| Description |
Quantum nuclear effects plays a central role in determining properties of systems that contain light nuclei. For instance, the large deviation of the heat capacity of a solid from the Delong’s Petit limit and of the particle momentum distribution from the Maxwell-Boltzmann behaviour are direct manifestations of the quantum nature of nuclei. While these effects can be accurately modelled in atomistic simulations by employing the imaginary time path integral (PI) technique [1], the high computational cost of running PI simulations has prevented their widespread use. In this talk, I will introduce molecular dynamics based methods [2, 3, 4, 5] that substantially reduce the computational cost of PI simulations and their implementation in an open source software i-PI. [6] Going beyond benchmarks, I will demonstrate the relevance of these advances by studying different properties and classes of materials – such as the proton momentum distribution in water that relates to the local structure of protons in ice [4] and facilitates interpretation of complex Deep Inelastic Neutron scattering experiments, quantum effects that facilitate isotope separation in porous organic cages [7], tuning of thermal properties of metal-organic cages loaded with greenhouse gases [5], and quantitative estimation of quantum mechanical effects that stabilize pharmaceutically active molecular crystals [8] – at a fraction of the computational cost if using conventional techniques.
References
|