Discovery of Record-Breaking Metal-Organic Frameworks for Methane Storage using Evolutionary Algorithm and Machine Learning

In the past decade, there has been a rise in a number of computational screening works to facilitate finding optimal metal-organic frameworks (MOF) for variety of different applications. Unfortunately, most of these screening works are limited to its initial set of materials and result in brute-force type of a screening approach. In this work, we present a systematic strategy that can find materials with desired property from an extremely diverse and large MOF set of over 100 trillion possible MOFs using machine learning and evolutionary algorithm. It is demonstrated that our algorithm can discover 964 MOFs with methane working capacity over 200 cm³ cm⁻³ and 96 MOFs with methane working capacity over 208 cm³ cm⁻³, which is the current world record. We believe that this methodology can facilitate a new type of a screening approach that takes advantage of the modular nature in MOFs, and can readily be extended to other important applications as well.