The goal of this work was to establish the efficiency of using computational design to sort through potential solid polymer electrolyte structures, and then verify the computational predictions with a traditional laboratory experimental program. The promise of this approach is demonstrated by the results presented. In this combined computational-experimental program a set of heteropolymer solid polymer electrolytes were first constructed in virtual space and the diffusion of lithium salts predicted as a function of chemical structure, including the salt, and other environmental conditions. These computational predictions were then verified by experimentation. More specifically, some previously untested heteropolymer structures have been designed using modeling, and their lithium ion diffusion rates predicted in comparison to those of known control structures. These new polymers were then synthesized and characterized in the laboratory with the results substantiating the predictions. This has led to an expanded synthesis program, one that is efficient and cost effective. (C) 2003 Elsevier Science B.V. All rights reserved.