Genetic algorithms in computational materials science and engineering: simulation and design of self-assembling materials

We introduce here two genetic algorithms that were developed in order to ai d in the design of molecules for self-assembling materials. The first const ructs molecules from sets of chemical building blocks, searching For candid ates that are determined by an ancillary modeling program to assemble into low-energy aggregates. The results of running this Genetic Algorithm (GA) o n a set of building blocks are discussed in the context of experimental obs ervations on molecules synthesized from these chemical components. The seco nd genetic algorithm attempts to find the most favorable configuration of f our molecules in space, as determined by an empirical molecular mechanics f orce field. We present the results of the application of this GA to molecul es that have been studied experimentally in our laboratory. The two genetic algorithms promise to be of use not only in the context in which they are presented, but also in a wide variety of future applications in molecular d esign and modeling. (C) 2000 Elsevier Science S.A. All rights reserved.