Due to the combinatorial complexity of chemical compound space, efficient algorithms and approaches are required to traverse the space in an informed manner. We compare a series of constrained optimization algorithms for the optimization of non-linear optical chromophores for maximal static hyperpolarizability under a transparency constraint. In the present work, various heuristic primal-dual approaches are introduced and compared with each other and prior work. Each of the algorithms has been analyzed for its relative performance and ability to find feasible chemical structures. Property calculations are performed using density functional theory and feed into the heuristic optimization algorithms. In addition, computational efficacy of each algorithm has been analyzed in order to assess the best possible approach among algorithms studied.

Constrained heuristic optimization of NLO chromophores

249th National Meeting (2015)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.