Multi-Scale Chemical Product Design using the Reverse Problem Formulation

The main objective for this research is to extend the reverse problem formulation algorithm to include aspects of multiple length scales, thereby creating a framework where product synthesis, design, and optimization can be achieved with significantly reduced computational time. In order to achieve this objective a centralized framework was developed that combines property clustering with chemometric techniques like principal component analysis (PCA) and partial linear regression onto latent surfaces (PLS) to solve the design problem in a property descriptor sub-domain. Information from the molecular scale on short range order, such as group structure, conformation, and stereoregularity, is combined with information from the mesoscale on long range order, such as the particle size, to design a set of alternative excipients that exhibit superior properties for use in an acetaminophen tablet without the use of parallel, grid, or supercomputing techniques.