Kinetic Analysis of Anaerobic Hydrogenesis-Methanogenesis Process for High Strength Brewery Wastewater

• Main Authors: Yi-Hua Chang, 張益華
• Other Authors: Kuo-shuh Fan
• Format: Others
• Language: zh-TW
• Published: 2002
• Online Access: http://ndltd.ncl.edu.tw/handle/66368965798907926608

碩士 === 國立高雄第一科技大學 === 環境與安全衛生工程所 === 90 === ABSTRACT Anaerobic fermentation is formed by many microorganisms, which convert organic materials into methane, carbon dioxide, hydrogen, acids and alcohols. Hydrogen is an intermediate that spontaneously transformed to methane by methane-forming bacteria in conventional anaerobic fermentation. Nevertheless, in two-phase anaerobic fermentation system, hydrogen can be detected in acid-phase digester. In order to research the optimization production rate of hydrogen and methane, this study carried out hydrogenesis-methanogeneis process. Microbial kinetic analyses were applied to study the kinetic model for the biochemical reactions. This study focused on the effects of HRT on hydrogen and methane production. Kinetic models were used to analyze the reaction steps in hydrogensis and methanogenesis process. The experiment was conducted in a two-stage system. The first stage with 4L in volume served as the hydrogen-forming digester and the second stage with 8L in volume served as the methane-forming digester. Both the reactors were completely stirred. Contrate brewery wastewater from a nearby plant was collected as the substrate and the concentration was adjusted to 110 g/l in COD for the feed. The two-stage system was operated in 9 consequent sets. In which, the hydrogenesis reactor was controlled at 3 different HRT, 24, 16,and 8 hr. The methaanogenesis reactor was fed with hydrogenesis reactor effluent and controlled at 3 different HRT, 10, 15, and 20 days. For the hydrogen-forming digester, the maximum performance occurred at HRT=8hr which determined that the production hydrogen rate, COD removal efficiency, volumetric conversion efficiency and substrate utilization rate were 5.42 L H2/L-reactor-day, 3.32%, 10.20 g/L-day ,and 0.25 gCOD/gVSS-day, respectively. Three microbial kinetic models including Monod, Contois and Chen & Hashimoto models were adapted to described kinetic condition of the hydrogen-forming digester operated at different HRT. The results showed that as HRT=24 hour, the transport rate coefficient (k), hydrolysis rate coefficient (Kh), maximum specific growth rate (μm), half-saturation coefficient (Ks) and growth yield (Y) were 1.68hr-1, 0.0024hr-1, 0.107hr-1, 0.598mg/L and 0.047 VSS/COD, respectively. As HRT=16 hour, k, Kh, μm, Ks and Y were 4.03h r-1, 0.0034hr-1, 0.241hr-1, 0.739mg/L and 0.107 VSS/COD, respectively. As HRT=8 hour, k, Kh, μm, Ks and Y were 4.73h r-1, 0.0158hr-1, 0.368hr-1, 0.773mg/L and 0.111 VSS/COD, respectively. As compared with these kinetic parameters, it demonstrated that the hydrolysis should be the rate-limiting step for the hydrogen fermentation. The affinity for substrate of hydrogen-forming bacteria was affected by butyric acid. It showed reverse relationship as butyric acid increased. Nevertheless, butyric acid became dominant as HRT closed to the limit. The developed model showed a good fitness with the observed data. The correlation coefficients were between 0.84 and 0.86 for the studied HRTs. Methane production rate from Methane-forming digester was low, because it was affected by influent substrate that contains high VFA. Comparison with different operating indicators (composition of feeding substrate, methane production rate, etc.) of methane-forming digester in each run, it illustrated that high butyric acid reduced methanogenesis efficiency. Due to the high strength VFA in substrate inhibited methanogenic activities, Andrew model was employed to analyze the kinetic reaction of methanogenesis. The results indicated that the maximum specific growth rate (μm) was 0.32-0.09 day-1. The half-saturation coefficient (Ks) was 10.35-1.27 g/L. The inhibition coefficient (Ki) was 90.85-17.62g/L. Ki increased with the extension of HRT. On the other hand, it decreased with increasing VFA. However, the variations of μm and Ks showed an inverse pattern with that of Ki . The model showed a reasonable prediction. The R2 were greater than 0.74