2004-10-012024-05-17https://scholars.lib.ntu.edu.tw/handle/123456789/670781摘要：於本計劃，我們將應用最新穎的「多重尺度電腦模擬」的概念，預測複雜化學品及材料的熱物性質與這些物質在純或混合狀態下的相平衡。這些性質在許多領域裡，都有非常重要的應用。例如，混合物的相平衡是化工程序如蒸餾塔、萃取槽之設計中，必要的資料。化學品的熱物特性，如蒸氣壓、無限稀釋下之活性系數及水與正辛純之分佈係數，則是預測化學污染物排放至環境後最終聚集地，不可或缺的要件。同時，水與正辛純之分佈係數，在製藥業也被用做研發新藥的基本數據。透過現代計算化學理論並配合高速電腦的運算，我們可以很快速的預測出這些性質。研發更快更準的計算方法，將對上述產業有深遠的影響。 我們的目標是進一步研發新提出的 COSMO-SAC 模型，以期作更精確的物性預測，並推廣其應用的範圍。COSMO-SAC 是依照「多重尺度電腦模擬」的概念，基於連續態溶液熱力學理論，所發展出來的模型。其原理是先將溶液簡化成一介電材料,再利用量子化學計算溶質分子與介電材料間的作用力，最後透過嚴謹的統計力學模型，計算出溶液以及其成份的各種熱物性質。在本計劃中，我們將配合真實分子的分子動力學電腦模擬，與量子化學的計算，發展新的技術，以提昇COSM<br> Abstract: In this project, we propose to utilize the novel concept of computational hierarchy for the prediction of the thermophysical properties of chemicals and complicated materials and their phase equilibria. This is of great important for various disciplines, including the chemical, environmental, and pharmaceutical industries, as a priori estimation of the properties for compounds of interest allows for the design and development of a new process new drug at a much lower cost. Here, we are particularly interested in applying and improving the COSMO-SAC model, recently developed by us, for making property predictions. This new method, based on molecular solvation, is one of the most successful approach for making predictions of thermophysical properties from first principle calculations, i.e., without the need of experimental input. In this model quantum mechanical COSMO calculations are performed to obtain the screening charges for molecules in a perfect conductor. A statistical mechanical model that considers molecules to be a collection of surface segments is used for the calculation of segment activity coefficients using the screening charges. Activity coefficients for molecules are then obtained by summing the contributions from the segments. The activity coefficients are the most important quantities in making fluid phase equilibrium. Furthermore, properties such as the infinite dilution activity coefficient, the octanol-water partition coefficient, the excess enthalpy and entropy, etc., which are needed in the prediction of chemical fate in the environment and in the design of new drugs, can be readily derived from the activity coefficient. Our focus in this project is to extend the capability of the COSMO-SAC model to a broader range of problems, including supercritical phenomena, complicated systems such as polymers and protein-ligand binding, and also improving its accuracy. We are also interested in examining the fundamental assumptions such as the linear response approximations by using atomistic molecular dynamic simulations, and attempt to determine all the model parameters from first principles.