There are many problems at the forefront of materials chemistry whose solution is stymied by its inherent complexity. Such problems are characterized by a rich landscape of parameters and processing variables that is combinatorially too large for either an experimental or a computational approach to solve through an exhaustive search. In such cases, the usual approach is an Edisonian trial-and-error approach, which inevitably leaves areas of parameter space unexplored. The problems that we have explored are also characterized by a scarcity of data, since the data are expensive to acquire, both experimentally and computationally. This makes it an ideal candidate to solve using a Bayesian optimization approach. So far, we have used a Bayesian approach to study several problems in materials chemistry. As a first test case, we have studied how to optimize the choice of solvent or additive to produce high quality thin films of lead-based hybrid organic-inorganic perovskites. We identified the best out of a list of around 12 solvents, and how to optimize the best mixed-halide system to solubilize the lead salt. We have also shown how we can use the Bayesian-generated ranked list of “best performers” to identify when a perovskite precursor is likely to form, which adversely affects thin film quality. We confirmed our ability to determine whether a precursor forms in six cases through complementary XRD experiments by the Estroff group at Cornell. We have also used the approach to identify all the polymorphs that can exist in a high performing p-type organic semiconductor, di-trimethylsilane benzothiophene (diTMS-BTBT). These polymorphs can exhibit differing ability to transport charge, opening up the possibility of using polymorphs as molecular “switches.” Our Bayesian approach identified two polymorphs that correlated well with new experimental observations by the Diao group at UIUC and we posited the existence of a third high-temperature polymorph as yet unconfirmed in the laboratory. In both these examples at least, the acceleration made possible by a Bayesian Gaussian Process approach allowed us to run expensive simulations for only a small fraction (~15-20 %) of the entire set of possible candidates (over 1000 structures). Our approach uses decision theory to choose an optimal material design that minimizes the number of experiments needed, without user intervention, and one that is broadly applicable across a range of chemical systems.

How Bayesian optimization approaches can transform our ability to reduce the complexity of materials chemistry problems

Macromolecular Chemistry: The Second Century

[Author Benefits](https://scimeetings.acs.org/?utm_source=pubs_content_marketing&utm_medium=website&utm_campaign=0320_MCF_NPI_Launch_Spring_Homepage&ref=pubs_content_marketing)

[How to use SciMeetings](https://storage.googleapis.com/dl.morressier.com/acs/SciMeetings-ACSSpring2020-HowToGuide_M3020-Final.pdf)

[Frequently Asked Questions](https://storage.googleapis.com/dl.morressier.com/acs/SciMtng_ACSNMSp2020_FAQ_M3120_Final.pdf)

ACS Spring 2020 National Meeting & Expo

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.