[Calculation Example Released] Data-Driven Method for Discovering Low Adsorption Energy Molecules on Si Surfaces

We have published a new Matlantis application example on the calculation examples page.
In this case study, Bayesian optimization and universal machine learning potential (PFP) are combined to search for molecules exhibiting low adsorption energy for use as chemicals in semiconductor manufacturing processes. Bayesian optimization makes it possible to narrow down candidate molecules with 39 times more efficiency compared to random sampling.

Read the full article here: Data-Driven Method for Discovering Low Adsorption Energy Molecules on Si Surfaces