| Supercritical fluid CO2 extraction (SCFE) is an environmental friendly technology, and shows an attractive future in separation process. In this thesis, a SCFE set-up with an extraction volume of 1L was established, with which both hippophae rhamnoides L. seed oil and soybean oil were extracted using supercritical CO2. The experiments show that many factors have impacts on the oil yield, such as extraction pressure, temperature, and extraction time, as well as seed particle size and charge quantity. For the extraction process of hippophae rhamnoides L. seed oil, the optimum conditions were as follows: extraction pressure of 20MPa~30MPa, extraction temperature of 35癈~40癈, supercritical CO2 flow rate of 0.15m3/h~ 0.3m /h, and extraction time of 4-5h. Under such a condition, the oil obtained is very lucid and good quality, and the yield is as high as above 90%. GC analysis shows that the oil contains 12.61% of saturated fatty acid and 86.93% of unsaturated fatty acid. For the SCFE of soybean oil, the highest oil yield was obtained at a pressure of 30MPa and a temperature of 35癈, and the soybean flakes thickness of 0. 4mm. The soybean oil obtained from SCFE process contains 16.089% of saturated fatty acid and 83.567% of unsaturated fatty acid.From the changes of oil yield with the extraction time, it can be concluded that the extraction process contains three stages: fast extraction stage (line); transitional stage and the slow extraction stage. At the first stage, 75%-80% of the oil has been extracted out.The solubility of vegetable oil in supercritical CO2 plays the most important role in the design of SCFE process. In this work, we found that there exists a great difference between ChrastiPs formula and del Valle's formula. When the temperature varies from 303K to 333K, nearly the same results are obtained from the two formulas. However, when the temperature range became wider, especially at the near critical zone, the results calculated from the Chrastil's formula became strange. In contrast, del Valle's formula has a more extensive physical meaning. However, in delValle's formula, some characteristic parameters were set as constant, which confined its application limit. In this thesis, based on our experimental results and the literature work, we proposed the solubility formula, and fit the coefficients of the formula according to the measured values of solubility of hippophae rhamnoides L. seed oil and soybean oil. The following results are obtained:C: solubility; kg/m3k: association number of solute molecules,T: temperature of supercritical fluidA1: physical constant related to the MW of solute or solventA2: physical constant related to the reactive heat of soluteA3: physical constant related to the evaporation heatFor the particular system, the above coefficients were obtained by the fitting experimental data.In this work, the artificial neural network (ANN) technology was for the first time applied to the simulation of SCFE process of hippophae rhamnoides L. seed oil and soybean oil., With the normalization pretreatment and the reasonable choice of expression style of the initial input data, we developed the "neural-regression hybrid prediction system" into the real ANN method.With the MATLAB software as platform, establish an ANN model for the SCFE kinetic simulation. With this ANN-SCFE system, we can simulate and predict the SCFE process of hippophae rhamnoides L. seed oil and soybean oil.In the ANN-SCFE system, a 3-layer BP network structure is applied, and the operation factors such as pressure, temperature and extraction time are used as input variables of the network, whereas the oil yield of extraction is treated as output values of the network. Optimization of the topological structure of the net, 6 units of middle hidden layer had been proved to be the optimum value. With thenormalization pretreatment of the initial input data, not only the convergence speed and accuracy has been improved greatly, but also the problem of derivative at zero...
|
PDF Full Text Request