(604j) A Molecular Signature Based Approach for the Design of Reactive Systems

The simultaneous consideration of process and product design aspects is essential for the optimal design of a process. A variety of different algorithms have been published in the last decade that addresses this issue. However, the existing techniques for solving such problems are limited to non-reactive systems. In this work, we are introducing an algorithm that can be used for solving integrated process and molecular design problems of reactive systems on a property based platform. Property clustering is a well-known technique for a component independent design. Property clustering technique will be extended into identifying a set of target properties for the components in the system that provides the optimum process performance. Once the property targets have been identified, a molecular design problem can be formulated to identify the potential candidate molecules that meet the targets identified in the process design problem. The molecular design involves the identification of potential molecules possible from the specific types of reactions in the process. To design molecules, a recently introduced concept known as molecular signature descriptors have been used. The signature is a systematic coding system of atom types and the signature of a molecule can be obtained as a linear combination of its atomic signatures (Visco et al 2002). The molecular signatures can be tailored to track the change in molecular groups in a molecule resulting from different types of chemical reactions. This task is achieved by forming the signatures for both reactants and products and by considering them as un-connected graphs. The changes in the chemical structure can be correlated with the changes in the properties of the molecule. Therefore, the changes in the molecular structure due to reactions can be represented as a function of the property. The developed algorithm applies different qualitative structure activity/property relationships (QSAR/QSPR) to estimate properties from the molecular structure. QSAR relationships make use of different topological indices and it has been proved that a number of topological indices of molecules can be represented in terms of molecular signatures and it is possible to correlate the topological indices to the actual properties and biological activities (Faulon et al 2003). Here, the new algorithm utilizes molecular property operators based on signatures for solving the inverse problem of obtaining the molecular structures that satisfy the property targets estimated in the process design step. A new set of equations will be employed to ensure that the molecule meets the safety and environmental constraints as well.  The principles in graph theory are incorporated to track signatures and to avoid the generation of infeasible molecular structures. This contribution will illustrate the developed methods and highlight their use through a case study.