High-throughput screening of tribological properties of monolayer films using molecular dynamics and machine learning

Abstract

Monolayer films have shown promise as a lubricating layer to reduce friction and wear of mechanical devices with separations on the nanoscale. These films have a vast design space with many tunable properties that can affect their tribological effectiveness. For example, terminal group chemistry, film composition, and backbone chemistry can all lead to films with significantly different tribological properties. This design space, however, is very difficult to explore without a combinatorial approach and an automatable, reproducible, and extensible workflow to screen for promising candidate films. Using the Molecular Simulation Design Framework (MoSDeF), a combinatorial screening study was performed to explore 9747 unique monolayer films (116 964 total simulations) and a machine learning (ML) model using a random forest regressor, an ensemble learning technique, to explore the role of terminal group chemistry and its effect on tribological effectiveness. The most promising films were found to contain small terminal groups such as cyano and ethylene. The ML model was subsequently applied to screen terminal group candidates identified from the ChEMBL small molecule library. Approximately 193 131 unique film candidates were screened with approximately a five order of magnitude speed-up in analysis compared to simulation alone. The ML model was thus able to be used as a predictive tool to greatly speed up the initial screening of promising candidate films for future simulation studies, suggesting that computational screening in combination with ML can greatly increase the throughput in combinatorial approaches to generate in silico data and then train ML models in a controlled, self-consistent fashion.

A review of recent advances and applications of machine learning in tribology

“…To that extent only few studies have shown the potential to explore this data wisely and efficiently in the field of tribology. [315][316][317][318] Table S8 (ESI †) tabulates these four articles describing their tribological investigations on different materials using ML algorithms in MD and DFT.…”

“…Besides, the RF models trained on a limited dataset could still give meaningful predictions for monolayer designs outside the training set, which suggested a good transferability. 315 Kadupitiya et al conducted NEMD simulations and applied PCA on the data obtained from these simulations. 316 Specifically, a study was conducted to investigate the rheological properties of squalane under elastohydrodynamic lubrication (EHL) conditions, the lubricant under large pressure of more than 500 MPa and strain rate greater than 10 5 s À1 , which is difficult to achieve in an experiment at the same time.…”

“…It should be noted that although the use of ML has been extensively examined in experimental tribological studies, only four papers, [315][316][317][318] to the best of our knowledge, have implemented ML in computational modeling studies. The combinatorial approach of ML with MD simulations demonstrated the efficient exploration of MD-obtained data.…”

“…Quach et al 315 applied a Python-based Molecular Simulation and Design Framework (MOSDeF) 319 with a signac data and workflow management framework 320,321 to perform high-throughput screening of CoF and adhesion force for about 10 000 monolayer films. 321 These two properties are then used as the expected labels for the ML model using the RF algorithm.…”

“…Besides, the RF models trained on a limited dataset could still give meaningful predictions for monolayer designs outside the training set, which suggested a good transferability. 315…”