Towards a Predictive First-Principles Description of Solid Molecular Hydrogen with Density-Functional Theory

TL;DR

This paper investigates the effects of quantum and thermal approximations in density-functional theory on predicting the properties of solid molecular hydrogen, highlighting the importance of quantum effects and the limitations of common approximations.

Contribution

It provides a comprehensive analysis of how nuclear quantum effects and exchange-correlation functionals influence the predicted properties and phase diagram of solid hydrogen from first principles.

Findings

Nuclear quantum effects significantly impact intramolecular properties.

Thermal and quantum effects drastically alter optical property predictions.

Quasi-harmonic approximation achieves ~10 meV accuracy but is insufficient for reliable phase diagram predictions.

Abstract

We examine the influence of the main approximations employed in density-functional theory descriptions of the solid phase of molecular hydrogen near dissociation. We consider the importance of nuclear quantum effects on equilibrium properties and find that they strongly influence intramolecular properties, such as bond fluctuations and stability. We demonstrate that the combination of both thermal and quantum effects make a drastic change to the predicted optical properties of the molecular solid, suggesting a limited value to dynamical, e.g., finite-temperature, predictions based on classical ions and static crystals. We also consider the influence of the chosen exchange--correlation density functional on the predicted properties of hydrogen, in particular, the pressure dependence of the band gap and the zero-point energy. Finally, we use our simulations to make an assessment of the accuracy of typically employed approximations to the calculation of the Gibbs free energy of the solid, namely the quasi-harmonic approximation for solids. We find that, while the approximation is capable of producing free energies with an accuracy of ~10 meV, this is not enough to make reliable predictions of the phase diagram of hydrogen from first-principles due to the small free energy differences seen between several potential structures for the solid; direct free energy calculations for quantum protons are required in order to make definite predictions.