Combined property clustering and GC+ techniques for process and product design

Recent developments in the area of process and product integration have enabled the systematic identification of suitable candidate molecules to meet certain process performance. In this approach, the property targets for the input molecules corresponding to the optimum process performance have been identified in the first step and the molecules that have the target properties have been designed in the next step. The focus of this work is to develop a combined property clustering and GC+ algorithm to identify molecules that meet the property targets identified during the process design stage. In our earlier works, a methodology was introduced for identifying molecules with a given set of properties by combining property clustering and group contribution methods. Yet, there are situations when the property contributions of some of the molecular groups of interest are not available in literature. To address this limitation, an algorithm has been developed to include the property contributions predicted by combined group contribution and connectivity indices methods into the cluster space. For the design of simple monofunctional molecules, a modified visual approach has been used, while for the design of more complicated structures an algebraic method has been developed. The applicability of the algebraic method has been increased by including the property contributions from second and third order groups.