Pervaporation for Effective in-situ Butanol Recovery in ABE Fermentation

Promoted by the increase of industrialization and population the global demand for energy and material products is steadily rising. Since the primary sources for energy and chemicals are mainly based on fossil resources, this growth provokes significant issues in terms of environment, economy and society. Thus, the supply of renewable energy and sustainable chemicals became one of the major societal targets of the current century. Energy and chemicals from biogeneous resources are accepted amongst the most promising pathways for covering the growing demands both in a sustainable and economic way. Biobutanol can be treated as a possible key substance in these regards as it acts as an energy carrier or fuel (supplement for gasoline, diesel and kerosene superior to ethanol) and as a commodity chemical (base chemical for organic synthesis, solvent in chemical industry).

While Acetone-Butanol-Ethanol(ABE)-fermentation as the main production route of biobutanol is a well-established technology, experts agree that still remarkable optimization potential is given. Significant efforts in contemporary research are undertaken towards the selection of fermentation feedstock, development and optimisation of biomass-pretreatment and hydrolysis as well as metabolic engineering of the involved microorganism strains (clostridia) to enhance butanol yield and substrate utilisation. Persistent challenge is the limitation of ABE fermentation by high product toxicity of butanol resulting in reasonably dilute product concentrations (up to 20 g/l). Consequently, constant in-situ separation of butanol from the fermentation broth is mandatory to sustain stable production. Furthermore, butanol has to be concentrated from fermentation levels up to 99.9 wt% for further utilisation (downstream processing). Distillation is no option as the energy demand for separations at these low product concentratios would exceed the thermal energy content of the product by more than 200%. Thus, alternative technologies for purification or pre-enrichment like gas stripping, adsorption, extraction or membrane separation have to be developed and applied.

The scope of current work is the analysis and development of membrane-based pervaporation process for in-situ butanol recovery from ABE fermentation. Results from extensive lab-scale experiments shall be used for development of mathematical models for improved process description targeting for optimisation and upscaling of the pervaporation process.

Experimental analysis is performed using a laboratory pervaporation setup with a feed flowrate in the range of 100 to 200 l/h in a closed loop. Feed temperature is maintained by a thermostat and heat exchanger. Feed depletion from the feed tank is monitored with a balance. The setup can be equipped with flat-sheet membranes with an active area of 144 cm² or with commercial hollow-fiber membrane modules of appropriate size. Vacuum in the range of 10 to 20 mbar(a) is applied on the permeate side with a rotary vane vacuum pump providing the driving force for the process. Transmembrane or permeate flux is totally condensed in cold traps cooled with liquid nitrogen and weighted out at given timestamps. Compositions of feed and permeate are analyzed by GC.

This setup is used to analyze the pervaporation performance of different membrane materials and membrane modules under fermentation conditions in terms of specific transmembrane fluxes, butanol enrichment in the permeate as well as permeance and selectivities for ABE solvents and water. The influence of feed solvent concentration, residual glucose concentration, presence of salts and organic acids, feed temperature and pH value is studied. Regarding membrane materials, both commercially available as well as experimental materials like PDMS and POMS are considered.

Results indicate that the butanol enrichment factor in the permeate reaches values of around 20 in a single separation step (1 wt% BuOH in the feed, 20 wt% BuOH in the permeate) depending on the membrane material. This remarkable enrichment is mainly contributed to the activity coefficients, vapor pressures and membrane selectivity which prefers solvent over water transport due to its hydrophobic nature. Furthermore, the influence of additional components in synthetic fermentation broth is of minor importance in the investigated ranges. No interference on pervaporation performance has been detected for glucose contents of up to 50 g/l, ammonium chloride contents of up to 3 g/l, and organic acids contents of up to 4 g/l (acetic acid and propionic acid). Above all, no flux decline after experiments with synthetic fermentation broth has been detected from which follows that fouling phenomena did not occur. Experiments with real fermentation broth applying effective cell retention are scheduled in order to validate this conclusion also for the real process. Deterioration of membrane performance has been detected at very harsh pH values in the acidic range that need to be avoided in a real coupled fermentation-separation process. In conclusion, current work shows that membrane-based pervaporation represents a highly effective technique for in-situ recovery and pre-concentration of butanol produced in ABE fermentation.