"Computer simulations of matter at the atomic level provide insights complementary to experiments in a number of different physical, chemical and biological problems. The quantum nature of nuclei is usually disregarded, because of the computational burden inherent in an accurate treatment. However, these nuclear quantum effects have important, qualitative consequences, in particular when light atoms such as hydrogen are present. Moreover, inelastic neutron scattering techniques have been developed in the past years, which probe experimentally the quantum distribution of proton velocities, and would benefit from the availability of inexpensive modeling tools.
Here we propose the development of a new methodology for treating nuclear quantum effects, which reduces computational effort without sacrificing accuracy by combining in a unified formalism path integral molecular dynamics and the ""quantum thermostat"", based on correlated-noise stochastic equations.
We expect in this way to be able to treat routinely nuclear quantum effects together with an accurate, first-principles evaluation of inter-atomic forces, while so far this has only been attempted occasionally and with great computational effort.
These theoretical developments will be accompanied by applications to problems of fundamental and technological relevance. After having demonstrated the reliability of our approach on a few simple, controlled examples, we will tackle the problem of ab-initio water, in particular with reference to recent inelastic neutron scattering experiments performed at ISIS.
At a later stage we will choose a more complex problem to demonstrate our approach. Among the possible applications, hydrogen-storage materials are particularly promising, also considering the opportunity of a collaboration with the experimental group of Prof. Bill David, who has been performing leading-edge experiments on complex hydrides in the past few years at the Rutherford Appleton Laboratories in Oxfordshire."
Fields of science
Call for proposal
See other projects for this call