Inverse Design Method with Enhanced Sampling for Complex Open Crystals: Application to Novel Zeolite Self-Assembly in a Coarse-Grained Model

TL;DR

This paper presents an innovative inverse design workflow with enhanced sampling for complex zeolite frameworks, enabling efficient reproduction, optimization, and discovery of novel structures within a coarse-grained model.

Contribution

The authors develop a new inverse design method combining evolutionary optimization and enhanced sampling to target complex zeolite structures in a coarse-grained simulation.

Findings

Successfully reproduces known zeolites Z1 and SGT

Discovers new frameworks like Z5

Optimizes parameters for SOD and CFI phases

Abstract

Optimizing the synthesis of zeolites and exploring novel frameworks offer pivotal opportunities and challenges in materials design. While inverse design proves highly effective for simpler crystals, its application to intricate structures like zeolites poses severe challenges. Here, we introduce an innovative inverse design workflow tailored to efficiently reproduce target zeolite frameworks in a binary coarse-grained model using enhanced sampling molecular dynamics simulations. This workflow integrates an evolutionary parameter optimization strategy with a variant of the seeding approach. Using this method, we successfully reproduce Z1 and SGT zeolites, and Type-I clathrates, find new optimal parameters for known phases, such as the SOD and CFI, and even discover novel frameworks, such as Z5. This is done within a simple coarse-grained model for a tetrahedra-forming component and a structure-directing agent. Our methodology not only enables the screening of synthesis protocols but also facilitates the discovery of hypothetical zeolites.