On the Ewald summation of Gaussian charges for the simulation of metallic surfaces

► Gaussian charges influence only real-space Ewald terms. ► k = 0 divergences are identical between Gaussian and point charges. ► Continuum results arise naturally from representing a metal by Gaussian charges. ► Classical model convergence to long-range continuum results insensitive to Gaussian spread. ► Comparisons with continuum theory may not be the optimal way to parameterise Gaussian spreads of a Siepmann/Sprik model.