Gas phase production of mixed culture phototrophic bacteria

Abstract

H2S is one of the major contaminants in biogas, landfill gas (up to 700 ppmV), and natural gas (up to 200 ppmV), which are otherwise rich in CH4. H2S impedes any further use of the gases due to its highly corrosive nature. Typical H2S removal processes are based on absorption or adsorption, where the sulfide is removed either physically or biologically. These processes generally require chemicals and often produce waste gases and solids. This thesis aims to investigate a novel H2S removal process employing a mixed culture of purple phototrophic bacteria (PPB). PPB are mostly anaerobic and can grow photo-autotrophically with H2S as an electron source, oxidising the sulfide to sulfate. Thus, a PPB driven desulfurisation process was developed as a one-stage sulfide absorption and removal system. In a scaled application, it could potentially use sunlight as an energy source with minimised chemical input. Additionally, the process may yield microbial protein from PPB as a high-value product besides the desulfurised gas. PPB were grown photo-autotrophically with sulfide as electron donor and inorganic carbon as carbon source, supplying macro-nutrients with a synthetic medium and centrate in separate experiments. The experiments were conducted with low (27 ± 3 Wm-2) and high (56 ± 11 Wm-2) infra-red (IR) irradiance. The results showed a maximum sulfide removal rate of 1.79 ± 0.16 mg-S (Lh)-1 for the low and 2.9 mg-S (Lh)-1 for high irradiance. Sulfide removal rates using centrate were similar in the low irradiance case; however, additional micro-nutrients were required (Fe and Mn). The microbial community shifted to Chromatiaceae, a family of purple sulfur bacteria (PSB), capable of sulfide oxidation. The sulfide removal was identified to be a multi-step process, using the experimental results and process modelling. Subsequently, a continuous PPB desulfurisation bubble column was set up and fed with a biogas mixture (2000 ppmV H2S and 30% CO2). Municipal digester centrate was used as a macro-nutrient source. Removal rates up to 6.62 ± 0.05 mg-S (Lh)-1 were achieved. Attached and suspended biomass removed sulfide at the same rate, indicating that biofilm formation did not adversely affect the removal. PPB oxidised sulfide to sulfate, which resulted in a faster pH decrease at higher sulfide loading rates. The feasibility of a scaled-up process was investigated with a continuous process model, revealing pH limitation with increasing sulfide loading rate, avoidable by decreasing the liquid hydraulic retention time (HRT). A major limitation of any gas-fed process is gas-liquid mass transfer. Increasing the specific bubble surface area by decreasing bubble size, increases mass transfer. However, bubbles scatter the required radiation for a phototrophic process, and small bubbles potentially scatter it throughout the bulk. A computational fluid dynamics (CFD) model was set up to investigate the refraction and reflection on bubble surfaces and determine a scattering phase function dependent on bubble size. The resulting function described the asymmetry factor of the Henyey-Greenstein phase function model depending on bubble size with an exponential decay model. Bubble size-dependent scattering was implemented in a CFD model combining radiative transfer with gas-liquid mass transfer. Simulations with large bubble size and small bubble size were carried out to quantify the effect of bubble size on mass transfer and radiation, which revealed that a larger bubble size increases radiative energy locally, due to scattering. Overall, the experiments and modelling confirmed the biological capacity of PPB for gas desulfurisation, the possibility of using centrate as a low-cost nutrient source, and the feasibility of a scaled-up continuous process. Besides the desulfurised gas, microbial protein from the PPB may be a valuable product stream and, other than centrate, the process does not require chemical input. However, major hurdles include light supply at night, low rates, and low overall biomass productivity. The application as a standalone desulfurisation unit is questionable, especially compared to current conventional and biological technology.