This proposal seeks to develop a set of modelling protocols to design, characterize and invent macromolecular materials for molecular capture, separation and detection. The approach combines multi-scale modelling of the structural, dynamic, electronic and optical properties of the target materials with an evolutionary algorithm (EA) approach to the selection of material designs with optimised functionality. Microscopic modelling will provide the relationship between chemical and physical structure and the fitness parameters to be optimized during the EA, while multi-scale modelling and comparison with experiment allow evaluation of the proposed structures. As examples of technologically relevant material systems, we will first study membranes for molecular separations, including small molecule separation and desalination. The methods will then be adapted to other applications, specifically porous polymer materials for photocatalysis and optical sensing, and conjugated polymer based ion sensors. An ancillary aim is to evaluate the EA approach as a tool for materials discovery.