The Modelling of Thermodynamically Difficult Systems

9521406 Sandler实验测量非常耗时，并且测量所有物质和混合物的热力学性质和相行为是不切实际的，这对于化学和制药加工，环境和安全工程以及化学加工的其他方面都很重要。因此，有准确的预测或至少相关的方法是很重要的。本建议涉及工业上重要但复杂的混合物的热力学性质和相平衡的预测模型的发展研究。将考虑三种不同类型的混合物：氢键流体，蛋白质溶液和胶体系统。目前的基团贡献方法对氢键流体的描述很差，正如本提案所记录的那样。原因是氢键的程度和强度受到与氢键基团相邻的分子上的官能团的强烈影响。我们已经通过实验和从头算量子力学计算证明了这一点。我们现在建议开发新一代的预测方法，该方法是基于可推广色散力的基团贡献方法和氢键的分子特异性量子化学计算的结合。蛋白质溶液及其纯化在许多生物加工应用中都很重要。这里提出的研究包括使用分子热力学和统计力学建模以及计算机模拟来开发水两相系统中蛋白质分配和蛋白质沉淀的预测模型。胶体介于物质的宏观和微观状态之间，它们的行为影响着化学、生化和制药加工中的许多新兴技术，因此引起人们的兴趣。液中液胶体体系的相行为有两个尺度。在宏观尺度上是液-液相分离，或宏观相平衡。微观尺度的行为是在某些（或全部）分离相中形成微观结构。液体胶体的许多重要特性是由形成的微观结构决定的。然而，为了研究微观结构，首先需要了解宏观相行为，因为这确定了微观结构形成的相边界。这类系统的热力学建模相当简单，并且没有捕捉到液体胶体系统宏观相边界的温度和压力依赖关系。在此，我们提出利用现代热力学方法对液中胶体宏观相行为进行建模研究，以便更准确地描述随着温度、压力和组成的变化，胶体体系宏观相边界的变化。＊＊＊

9521406 Sandler Experimental measurements are very time consuming, and it is not practical to measure the thermodynamic properties and phase behavior of all substances and mixtures, that are of importance for chemical and pharmaceutical processing, for environmental and safety engineering, and for other aspects of chemical processing. Therefore, it is important to have accurate prediction or at least correlation methods. This proposal deals with research on the development of predictive models the thermodynamical properties and phase equilibria of industrially important, but complex mixtures. Three different types of mixtures will be considered: hydrogen-bonding fluids, protein solutions and colloidal systems. Hydrogen-bonding fluids are poorly described by current group contribution methods as documented in this proposal. The reason is that the extent and strength of hydrogen bonding is strongly influenced by functional groups on a molecule adjacent to the hydrogen-bonding groups. We have shown this both experimentally and from ab initio quantum mechanics calculations. We are now proposing to develop a new generation of prediction methods based on a combination of group contribution methods for the generalizable dispersive forces, and molecule-specific quantum chemistry calculations for hydrogen bonding. Protein solutions and their purification are important in many bio-processing applications. The research being proposed here involves the use of molecular thermodynamic and statistical mechanical modeling and computer simulation to develop predictive models for the partitioning of proteins in aqueous two-phase systems and for protein precipitation. Colloids lie between the macroscopic and microscopic regimes of matter, and are of interest because their behavior influences many emerging technologies in chemical, biochemical, and pharmaceutical processing. There are two scales of phase behavior in liquid-in-liquid colloidal systems. At the macroscale is liquid-liquid phase separation, or macroscopic phase equilibrium. The microscale behavior is the formation of microstructures in some (or all) of the separated phases. Many important characteristics of liquid colloids are dictated by the microstructures that form. However, to investigate the microstructure, one first needs to have an understanding of the macroscopic phase behavior as this identifies the phase boundaries within which microstructures are formed. Thermodynamic modeling of such systems has been rather simple, and has not captured the temperature and pressure dependence of the macroscopic phase boundaries of liquid colloidal systems. Here we are proposing research on the modeling of macroscopic phase behavior of liquid-in-liquid colloids using modern thermodynamic methods to provide a more accurate description of the changes in the macroscopic phase boundaries of colloidal systems with variations in temperature, pressure and composition. ***