The integration of environmental aspects in computer aided molecular design framework

Chemical product design is the process of identifying optimal products for specific applications based on customer needs. This process is often done by Computer Aided Molecular Design (CAMD) technique, a powerful tool to determine molecular structures that meet targeted properties using property prediction models. Most of the CAMD problems focused on designing molecules with desirable physicochemical properties without considering the environmental aspects of molecules during the decision-making stage. However, chemical products with good functionalities do not imply that they are environmental friendly. Hence, besides physicochemical properties, it was also important to consider environmental properties of molecules as design criteria. The first focus of this research was to develop a mathematical model to integrate both physical and environmental properties in CAMD framework. The CAMD problem was formulated as a multi objective optimisation problem where targeted physicochemical and environmental properties were to be optimised simultaneously. Nevertheless, a major obstacle faced by multi-objective CAMD problems was the difficulty in assigning weightages to the multiple objectives. To address this issue, Analytic Hierarchy Process (AHP) method was integrated into multi-objective CAMD framework to provide a systematic way in assigning the consistent weightages to multiple objectives involved in CAMD formulation. The work was further extended to include safety and health aspects as design criteria in CAMD problems. However, there was an increase in complexity and fuzziness of CAMD problem when safety, health and environmental (SHE) aspects as well as physical properties were all considered as design criteria. This was because information needed for evaluating the relative importance between SHE properties were more uncertain and complex. Fuzzy Analytic Hierarchy Process (FAHP) was embedded into CAMD framework to handle the fuzziness and ambiguity of subjective opinions occurred during the pairwise comparison. Following that, it was realised that the attributes of a solvent not only affected its functionality but environmental performance of a process. Thus, the next focus of the work was to develop a single stage CAMD framework that can simultaneously identify solvent that meets predefined properties and improves environmental performance of its recovery process. The developed methodology integrated the quantification of environmental impact of solvent recovery process into CAMD framework. IChemE Sustainability Metrics was applied to calculate the total environmental burden of solvent recovery process. There was also a need to evaluate the effect of various solvent recovery processes on molecular design as each of these separation techniques required different amount of energy for solvent recovery. The total utility cost and energy required for solvent recovery were integrated into CAMD framework. With this approach, the generated solvents can be recovered through a cheaper and cleaner process and gave a better balance of performance for a set of predefined properties. In the last part, the thesis highlighted the potential extension and future works in this area of research work.