AI-Guided Design and Property Prediction for Zeolites and Nanoporous Materials

3 Machine Learning Search for Suitable Structure Directing Agents for the Synthesis of Beta (BEA) Zeolite Using Molecular Topology and Monte Carlo Techniques

María Gálvez-Llompart ^1,2 and German Sastre ¹

¹Instituto de Tecnologia Quimica (UPV-CSIC), Universidad Politecnica de Valencia; Valencia, Spain²Molecular Topology and Drug Design Unit, Department of Physical Chemistry, University of Valencia, Valencia, Spain

3.1 Introduction

Computer-aided material design is experiencing exponential growth with the introduction of high-performance computations. Brand new big data techniques can be used for the rational design of target materials, with machine learning (ML) being one of the most innovative and promising techniques [1–3]. Its strong predictive capability has been observed in areas such as molecular design, property optimization, and synthesis prediction [4–6]. Now is the time for materials science to unveil the potential of machine learning techniques [7].

ML techniques have already been used for the design of new materials, for instance, zeolites and related zeotype materials. Zeolites are crystalline, microporous materials extensively used in a variety of industrial applications such as catalysis, gas separation, and ion exchange. The industrial application of zeolites is determined by their structural and topological properties. As in many other areas of material design, the “trial-and-error” approach was common when searching for novel zeolites because the mechanisms ...