(138d) Membrane-Assisted Downstream Processing for Biobutanol Purification

One of the main challenges for chemical
industry is the change from fossil resources to a sustainable production of
chemicals. The change to biobased raw materials and biotechnological processes
often leads to product inhibitions of microorganisms and diluted system,
resulting in an expensive downstream processing. One bulk chemical that can be
produced in a sustainable way is biobutanol. One method of producing biobutanol
utilizes the so-called acetone?n-butanol?ethanol (ABE) fermentation that
can be carried out on basis of e.g. cellulosic and lignocellulosic biomass or
agricultural wastes. However, the productivity of this fermentation is still
limited, because n-butanol is toxic towards the microorganisms used in
ABE fermentation. That is why the concentration of n-butanol in the
fermentation broth is often limited to values lower than 2 wt.-percent [1]. For
n-butanol recovery mainly distillation processes are used which - as result of
the low concentrations - have a high energy demand. About 80 % of the total
energy demand of the downstream process is required for the separation of
acetone, n-butanol and ethanol from the fermentation broth. Alternatively to
distillation the fermentation can be coupled with a pervaporation unit, thus
n-butanol can be separated from the broth continuously to prevent the n-butanol
concentration from reaching the toxicity limit. Because of the mild operating
conditions, pervaporation enables an energy-efficient n-butanol separation and
facilitates the recycling of the broth with lower n-butanol content back to the
fermenter.

Therefore, the aim of this work was to
evaluate if it is more efficient to separate n-butanol by means of pervaporation
or distillation or a combination of both. For this purpose extensive
experimental studies were performed to characterize the permeation behavior of
a commercial available poly(dimethylsiloxane) (PDMS) membrane (Sulzer Pervap^TM
4060). These studies provide a wide data basis for the downstream process analysis
and are used to identify possible operating windows for the application of a
pervaporation. An experimental comparison between the pervaporation of a binary
mixture of n-butanol and water, the pervaporation of a synthetic medium and the
pervaporation of real fermentation broth was carried out. The synthetic medium contained
acetone, n-butanol and ethanol in mass ratios of 3:6:1 together with acetic and
butyric acid, which are the solvent precursors. Additionally process parameters
like feed composition, pervaporation temperature and permeate pressure were
varied. The detailed experimental procedure is described elsewhere [2]. Next to
the PDMS membrane, composite membrane materials containing ionic liquids are
promising for separation of n-butanol from an aqueous solution. These membrane
materials show advantageous permeation properties compared to conventional
polymer membranes [2]. Therefore different supported ionic liquid membranes
(SILMs) have been tested to evaluate their potential in n-butanol
pervaporation.

Based on the experimental results a
correlation has been developed describing partial fluxes of acetone, n-butanol,
ethanol and water. Fluxes of acetic and butyric acid were neglected. With the
help of this correlation modular-based process studies have been carried out
using Aspen Custom Modeler^TM to find an optimal process
configuration for n-butanol separation from the fermentation broth. It was
found that pervaporation as stand-alone process as well as distillation is
comparatively expensive. Because n-butanol should be almost completely
separated from the broth, the concentration at the outlet of the membrane
module becomes small, leading to a large required membrane area and a small
permeate concentration of n-butanol.

In contrast to the application of
distillation and pervaporation as stand-alone processes, a combination of
pervaporation and distillation was found to be best suited for efficient
n-butanol recovery. A possible membrane-assisted process is shown in Fig. 1.  Pervaporation
is used for continuous n-butanol removal in order to extend the fermentation
cycles. Because the mass fraction of n-butanol w_B,Rec in the recycle
stream is only slightly smaller compared to the mass fraction of n-butanol in
the fermentation w_B,Ferm, the required membrane area is
comparatively small and a higher permeate concentration can be obtained. The
complete removal of n-butanol from the broth is then carried out in
distillation columns. In this case, the extension of the fermentation periods
and the reduction of the stream to be processed in the energy-intensive
distillation lead to a reduced energy demand and lowered downstream costs for
n-butanol. The permeate stream and the solvent phase from the distillation can
be processed further in a conventional distillation sequence. As existing
biobutanol production plants mainly rely on distillation sequences, retrofit of
a pervaporation thus enables an increase in the biobutanol production efficiency,
while existing paid-off plants can still be used.

Fig. 1: Solvent removal from fermentation
broth by distilliation (a), pervaporation (b) and a membrane-assisted process
(c).

Acknowledgement:

The research leading to these
results has received funding from the European Union Seventh Framework
Programme (FP7/2007-2013) under grant agreement n° 241718 EuroBioRef. We
acknowledge Sulzer Chemtech AG and the Laboratory of Biochemical Engineering,
TU Dortmund University for providing PDMS membranes and fermentation broth.

[1] T.C. Ezeji, N. Qureshi, H.P. Blaschek,
Bioproduction of butanol from biomass: from genes to bioreactors, Curr. Opin.
Biotechnol. 3 (2007) 220?227.

[2] S. Heitmann, J. Krings, P. Kreis, A.
Lennert, W.R. Pitner, A. Górak, M.M. Schulte, Recovery of n-butanol using ionic
liquid-based pervaporation membranes, Sep. Purif. Technol. 97 (2012), 108-114.