Biobutanol production in a Clostridium acetobutylicum biofilm reactor integrated with simultaneous product recovery by adsorption

Abstract

Background: Clostridium acetobutylicum can propagate on fibrous matrices and form biofilms that have improved butanol tolerance and a high fermentation rate and can be repeatedly used. Previously, a novel macroporous resin, KA-I, was synthesized in our laboratory and was demonstrated to be a good adsorbent with high selectivity and capacity for butanol recovery from a model solution. Based on these results, we aimed to develop a process integrating a biofilm reactor with simultaneous product recovery using the KA-I resin to maximize the production efficiency of biobutanol.

Results: KA-I showed great affinity for butanol and butyrate and could selectively enhance acetoin production at the expense of acetone during the fermentation. The biofilm reactor exhibited high productivity with considerably low broth turbidity during repeated batch fermentations. By maintaining the butanol level above 6.5 g/L in the biofilm reactor, butyrate adsorption by the KA-I resin was effectively reduced. Co-adsorption of acetone by the resin improved the fermentation performance. By redox modulation with methyl viologen (MV), the butanol-acetone ratio and the total product yield increased. An equivalent solvent titer of 96.5 to 130.7 g/L was achieved with a productivity of 1.0 to 1.5 g · L-1 · h-1. The solvent concentration and productivity increased by 4 to 6-fold and 3 to 5-fold, respectively, compared to traditional batch fermentation using planktonic culture.

Conclusions: Compared to the conventional process, the integrated process dramatically improved the productivity and reduced the energy consumption as well as water usage in biobutanol production. While genetic engineering focuses on strain improvement to enhance butanol production, process development can fully exploit the productivity of a strain and maximize the production efficiency.

Figure 1

Kinetics of batch fermentation by planktonic culture. (A) pH, optical density (OD)_600nm and glucose consumption; (B) product concentrations during the fermentation process. Initially, experiments were performed in six Duran bottles under the same condition. At 40 h, three of the bottles were supplemented with KA-I resin (final concentration 50 g/L) (open symbol and dashed line), with the other ones left as control (solid symbol and solid line). The mean value (± SD) was calculated from the results of parallel runs. The vertical dashed lines represent the equivalent concentrations adsorbed by the resin (determined by dividing the adsorbed amounts by the volume of the fermentation broth) at the end of the fermentation.

Figure 2

Schematic diagram of a biofilm reactor coupled to fixed-bed adsorption. The KA-I resin was packed in glass columns of 250-mm length and 30-mm diameter. Approximately 50 to 60 g of the resin was loaded in each column. After autoclaving, the resin column was connected to a 2-L biofilm reactor to construct an online product removal system. With feeding of concentrated P2 medium, fed-batch fermentation can be performed. Operation details are described in the Methods section. SEM, scanning electron microscopy; T, temperature; P, pressure.

Figure 3

Kinetics of fed-batch fermentation in the biofilm reactor with fixed-bed adsorption. After 13 h of batch culture, concentrated medium was fed into the reactor to maintain desired glucose levels and the fixed-bed adsorption was started, as is indicated by arrows. Butanol was selectively adsorbed by the KA-I resin, whereas other solvents accumulated in the fermentation broth. OD, optical density.

Figure 4

Kinetics of the fixed-bed adsorption. The fixed-bed adsorption was started at13 h. A sample before the adsorption (influent, black square) and a sample after the adsorption (effluent, red circle) were withdrawn simultaneously every time to determine the extent of resin saturation. A higher sorbate titer in the effluent meant that the sorbate initially adsorbed onto the KA-I resin was subsequently eluted in the butanol adsorption process. The resin column was replaced with a new resin column when saturated by butanol.

Figure 5

Kinetics of fed-batch fermentation in the biofilm reactor with co-adsorption of acetone. After 8 h of batch culture, concentrated medium was fed into the reactor to maintain desired glucose levels and the fixed-bed adsorption was started, as is indicated by arrows. By co-adsorption of acetone using KA-I resin, the acetone titer in the fermentation broth was also maintained at a relatively stable level. OD, optical density.