Comparing Ethanol Production Of Carboxydotrophic Clostridium Strains During Syngas Fermentation With A Two-Stage Continuous Culture

Several strains of anaerobic bacteria in the genus Clostridium can convert syngas, which consists of CO, H2, and a small amount of CO2, into ethanol via the Wood-Ljungdahl pathway in a process called syngas fermentation. Process optimization to increase production rates, improve gas/liquid mass transfer, and to control product ratios is important to render the process efficient and economically viable. Strain differences in growth and product concentrations have been reported in the literature, yet experiments varied in reactor design, medium, batch vs. continuous fermentation, pH, syngas composition, and other parameters. To directly compare the productivity of these biocatalysts, some of which are indistinguishable based on 16S rRNA phylogeny, C. ljungdahlii ERI-2, C. ljungdahlii PETC, and C. autoethanogenum JA1-1 were separately cultured in a long-term continuous syngas fermentation system optimized for ethanol production. The pH was lowered during the fermentation runs in a controlled manner in an attempt to induce solventogenesis. For each strain, duplicate fermentation runs were conducted, and for all strains the results of these duplicates runs were reproducible. A shift to solventogenesis was not achieved with JA1-1 and lowering the pH adversely affected its growth and CO consumption. ERI-2 and PETC performed similarly. Ethanol production rates for ERI-2 and PETC were correlated to biocatalyst density more so than pH, whereas acetate production rates for these strains decreased when pH was lowered. The highest average ethanol production rate of 0.301 g/L/h was generated with PETC at pH 4.5 with a corresponding 19 g/L ethanol concentration and 5.5:1 ethanol/acetate ratio. Carbon conversion to ethanol for ERI-2, PETC, and JA1-1 was 70%, 75%, and 28% of theoretical maximum, respectively. A minimum mass transfer coefficient (KLa) of 40 h-1 was calculated for our second-stage bubble column when considering the maximum possible gas/liquid driving force; however, an apparent KLa value of 190 h-1 was found to be representative of normal continuous operation. The results provide information on long-term (>700 h), two-stage continuous syngas fermentation using a bubble column reactor, and on the dissimilar effect of pH 4.5 on C. ljungdahlii and C. autoethanogenum growth and ethanol/acetate production. Keywords: second-generation biofuel, biomass, renewable energy, syngas fermentation, continuous fermentation, two-stage fermentation, Clostridium ljungdahlii, Clostridium autoethanogenum, ethanol, acetate, 2,3 butanediol, carbon monoxide, mass transfer ii