生質酒精是利用微生物醱酵把生質料源中的醣分轉化為酒精。而製造生質酒精的料源大致區分為糖質原料、澱粉質原料、纖維質原料。其中的纖維質原料經過水解程序可得到五碳糖與六碳糖(大部分是木糖與葡萄糖)。 本研究利用不同比例的葡萄糖與木糖作為基質進行批次醱酵。實驗結果發現,Pichia stipitis會先消耗完葡萄糖再消耗木糖,依據此現象我們提出適合的動力學模式,並且以實驗數據為依據,利用混合差值演算法(Hybrid Differential Evolution, HDE)估計參數以建立模式。此模式可以預測混合糖的消耗、菌體和酒精的生成對時間的關係,模擬的結果準確預測實驗數據。接著利用建立的批式醱酵動力學模式設計饋料批式醱酵的最佳進料策略,並利用計算出的最佳進料策略進行饋料批式醱酵實驗,實驗結果得知饋料批式醱酵的酒精生產力為批式醱酵的1.18倍。
Microbe can be applied to convert biomaterials to produce bioethanol. Such raw materials are roughly categorized as sugar, starch and lignocellulose. Lignocellulose can be hydrolyzed to pentose and hexose, e.g. xylose and glucose in general. In this study, we use different synthetic glucose/xylose mixtures to ferment into ethanol in batch process. Pichia stipitis consume glucose first and then xylose. We propose a suitable kinetic model based on this phenomenon and use Hybrid Differential Evolution to estimate parameters. The model can predict the mixed-sugar consumption and the formation of biomass and ethanol with time-course data, and the simulation results accurately fit the experimental data. Then, the optimal feeding strategy of fed batch fermentation that we used to design fed batch experiment depends on the kinetic batch model we building up. Finally, the experimental results indicated that the ethanol productivity of fed-batch fermentation was 1.18-fold of batch fermentation.