Challenge: Given that lubricants are mixtures of predominantly alkanes, it is unclear whether contemporary lubricant formulations are the most optimal. Lubricants contain hydrocarbon molecules, therefore predicting the properties of hydrocarbons facilitates the development of base oils.
Solution: Deep learning algorithm Alchemite can exploit property-property correlations to predict the physical properties of alkanes. Alchemite inputs the molecular structure of alkanes to predict the boiling point, heat capacity, and vapor pressure as a function of temperature. The results reproduced by this algorithm are significantly more accurate and consistent than those reproduced by other methods.
Outcome: By combining sparse experimental data with molecular dynamics simulations to predict physical properties of alkanes, Alchemite accelerated the identification of optimal hydrocarbons tenfold. Alchemite also accurately estimated intractable properties including density and shear viscosity and produced results that were five times more accurate and consistent than those reproduced by other methods.
Swipe to navigate articles