On this page you will find more information regarding some

of our current research interests:




Chemical reactions occur in all states matter ranging from the gas phase of interstellar clouds to cytosol in living cells. Understanding and predicting chemical reactivity enables the rapid and focused design of e.g. efficient drugs, optimal catalysts and desired / undesired reactive pathways. We seek to overcome the cost of ab-initio calculations for reaction rate constants and networks of reaction rate constants by training and using machine learning algorithms to predict rate constants. 


Relevant publications

E. Komp and S. Valleau, "Machine learning quantum reaction rate constants" J. Phys. Chem. A.: 124:8607 (2020)



We are interested in understanding the kinetics of reaction cycles and networks using the combination of theoretical chemistry with biophysics methods using a bottom up strategy which combines atomistic models to flux models with machine learning. Here below we show images from our publications related to the calculation of reaction rate constants in the ground state both classically and quantum mechanically.

Example of a cycle of reactions

Ni(100) ground state potential energy surface which was used to study hydrogen diffusion

Computed minimum energy pathways for reactions when including solvent explicitly

Example of a cycle of reactions


Relevant publications

S. Valleau and T. Martínez., "Reaction dynamics of cyanohydrins with hydrosulfide in water", arXiv:1806.08841 (2018)

S. Mandrà, S. Valleau and M. Ceotto, "Deep Nuclear Resonant Tunneling Thermal Rate Constant Calculations", International Journal of Quantum Chemistry: 113,1722 (2013)

S. Valleau, "A Quantum Instanton study of the diffusion of hydrogen and its isotopes on Ni(100)" (2010) (download).