Reduced variance analysis of molecular dynamics simulations by linear combination of estimators.

Building upon recent developments of force-based estimators with a reduced variance for the computation of densities, radial distribution functions, or local transport properties from molecular simulations, we show that the variance can be further reduced by considering optimal linear combinations of such estimators. This control variates approach, well known in statistics and already used in other branches of computational physics, has been comparatively much less exploited in molecular simulations. We illustrate this idea on the radial distribution function and the one-dimensional density of a bulk and confined Lennard-Jones fluid, where the optimal combination of estimators is determined for each distance or position, respectively. In addition to reducing the variance everywhere at virtually no additional cost, this approach cures an artifact of the initial force-based estimators, namely, small but non-zero values of the quantities in regions where they should vanish. Beyond the examples considered here, the present work highlights, more generally, the underexplored potential of control variates to estimate observables from molecular simulations.

Electron transfer within a reaction path model calibrated by constrained DFT calculations: application to mixed-valence organic compounds.

The quantum dynamics of electron transfer in mixed-valence organic compounds is investigated using a reaction path model calibrated by constrained density functional theory (cDFT). Constrained DFT is used to define diabatic states relevant for describing the electron transfer, to obtain equilibrium structures for each of these states and to estimate the electronic coupling between them. The harmonic analysis at the diabatic minima yields normal modes forming the dissipative bath coupled to the electronic states. In order to decrease the system-bath coupling, an effective one dimensional vibronic Hamiltonian is constructed by partitioning the modes into a linear reaction path which connects both equilibrium positions and a set of secondary vibrational modes, coupled to this reaction coordinate. Using this vibronic model Hamiltonian, dissipative quantum dynamics is carried out using Redfield theory, based on a spectral density which is determined from the cDFT results. In a first benchmark case, the model is applied to a series of mixed-valence organic compounds formed by two 1,4-dimethoxy-3-methylphenylene fragments linked by an increasing number of phenylene bridges. This allows us to examine the coherent electron transfer in extreme situations leading to a ground adiabatic state with or without a barrier and therefore to the trapping of the charge or to an easy delocalization.