Monte Carlo potts and Phase field

Hi,

I have very recently started with Grain growth modeling and am currently learning the various methods.

I have managed to reproduce the results of Chen et al., PRB, 1994 using the phase field model. I was now trying

to focus my attention on reproducing the result using the Monte Carlo Potts model, essentially the formalism described

in Tikare et al., Acta Mat. 1999. using a python program. The code is still very rudimentary and am sure there is a lot

of optimization effort needed. With the Monte carlo potts (2D), even after reaching a large number of steps, i do not see

a structure even closely resembling grains. My question is

1.) is there a preferred strategy to initialize the system. Does it need to have a preferred distribution for it to converge faster.

I am currently using a uniform distribution of "Orientations". Am also attaching the file. The strategy is as I understood it to be described in the paper.

I am not sure where the error is.

Any help regarding 1.) would be greatly appreciated.

Sincerely,

Multi_scaler