Lecture
Van Marum Colloquium: Coupled Cluster Molecular Dynamics for Aqueous Solutions With Machine Learning
- Dr. Nore Stolte (Ruhr- Universität Bochum)
- Date
- Thursday 5 November 2026
- Time
- Location
-
Gorlaeus Building
Einsteinweg 55
2333 CC Leiden - Room
- CM.3.23
Abstract
The behavior of interfacial water is to a large extent determined by the interactions of water with the interface or solute, yet these interactions are often misrepresented by force fields and even by density functional theory. We have developed a data-efficient strategy to train a machine learning interatomic potential for neutral aqueous solutes at coupled cluster accuracy, CCSD(T), using only small molecular clusters. Applying the methodology to aqueous solutions of toluene, our coupled cluster molecular dynamics (CCMD) simulations reveal distinct π-contacts of water at the aromatic ring and quantify the hydrophobic solvation shell, resolving subtle differences in solvation structure that are not captured consistently by common force fields and even hybrid density functional theory. The workflow provides a general route to CCSD(T)-quality condensed-phase simulations of aqueous solutions, opening the door to systematic, transferable studies of solute{solvent interactions.