I have a 2D triangular lattice model with harmonic interactions I’d like to try running in GULP for phonon calculations. The unusual thing about my model is that it has a mixture of positive and negative k_2 and r_0 parameters. (k_3=k_4=0). Apart from dynamical instabilities I may encounter, is there any issue using negative parameter values in my 2-body interaction?
If not, is there an example input file available that is reasonably close to what I want to do?
Hi Tyler,
GULP should allow negative values for k_2 and r_0 in the harmonic potential. Of course this would lead to potentially unstable results for frequencies, as you say, but the calculation should be performed regardless. With a 2D triangular lattice and positive/negative values, this almost sounds like an Ising model. If so, then note that recent versions of GULP already support the idea of a spin on a lattice for Monte Carlo simulations of an Ising model. If this is just coincidence then ignore this info.
Regards,
Julian