Designing Conducting Polymers with Genetic Algorithms
- PDF / 55,257 Bytes
- 6 Pages / 612 x 792 pts (letter) Page_size
- 67 Downloads / 305 Views
Designing Conducting Polymers with Genetic Algorithms R. Giro, M. Cyrillo and D.S. Galvão Instituto de Física, Universidade Estadual de Campinas – UNICAMP, Campinas, São Paulo, CEP 13083-970, CP 6165, Brazil. ABSTRACT We have developed a new methodology to design conducting polymers with pre-specified properties using genetic algorithms (GAs). The methodology combines GAs with the Negative Factor Counting (NFC) technique. NFC is a powerful technique to obtain the eigenvalues of large matrices without direct diagonalization.We present the results for a case study of polyanilines, one of the most important families of conducting polymers. The methodology proved to be able of generating automatic solutions for the problem of determining the optimum relative concentration for binary and ternary disordered polyaniline alloys exhibiting metallic properties. The methodology is completely general and can be used to design new classes of materials. INTRODUTION Conducting polymers constitute a new class of electronic materials with unusual properties1. The discovery by MacDiarmid, Heeger, and Shirakawa2 that polyacetylene (PA) presented a dramatic increase in the conductivity when doped by charge-transfer reactions (with a well defined transition to metallic regime) started a great number of works on these materials. The interest on conducting polymers has experienced a continuous growth attracting researches from many different areas due to their large technological potential applications and new fundamental physical phenomena1. In the last years the possibility of creating new conducting polymers exploring the concept of copolymerization (different structural monomeric units) has attracted much attention from experimental and theoretical points of view3. As structural disorder is always present in such structures, systematic theoretical analysis is often prohibitive due to the huge number of possible configurations to be tested. In this work we propose a new methodology to address these problems. It combines the use of negative factor counting (NFC)4,5 technique with genetic algorithms (GAs)6. The NFC technique allows us to obtain the eigenvalues of very large matrices without direct diagonalization. It has been proved very effective in the study of electronic structures of disordered polymeric chains when coupled to tight-binding hamiltonians7,8. GAs originated from the studies conducted by John Holland in the 1970s9. The metaphor underlying GAs is that of natural evolution. The Darwinian theory of evolution depicts biological systems as the product of the ongoing process of natural selection10. GAs follow these ideas in a very simple way and allow us to use the computer to evolve automatic solutions over time. First, a population of individuals is created in a computer (typically stored as binary strings) and then the population is evolved using processes analogues to the biological ones of mutation and crossover10 (variation, selection and inheritance). GAs allow very efficient intelligent searches in huge phase spaces l
Data Loading...
