亚临界丙烷萃取米糠油及其微量活性组分动力学模型

为了揭示亚临界流体萃取米糠油及其微量活性组分的动力学规律，该文以亚临界丙烷为萃取介质，研究了米糠油、γ-谷维素及α-生育酚等活性组分在不同萃取温度和时间的萃取率，并基于 Baümler 模型对目标物的萃取率进行模型拟合，建立相应的动力学模型并验证。结果表明，亚临界萃取过程中目标物的相对萃取率随时间增长递增，且在初始阶段（洗涤过程）的增长速率快于第二阶段（扩散过程）的增长速率；同时，升高萃取温度可同时提高目标产物的扩散系数和相对萃取率；基于Baümler模型可较好拟合亚临界丙烷萃取米糠油、γ-谷维素和α-生育酚的动力学过程（R2>0.95），所得动力学模型具有较好的预测准确性。由Arrhenius方程可计算出米糠油、α-生育酚和γ-谷维素的亚临界萃取反应活化能，其中米糠油的活化能最低（5.23 kJ/mol），α-生育酚次之（7.05 kJ/mol），γ-谷维素最高（9.11 kJ/mol），表明γ-谷维素最难萃取且对温度依赖程度最高；根据拟合所得动力学模型，对米糠油提取率进行预测，结果表明，该模型对米糠油萃取率的预测准确率达95.8%，而对γ-谷维素和α-生育酚萃取率的预测准确度稍低，分别达94.7%和94.4%，对三者均具有很好的预测准确性。总之，基于 Baümler 模型可以较好地建立米糠油及其有益伴随物的亚临界萃取动力学模型，从而较为准确的预测米糠油及其微量活性组分的萃取率，对米糠油的亚临界萃取工艺具有一定的理论指导意义。

Extraction kinetics of rice bran oil and its minor bioactive compounds from rice bran by subcritical propane

Rice bran, the main by-product of rice process, contains oil and lots of bioactive compounds, such asγ-oryzanol andα-tocopherols. Preliminary study shows that the active ingredient in rice bran could be well preserved during the subcritical fluid extraction. In order to reveal the extraction kinetics of rice bran oil (RBO) and its soluble active ingredients during subcritical propane extraction process, rice bran oil and active ingredient (γ-oryzanol andα-tocopherols) were extracted from fresh rice bran using subcritical propane as medium at different temperature and time. We developed a two-dimensional transitory mathematical model by using Baümler model to simulate the kinetics of RBO andγ-oryzanol,α-tocopherols during subcritical propane extraction process. The diffusion model of Fick's law, which fit for the system of suspending cylindrical particles in a dilute solution, was used to describe the diffusion process of oil and its minor compounds. The results showed that,Yt/Y∞(the mass ratio of substance that has diffused at time t and infinity time) increased with extraction time. Moreover, the free lipid on rice bran surface was extracted more easily than the cell bound lipid inside. As a result, the increasing rate of Yt/Y∞ and the effective diffusion coefficient of RBO at initial stage were both faster than that of second stage during the extraction process. Similarly, the trends ofγ-oryzanol andα-tocopherols were similar. However, due to the higher energy barrier ofγ-oryzanol than rice bran oil, the rate ofYt/Y∞increase was slower than that of rice bran oil in the same extraction condition. In general, as temperature increases, the vapor pressure of the extracts increase, resulting the increase of diffusion coefficient and mass transfer coefficient. This can be used to explain the result that higher temperatures enhanced oil and minor compounds extraction from rice bran as theYt/Y∞and effective diffusion coefficient increased. The De value ofγ-oryzanol was lower than that of rice bran oil. Only when the extraction temperature was 55℃, the De value ofγ-oryzanol (3.415×10-8m2/s) and RBO(3.461×10-8 m2/s)be similar. It indicated that the extraction ofγ-oryzanol was more dependent on the temperature. Theα-tocopherols were extracted more difficult because of their amphiphile structure. The effective diffusion coefficient was positively correlated with the temperature. The results also showed that, the extraction ofα-tocopherols was significantly influenced by the temperature. The Baümler model fit well the extraction kinetics of rice bran oil,γ-oryzanol andα-tocopherols (R2>0.95)during the subcritical propane extraction process. The temperature dependence of the diffusion coefficient was represented by an Arrhenius-type equation for target extract, and the extraction ofγ-oryzanol depended on temperature mostly. Activation energy ofγ-oryzanol (9.11 kJ/mol) was higher than that of oil (5.23 kJ/mol) andα-tocopherols (7.05 kJ/mol). Based on the extraction kinetics model obtained, high predicted precision of rice bran oil,γ-oryzanol andα-tocopherols from rice bran extracted by subcritical propane, which were 95.8%, 94.7% and 94.4%, respectively. In conclusion, based on Baümler model, the extraction kinetics of rice bran oil and its concomitant subcritical can be better revealed, and the model can predicted the Yt/Y∞of rice bran oil and its trace active components accurately to guide the subcritical extraction process of rice bran oil and its trace active components.