(18c) Towards Automated and Comprehensive Exploration of the Methane Physisorption Space Using an Iterative Mutation Search Algorithm

In materials design, we often hope to find the “best” material for a given task, but sometimes it can be difficult to know the theoretical performance ceiling that defines "best" for a given application. Attempting to find such limits can be non-trivial: you need to test a perfectly diverse set of materials that both a) encompasses the entire range of what is structurally possible, and b) has properties that vary continuously across that range to establish meaningful trends. In this talk we discuss a novel approach to this challenge where we relax the constraint that materials must be physically realizable. We use randomly generated Lennard-Jones crystals to quickly illuminate limits and the structure-property space. We will discuss our new iterative-mutation search algorithms using methane physisorption as a test problem. We demonstrate that this method compares well to prior data on real materials and due its speed and flexibility, can provide quick insight into theoretical limits and to why the limits exist.