High gravity extractive fermentation for enhanced productivity of bioethanol : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Biochemical Engineering at Massey University, Palmerston North, New Zealand

Abstract

Bioethanol is a renewable alcohol fuel produced from sugary substrates via fermentation processes. Accumulation of ethanol in the fermentation broth inhibits cell growth and further production of ethanol. Recovering ethanol from a dilute broth is expensive. Ethanol inhibition of fermentation may be reduced by continuously removing it as it is formed. This work focussed on production of bioethanol from glucose using the anaerobic bacterium Zymomonas anaerobia. This bacterium produces ethanol more rapidly than does the conventional yeast fermentation. The aim was to assess the impact of continuous in-situ removal of ethanol on the productivity of batch and continuous high-gravity fermentations. High-gravity fermentations use a medium with a high concentration of sugar to reduce production volume (bioreactor size) and potentially achieve a high productivity of ethanol, if the ethanol concentration in the broth can be kept to below inhibitory levels. First, the batch fermentation was characterized for glucose tolerance, ethanol tolerance, optimal production temperature and biocompatibility with solvents and adsorbents that could be used for in-situ removal of ethanol. High gravity media containing 50–300 g L-1 glucose were used to characterize batch and continuous fermentations with a view to identifying the best fermentation conditions for a detailed study. The optimal fermentation temperature was found to be 35 [deg]C and the maximum tolerable initial glucose (i.e. without causing substrate inhibition) was 150 g L−1. In continuous high-gravity fermentations, six different dilution rates (D = 0.05– 0.30 h-1) were tested, but steady-state operation proved to be impossible at the lowest dilution rate: the fermentation showed a highly consistent oscillatory behaviour that was ascribed to ethanol toxicity. Use of higher dilution rates could overcome oscillations by washing out the ethanol from the bioreactor, but this reduced ethanol productivity as glucose and biomass also washed out. Strategies for removing ethanol in-situ while operating at such a dilution rate as to achieve a high ethanol productivity, were assessed by using liquid-liquid extraction and adsorption on polymer resins as methods for removing ethanol as it was being produced. High gravity extractive fermentation for enhanced productivity of bioethanol In the absence of in-situ removal of ethanol, the lowest operable steady-state dilution rate (i.e. without oscillations) was 0.15 h-1. With in-situ removal of ethanol, the dilution rate for stable steady state operation could be reduced to 0.05 h-1. At a dilution rate of 0.15 h-1, the steady-state ethanol concentration was 42.5 g L-1 and the biomass concentration was 1.49 g L-1. In the absence of in-situ ethanol removal, the ethanol concentration, but not ethanol productivity, was highest at a dilution rate of 0.3 h-1 although much residual glucose remained. In in-situ batch extractive fermentations, all extraction solvents tested improved biomass concentration, glucose consumption and ethanol concentration relative to control, but iso-octadecanol was clearly the most effective solvent. For batch in-situ extractive fermentation with iso-octadecanol, the ethanol yield on glucose was 0.485±0.005 g g-1, or comparable to to a yield of 0.468±0.005 g g-1 for the control culture, but the ethanol productivity was distinctly higher than for the control culture. Of the various polymer resins tested in batch fermentations for in-situ removal of ethanol by adsorption, Dowex Optipore L-493 appeared to be somewhat better than the control (i.e. no resin). The best extraction solvent (i.e. iso-octadecanol) and the best adsorption resin (i.e. Dowex Optipore L-493) were separately assessed for ethanol removal in continuous fermentations. Continuous removal of ethanol both by adsorption and solvent extraction allowed a steady-state operation of the continuous fermentations at a dilution rate of 0.05 h−1 — the dilution rate at which steady-state operation had proved impossible in control fermentation (i.e. without in-situ removal of ethanol). This confirmed the mechanism used to explain the oscillatory behaviour of the fermentation and showed that in-situ ethanol removal permitted steady-state operation at dilution rates that would not allow such operation in the absence of ethanol removal. At a dilution rate of 0.05 h−1, an extraction solvent flow rate of 300 mL h−1 provided the highest total ethanol productivity and ethanol yield on glucose while keeping the solvent use to a minimum.