由實驗結果可知：無論是否存在觸媒，溫度對轉酯化反應的影響皆遠大於壓力的影響；且壓力對轉酯化反應的影響只有在當壓力遠小於甲醇的臨界壓力且醇/油混合物未形成單一均相時方為顯著。由視窗反應器目視觀察發現：在200 °C及4.14 MPa時，甲醇/椰子油混合物形成單一均相；且其視活化能亦由在低溫（180 °C以下）時之107.7 kJ/mol減少為在高溫（220 °C以上）時的35.3 kJ/mol，變得更有利於此轉酯化反應的進行。以上結果顯示：此轉酯化反應並不一定需要在超臨界甲醇、或是醇/油混合物的超臨界狀態下進行，而是只要有足夠高的溫度及壓力可以使得醇/油混合物形成均一相即可。
在高溫操作時的最佳滯留時間與溫度成反比；溫度越高，則所需滯留時間越短。另在我們實驗所採用的12/1到60/1的醇/油莫耳比範圍中，轉化率及視反應速率k皆隨著醇油比的增加而增加，例如：當甲醇/椰子油莫耳比從12/1增加到60/1（5倍）時，其視反應速率k值亦自0.00476 s–1增為0.02118 s–1（4.45倍）；我們並未發現有所謂最佳醇油比的存在。
In this dissertation, both types of supercritical fluid reactions, “reaction with supercritical fluid” and “reaction in supercritical fluid”, were adopted for the study of preparation of biodiesel and improvement of resistant random access memory (RRAM) performance, respectively.
In Chapter 1, a brief introduction to the development, properties, and applications of supercritical fluid was made.
In the beginning of Chapter 2, the catalytic effect of metal reactor surface was investigated. Ever since Saka and Kusdiana proposed the method of preparation of biodiesel by supercritical fluid technology, “non-catalytic” has been recognized as one of the most advantages of this process. Nevertheless, our experimental results showed that, in the transesterification of vegetable oils with supercritical methanol, the reaction rate was indeed accelerated by the catalytic effect of stainless-steel reactor surface, resulting in a high conversion; after the deactivation of this catalytic ability, the biodiesel yield was decreased.
Then we went on the screening of catalysts. Among various metal oxides tested in this study, MnO2 was found to be the most effective catalyst. The presence of MnO2 was essential for the complete conversion of vegetable oils to biodiesel under mild conditions; the conversion was relatively low during catalyst-free operation.
Thereafter, transesterification of supercritical/subcritical methanol with coconut oil and jatropha oil were conducted in a continuous operation system. With or without the addition of catalyst and co-solvent, the effects of the operating variables, namely the temperature, pressure, residence time, methanol-to-oil molar ratio, on the yield of biodiesel were systematically examined.
Our experimental results indicated that: regardless of the presence of catalyst, the effect of temperature on transesterification was more pronounced than that of pressure; the latter was apparent only at pressures far below the critical pressure of methanol and before the formation of a homogeneous liquid phase from the methanol/oil mixture. Through visual observation in a windowed-reactor, at 200 °C and 4.14 MPa, the methanol/coconut oil mixture formed a homogeneous liquid phase; the apparent activation energy decreased from 107.7 kJ/mol at temperatures below 180 °C to 35.3 kJ/mol at temperatures above 220 °C, more favorable for transesterification. The obtained results revealed that this transesterification does not necessarily have to be performed in supercritical methanol, nor in supercritical methanol/oil mixtures, but only at temperatures and pressures where a homogeneous liquid phase exists.
The optimal residence time for the transesterification was dependent on the reaction temperature; higher temperatures required shorter residence times. The FAME yield and the apparent rate constant k both increased upon increasing the molar ratio, for example, when the molar ratio of methanol to coconut oil increased from 12/1 to 60/1 (fivefold), the apparent rate constant (k) also increased from 0.00476 to 0.02118 s–1 (4.45-fold); we did not, however, observe an optimal molar ratio within the range from 12 to 60.
The effect of co-solvent in a continuous operation mode was investigated at the end of Chapter 2, and the experimental results showed that the effect of co-solvent on transesterification was negligible or even negative. The addition of co-solvent might enhance the miscibility between oils and methanol; on the other hand, it also increased the flux volume of transesterification , resulting in decreases in both the concentrations of reactants and residence time. Therefore, the overall effect of co-solvent in a continuous operation mode might be negligible or even negative.
In Chapter 3, the operating current of silicon oxide-based RRAM was reduced by supercritical fluid processing (SFP) technology. At a temperature of 120 oC, with the facilities of low viscosity, low surface tension and high diffusivity of supercritical carbon dioxide, the water molecules could easily diffuse into the film and repair the dangling bonds of grain boundary; by SFP, the conduction path of RRAM film became discontinuous and its conduction resistance also increased due to the reduce of defects in the film, resulting in a significant decline in operating current. With the reduction of operation power consumption of RRAM, the degradation of IC caused by the Joule heat would therefore be improved. Thus, SFP techniques can improve the switching characteristics of RRAM and its operation performance, showing a great benefit on the development and applications of RRAM as next-generation non-volatile memory.
In the experiment, the dangling bonds of Tin-doped Silica (Sn:SiO2) film were repaired by supercritical carbon dioxide (SCCO2). A discontinuous metal filament would be formed in Sn:SiO2 film through SCCO2 passivation process, causing the device current declined. In addition, we also use this technique to treat the RRAM with ITO transparent conductive electrode to effectively reduce the power consumption and operating voltage of device.
At last, SCCO2 treatment technology was used to manipulate the temperature coefficient of resistance (TCR) of TaN thin-film resistors. After annealing process, the TCR value of TaN film resistor was changed from negative to positive; by SCCO2 treatment, the positive TCR value turned back to negative again. Through optimization of supercritical fluid technology combined with thermal annealing method, the TCR value of TaN thin-film resistor could be modulated to close to zero, making it conform the requirements of a stricter specification for car-used electronic applications or other harsh environments of high temperature.
In Chapter 4, a summary of the content in this dissertation was made.