Swap Monte Carlo algorithm for diatomic molecules

In recent years the swap Monte Carlo algorithm has led to remarkable progress in equilibrating supercooled model liquids at low temperatures. Applications have so far been limited to systems composed of spherical particles, however, whereas most real-world supercooled liquids are molecular. We introduce here a simple size-polydisperse molecular model that allows for efficient thermal equilibration in silico with the swap Monte Carlo method, resulting in an estimated speedup of 10³--10⁶ at moderate polydispersity (5%--10%). The model exhibits little difference between size-resolved orientational time-autocorrelation functions.