These are preliminary reports that have not been peer-reviewed. They should not be regarded as conclusive, guide clinical practice/health-related behavior, or be reported in news media as established information. For more information, please see our FAQs.
Preprints are manuscripts made publicly available before they have been submitted for formal peer review and publication. They might contain new research findings or data. Preprints can be a draft or final version of an author's research but must not have been accepted for publication at the time of submission.
We show here that machine learning is a powerful new tool for predicting the elastic
response of zeolites. We built our machine learning approach relying on geometric features
only, which are related to local geometry, structure and porosity of a zeolite, to
predict bulk and shear moduli of zeolites with an accuracy exceeding that of force field
approaches. The development of this model has illustrated clear correlations between
characteristic features of a zeolite and elastic moduli providing exceptional insight into
the mechanics of zeolitic frameworks. Finally, we employ this methodology to predict
the elastic response of 590 448 hypothetical zeolites, and the results of this massive
database provide clear evidence to stability trends in porous materials.