Optimizing density-functional simulations for two-dimensional metals

Unlike covalent two-dimensional (2D) materials like graphene, 2D metals have nonlayered structures due to their nondirectional, metallic bonding. While experiments on are still scarce and challenging, density-functional theory (DFT) provides an ideal approach predict basic properties assist in design. However, DFT methods rarely been benchmarked against bonding at low dimensions. Therefore, identify optimal attributes for a desired accuracy, we systematically benchmark exchange-correlation functionals from LDA hybrids basis sets plane waves local with different pseudopotentials. With 1D chain, honeycomb, square, hexagonal, 3D bulk systems, compare the using bond lengths, cohesive energies, elastic constants, densities of states, computational costs. Although today most studies use waves, our comparisons reveal that often-used Perdew-Burke-Ernzerhof exchange correlation is quite sufficient purposes, while hybrid bring limited improvement compared greatly increased cost. These results ease demands generating data better interaction data-driven discoveries incorporating machine learning algorithms.