A study of butanol production in a batch oscillatory baffled bioreactor

Abstract

As with many bioprocesses, the acetone-butanol-ethanol (ABE) fermentation faces a number of economic drawbacks when compared to the petrochemical route for butanol production. In the 1920s biobutanol was the second largest biotechnology industry, after bioethanol production. However it became difficult to compete against the petrochemical route for reasons including the low product butanol concentration, because of product inhibition resulted in low butanol productivity and due to slow fermentation and low ABE yields. These lead to uneconomical butanol recovery by the conventional method, distillation, due to the high degree of dilution. Recent interest in biobutanol as a biofuel has led to re-examination of ABE fermentation with the aim of improving solvent yield, volumetric productivity and final solvent concentration to reduce the cost of production and thereby produce biobutanol that is cost-competitive with the chemical synthesis butanol. ABE fermentations were carried out in an intensified plug flow reactor known as the batch oscillatory baffled bioreactor (BOBB). The “BOBB”s were designed and built for this project. The effect of oscillatory flow mixing on ABE fermentation was compared to that of conventional stirred tank reactors (STRs) at power densities in the range 0 to 1.14 Wm⁻³. The maximum butanol concentration in this range in a BOBB was 34% higher than the STR. Some increase in butanol productivity was also observed: 0.13 gL⁻¹h⁻¹ in BOBBs, compared to 0.11 gL⁻¹h⁻¹ in the STRs. It can be concluded that at similar power densities, BOBB fermentation shifts to solventogenesis earlier than in STRs, resulting in higher solvent productivity. It is hypothesised that the reason for early solventogenesis in the BOBB was the higher solvent-producing cell concentration, due to the more uniform shear field in the BOBB, so the cell would be less exposed to high shear thereby reducing the risk of cell lysis. Two-stage ABE fermentations in BOBB increased the butanol productivity by up to 37.5% over the one-stage fermentation. Butanol productivity was further increased by 97% when gas stripping was integrated to the two-stage ABE fermentation. While the one-stage fermentation integrated with gas stripping increased the butanol productivity by 69% to 0.12 gL⁻¹h⁻¹ (as opposed to 0.071 gL⁻¹h⁻¹ in a similar fermentation without gas stripping). A simple model to describe the one-stage at oscillatory Reynolds number (Re₀) 0 and 938, and the two-stage ABE fermentation in BOBB II was developed. The model summarizes the physiological aspects of growth and metabolite synthesis by Clostridium GBL1082. The prediction of the models were in good agreement with experimental results incorporating mixing (Re₀938) and moderately agreed with results from Re₀0 and the two-stage fermentation.