本研究使用一種新方法製備生質柴油。以直接皂化─酯化的方式、在高含水量及高游離脂肪酸含量的條件下,能有效率地快速破碎細胞並轉換其內的油脂為生質柴油,且整個製程可在單一的反應器中進行。本製程具有以下幾個優勢:首先,直接皂化─酯化製程省卻了油脂萃取及精煉的步驟,節省了相當高的預處理成本;其次,與鹼催化直接轉酯化的製程相比,本製程可處理高含水量及高游離脂肪酸含量的系統,不受其限制;再者,與鹼、酸催化轉酯化製程所需時間相比,本製程可更快地進行生質柴油的生產。 本研究使用了小球藻(Chlorella sp. ESP-6)及黏紅酵母菌(Rhodotorula glutinis)作為生產生質柴油的原料,並以此研究直接皂化─酯化製程反應機制及其動力學。本研究更進一步探討來自於批次間、跨系統生物含油量的差異,來驗證所提出的反應機制。以微藻和酵母菌原料進行反應動力學的模擬發現,除了油脂萃取及酯化的平衡常數,所有動力學模型皆可使用相同的參數得到不錯的模擬效果。油脂萃取的不同主要是由於細胞本身機械強度不同。微藻細胞相較酵母菌細胞機械強度較高,造成細胞不易破碎,所需萃取時間較長;而酯化平衡常數的不同推測是由於脂肪酸組成本質上的不同所致。微藻及酵母菌為原料所生成的生質柴油平均分子量分別為283及291,此乃源自於脂肪酸碳鏈長度的差異,且兩種原料的不飽和脂肪酸組成也有接近10個百分比的差異。然而,在兩個系統中,本研究提出的動力學機制皆被成功地驗證且根據動力學參數模擬實驗值也得到良好的結果。
We have proposed a novel DSEC (direct saponification-esterification conversion) strategy that could directly disrupt cells and convert lipid into biodiesel efficiently in a single vessel under the conditions of high water and fatty acid. It is advantageous at least for several reasons. First of all, DSEC eliminates lipid extraction and purification steps in the conventional process, which usually accounts for more than half of production cost. Secondly, compared with the alkali catalyzed transesterification process, DSEC completely alleviated the restriction of incapable of dealing with high content of water and free fatty acid. Thirdly, the required reaction time would be a fraction of traditional acid transesterification and similar to, if not better, traditional alkaline transesterification. In this study, both Chlorella sp. ESP-6 and Rhodotorula glutinis were used as raw materials for biodiesel production to investigate and verify DSEC kinetics. With the consideration of lipid content variation, the mechanisms were confirmed and same kinetic parameters could be obtained for both yeast cells and microalgae except cell disruptions and esterification. The difference in cell disruption parameter would be obvious due to the rigidity of cell wall of microalgae compared with yeast cells. And the difference of esterification constants may be due to different carbon length of fatty acids (the average molecular weight 283 for Chlorella and 291 for Rhodotorula) and different degree of saturation. However, in both cases, our proposed mechanism were satisfactorily confirmed and the kinetic parameters with good fitting against experimental data were obtained successfully.