In silico veritas.

Abstract

Innovations in computational nanoscience have traditionally come in conjunction with experimental innovations, but uncertainty often surrounds the trustworthiness of in silico studies. While the accuracy of simulations has been improving every year, considerably less attention has focused on dealing with increasing complexity, which may be the source of concern. Creating more realistic virtual experiments (without sacrificing theoretical and numerical accuracy) remains challenging, particularly when we are confronted with the polydispersivity characteristic of extra silico samples. Fortunately, there are various theoretical methods that can be used in conjunction with first-principles simulations, not the least of which are the statistical tools and techniques promised by the emerging fields of materials informatics and data-driven sciences.