Colin says the supercell was originally an attempt to improve accuracy via averaging. It might not be necessary.
the questions to answer are:
for really small unit cells (e.g. AB graphene), does LAMMPS use enough images for van der walls? does the structure converge into more-or-less the same result as it would without a supercell?
Colin says the supercell was originally an attempt to improve accuracy via averaging. It might not be necessary.
the questions to answer are: