent research results indicate that microwave radiation might affect the thermal separation of molecules and accelerate chemical reactions. To determine the potential of microwave assisted reactive distillation in a first instance the separation of binary mixtures with microwave and conventional heating is investigated experimentally. The binary mixtures consist of ethanol which is a good microwave absorber, and a carbonate which is relatively transparent to microwaves-thus aiming at the preferred evaporation of ethanol. The experimental results indicate that there is no evidence for the influence of the applied microwave field on the thermal separation of these binary mixtures. The composition of distillate and bottom does not differ for conventional and microwave heating at macroscopic scale. Thus no enhancement of separation efficiency in distillation could be observed comparing the conventional and microwave distillation experiments. The potential impact of microwave radiation on the reaction is studied theoretically considering selective superheating of the catalyst by microwaves. The simulation results of the reactive distillation process indicate that no significant enhancement of conversion is possible. Consequently the performance of the reactive distillation process could not be improved by microwaves, neither on separation nor reaction level.

Full text

Process analysis of microwave assisted reactive distillation
Kathrin Werth1, Anton A. Kiss2, Giorgos Stefanidis3, Philip Lutze1
1Laboratory of Fluid Separations, TU Dortmund University, Dortmund, Germany;
2AkzoNobel Research, Development & Innovation, Deventer, Netherlands;
3Laboratory of Intensified Reaction & Separation Systems, Delft University of
Technology, Delft, Netherlands
Abstract
A promising approach to intensify reactive distillation processes is
the application of microwave fields. Recent research results indicate
that microwave radiation might affect the thermal separation of
molecules and accelerate chemical reactions. To determine the
potential of microwave assisted reactive distillation in a first instance
the separation of binary mixtures with microwave and conventional
heating is investigated experimentally. The binary mixtures consist of
ethanol which is a good microwave absorber, and a carbonate which
is relatively transparent to microwaves – thus aiming at the preferred
evaporation of ethanol. The experimental results indicate that there is
no evidence for the influence of the applied microwave field on the
thermal separation of these binary mixtures. The composition of
distillate and bottom does not differ for conventional and microwave
heating at macroscopic scale. Thus no enhancement of separation
efficiency in distillation could be observed comparing the
conventional and microwave distillation experiments. The potential
impact of microwave radiation on the reaction is studied theoretically
considering selective superheating of the catalyst by microwaves.
The simulation results of the reactive distillation process indicate that
no significant enhancement of conversion is possible. Consequently
the performance of the reactive distillation process could not be
improved by microwaves, neither on separation nor reaction level.
Keywords
Distillation, microwave assisted processes, reactive distillation
1. Introduction
Reactive distillation (RD) integrates reaction and separation into one apparatus this
leading to increased conversion and selectivity by overcoming equilibrium limitations.
However, its application is restricted by a common operating window between
reaction and separation, while the achieved purities may still be limited by the
occurrence of azeotropes. In chemical synthesis it is known that microwaves (MW)
have the potential to enhance reaction rates [1]. Furthermore, recent research results
indicate that MW radiation may affect the thermal separation of mixtures by e.g.
increasing the concentration of the low boiling components in vapor [2], and may
even influence the position of an azeotrope in terms of temperature and composition
[3]. Therefore, a promising approach to intensify RD and widen its application window
is the integration of MW fields into RD columns.
To prove the feasibility and the potential of the concept of MW assisted RD a
systematic investigation of the impact of MW radiation on the thermal separation of
691

mixtures, and on the reaction is necessary. For that reason the transesterification of
dimethyl carbonate with ethanol to produce diethyl carbonate is chosen as case-
study. This reaction has been successfully realised in a RD column already [4]. Since
the effects and mechanisms of MW heating are still not understood in the first step
both separation and reaction are analysed separately. In the following an
experimental investigation of the influence of MW radiation on the VLE and distillation
of binary mixtures is presented. Subsequently a theoretical study of the MW assisted
RD process based on current experimental data is described briefly.
2. Materials and methods
It is shown in literature that the boiling behavior of pure components can be affected
in an electromagnetic field due to a number of reasons, e.g. superheating [5] or
surface hydrodynamic instabilities [6]. The change in boiling behavior under MW
radiation might also influence the evaporation of mixtures. Therefore, we wanted to
study whether the thermal separation of binary mixtures can be favorably disturbed
(i.e. enhancement of separation efficiency) in the presence of a MW field.
In literature it is claimed that one possible reason for the effect of MWs on the
thermal separation of mixtures might be selective heating of components in MW
fields caused by their different dielectric properties. Components which are capable
to convert electromagnetic energy into heat very easily (e.g. polar components with a
high dielectric loss) might be evaporated preferentially [3]. However, components
which are not affected strongly by the MW field - such as nonpolar components with
low dielectric loss - may remain mainly in the liquid phase. The dielectric loss
describes the amount of input MW energy which is lost to the sample by heat
dissipation [7]. To identify the effect of MW radiation on the evaporation behavior two
different binary mixtures were chosen, each consisting of one component which is
strongly affected by MWs (alcohol) and a second component which is relatively
transparent to MWs (carbonate). Because of its high ability to store and convert
electromagnetic energy into thermal energy ethanol (EtOH) was selected as the
component which is affected by MWs very well. Dimethyl carbonate (DMC) and
diethyl carbonate (DEC) represent components which would not be affected that
much because a low dielectric loss of the components can be assumed. All of these
components are present in the transesterification reaction of dimethyl carbonate with
ethanol which is a reaction conducted in RD columns [4]. The system of DEC and
ethanol shows ideal thermodynamic phase behavior with ethanol as low boiling
component. In the DMC/ethanol systems a minimum azeotrope is present.
The experiments were performed in a small lab scale distillation setups. To evaluate
the results experiments were conducted with both MW heating and conventional
heating. The distillation setup (Figure 1) consists of a glass flask which was placed in
the reboiler and was connected via glass tube to a distillation head with total
condenser. In the MW experiments a Discover CEM single-mode microwave cavity
was used as reboiler while the conventional heating experiments were performed
with a heating mantle. The temperature is measured near the vapor liquid interface
with fiber optic sensors which are appropriate for applications in MW fields.
Binary mixtures of different compositions were filled into the glass flask and the liquid
was stirred continuously to guaranty a good temperature distribution. All experiments
were run under total reflux conditions at atmospheric pressure. For the MW heating
experiments a constant MW power was applied. MW power between 50 and 100W
were chosen depending on the composition of the feed mixture to control the
692

evaporation rate and minimize the superheating of the liquid. When steady state was
reached samples of distillate and bottom were taken. The distillate sample is taken at
the top of the column from the condensed vapor and the bottom sample is taken
directly from the reboiler. The composition of both samples was determined via gas
chromatography.
Figure 1: Lab scale distillation setup for microwave experiments, flowing conditions indicated.
3. Experimental results
Following the results of the conventional and microwave assisted distillation
experiments for the binary systems (DMC/ethanol and DEC/ethanol) are presented.
Figure 2 shows the results for the distillation of DMC and ethanol for some selected
experiments. The composition of distillate and bottom is displayed as molar fraction
of ethanol at the top of the column or respectively in the reboiler similar to a common
concentration profile in distillation columns.
Figure 2: Molar fraction of ethanol in distillate and bottom for different feed composition of
ethanol and DMC, each for conventional and MW assisted distillation experiment.
693

For a given bottom composition the molar fraction of ethanol in distillate is similar for
conventional and MW assisted distillation experiments. Considering the uncertainties
of measurements the slight difference in compositions cannot be regarded as a MW
effect. One key issue is the temperature measurement because the measured values
strongly depend on the position of the fiber optic sensor in the liquid. Thus no similar
temperature in the reboiler can be ensured in the conventional and MW experiments
(deviation around 1-2°C). It should be noted that in the setup also the temperature
profile in the column cannot be measured. Furthermore, the analysis error of the gas
chromatography (GC) should be taken into account (1-2.2%). There are also some
unknown parameters which should be considered here. Since the absorbed MW
power cannot be measured in the CEM microwave cavity the energy input and thus
the reboiler heat duty is unknown. In addition, no information is available about the
MW field – such as uniformity, stationarity or geometry.
An overview of all experimental results comparing the distillation under conventional
and MW heating is provided in Figure 3. The molar fraction of ethanol in the distillate
at the top of the column is plotted versus the corresponding molar fraction of ethanol
in the bottom. One should keep in mind that in this set of experiments no vapor liquid
equilibrium (VLE) was measured but a multistage distillation was considered.
The molar fraction of ethanol in distillate and bottom are in good agreement for
conventional and MW assisted distillation. The separation behavior of both
investigated binary mixtures in the distillation column setup does not change
significantly for conventional and MW heating. If MW radiation has a strong influence
on the evaporation in the bottom a change of vapour composition at the top of the
column would be expected.
Figure 3: Molar fraction of ethanol in distillate and bottom for conventional and MW assisted
distillation experiments for the binary system DMC/ethanol (left) and DEC/ethanol (right).
694

4. Discussion
MW assisted RD aims at a more efficient separation and accelerated chemical
reaction leading to an improved process performance. Following the influence of
MWs on the separation and reaction is discussed.
4.1 Influence of microwaves on the separation
In the conducted experiments the targeted impact that MWs enhance the separation
efficiency in thermal separation is not observed on the macroscopic distillation scale.
Also if there might be a small effect on the equilibrium composition at the vapor liquid
interface when a MW field is applied the effect is dominated by the mass transfer in
the still and in the column. However, it cannot be excluded that MW radiation has an
effect on the separation when using MWs at micro scale, directly at vapor liquid
interface.
Furthermore, the hypothesis that MWs might shift the azeotrope as proposed in
literature [3] was investigated for the non-ideal system of DMC and ethanol. The
results indicate that an influence of MW radiation on the VLE cannot be verified
experimentally because otherwise the azeotrope would have been shifted, which is
not the case here. However, no influence on the position of the azeotrope in terms of
temperature and composition could be observed. In none of the distillation
experiments an enrichment of distillate above the azeotropic composition was
possible. But it should be noted that the dielectric properties which are changing with
composition and temperature are unknown for the DMC/ethanol system. It might be
possible that there are more appropriate systems for which the effect of selective
heating (strongly depending on the dielectric properties of the mixture) may be more
pronounced. The measurement of the dielectric properties would be necessary in
order to get further insight.
4.2 Influence of microwaves on the reactive distillation process
Although no effect of an applied MW field in the separation was found, MWs may still
have an impact on RD processes. Appling a MW field in the reactive section of the
RD column could lead to local heating (e.g. hot spots on the solid catalyst) and thus
enhanced reaction rates and higher conversion. To determine the potential of MW
assisted RD processes an adapted model was developed based on a detailed model
of conventional RD [8]. The adapted reaction model integrates a selective
superheating of the catalyst into the Arrhenius approach.
Conventionally the transesterification of DMC with ethanol is catalyzed homo-
geneously by sodium ethoxide because the reaction rates of common heterogeneous
catalysts are too low [9]. In homogeneously catalyzed RD no selective heating of
catalyst would be possible due to good heat transfer in liquid phase. Therefore,
reaction temperature and thus reaction rate is limited by the boiling temperature of
the system. Even though MW heating might lead to elevated boiling temperatures, for
this system around 10K higher [5], this is not enough to remarkably improve column
performance since the reaction rate is already fast under conventional conditions.
But the local heating of solid catalysts might be an option to enable a
heterogeneously catalyzed reaction which would allow a simplified catalyst recycling.
In the simulation studies a MW field was integrated in the reactive section of the RD
column to generate an additional power input and simulate a local heating of the
catalyst. The results showed that very high superheating (ΔT>20K) would be
necessary to enhance reaction rates significantly. However, conversion is still very
low. Furthermore, the realization of such high superheating in the column would not
695

be possible so heterogeneous catalysts would not be appropriate for this MW
assisted RD process.
5. Conclusions
In this study the feasibility of MW assisted RD was evaluated. First the influence of
MW radiation on the separation of binary mixtures was investigated experimentally,
by comparing distillation using conventional and MW heating. Then the influence of
MWs on the chemical reaction was determined theoretically.
The distillation with conventional and MW heating showed comparable results for the
compositions of distillate and bottom for a given feed mixture. In general an influence
on the evaporation behavior by applying a MW field could not be observed. Taking
the uncertainties of measurement and the unknown parameter into account the
deviation in distillate and bottom composition for the conventional and MW
experiment is negligible and no microwave effect could be assumed. In the
theoretical study of the MW assisted RD it was investigated if enhanced reaction
rates and thus higher conversion could be reached by applying a MW field in the
reactive section of the column. But no remarkable effect of MWs on the performance
of the RD process could be found.
Hence, the current results indicate that for the investigated chemical system the
integration of a MW field into the RD column will neither enhance the separation
efficiency at macroscopic scale as present in a RD column nor enhance the reaction
rate significantly.
Acknowledgements
The research leading to these results has received funding from the European
Community‘s Seventh Framework Programme under grant agreement no. FP7-NMP-
2012-309874.
References
[1] P. Lidström et al., Microwave assisted organic synthesis – a review, Tetrahedron 57 (2001) 9225-
9283.
[2] E. Altman et al., Process Intensification of Reactive Distillation for the Synthesis of n-Propyl
Propionate: The Effects of Microwave Radiation on Molecular Separation and Esterification
Reaction, Ind. Eng. Chem. Res. 49 (2010) 10287-10296.
[3] X. Gao et al., Influence of a microwave irradiation field on vapor–liquid equilibrium, Chem. Eng.
Sci. 90 (2013) 213-220.
[4] T. Keller et al., Transesterification of dimethyl carbonate with ethanol in a pilot-scale reactive
distillation column, Chem. Eng. J. 180 (2012) 309-322.
[5] F. Chemat et al., Microwave super-heated boiling of organic liquids: Origin, effect and application,
Chem. Eng. Technol. 24 (2001) 735-744.
[6] D. Stuerga et al., Microwave heating as a tool for coupling Marangoni and Hickman instabilities,
Springer, 1996.
[7] C.O. Kappe et al., Practical Microwave Synthesis for Organic Chemists, Wiley-VCH, 2009.
[8] T. Keller and A. Górak, Modelling of homogeneously catalysed reactive distillation processes in
packed columns: Experimental model validation, Comput. Chem. Eng. 48 (2013) 74-88.
[9] T. Keller et al., Transesterification of Dimethyl Carbonate with Ethanol To Form Ethyl Methyl
Carbonate and Diethyl Carbonate: A Comprehensive Study on Chemical Equilibrium and
Reaction Kinetics, Ind. Eng. Chem. Res. 50 (2011) 11073-11086.
696