Code-Monte-Carlo-SRO

This repository includes scripts/codes that we used to calculate temperature-dependent short-range-order parameters for complex concentrated alloys. The open-source code is free to distribute and share under the MIT license.

This is mainly a Python code tested on Python 3.6 with Mac OS 11.0.1. It is expected to run in different operation systems as long as Python3 works.

Instruction

Calculate the contribution of the lattice distortion to the Hamiltonian

cd Lattice-Green-Function-NiCoCr
python LGF.py

Metroplis MC simulations

Once the Hamiltonian is ready, put the interaction parameters in the file Metroplis_MC_simulations/eci.out Then submit the simulation as a job to a cluster of supercomputer.

cd Metroplis_MC_simulations
sbatch sb_mc.sh

Once the job finishes, change the format of restart.struct to POSCAR. Put the file in another folder calculate_SRO_k-space. The order parameters in the reciprocal space can be calculated by

cd calculate_SRO_k-space
sbatch sb_sro_k.sh

or submit it as a job to run it remotely in a cluster.

python sro_k_by_R.py restart.struct 0

The last parameter "0" means starting the calculation from scratch and then plotting a figure based on the result. If "1" is given, the code will search for the calculated short-range-order parameter in reciprocal and plot the figure based on the data. The calculation can take a while.

Wang-Landau simulations

Download the Wang-Landau Monte-Carlo code (see link below). Copy the interaction parameters (eci.out), structure file (init.struct) and other related files to a new folder or examples/example-ternary-NiCoCr/. Then submit it as a job to a cluster.

cd examples/example-ternary-NiCoCr/
sbatch qsub.sh (This is for slurm manager; need to adjust this to specific job manager)

To accelerate the calculation, we divide the whole calculation into multiple ones and each one correspond to a specific energy range. This can be done with the following command, but the energy range (minimum of ordered and maximum of random states) need to be provided.

python gen_energy_array.py

One the multiple jobs are done, all important information is written to restart.log.wlce, including the density of state. Then copy the output files restart.log.wlce to Wang-Landau-data-processing.

First, use the code to generate a combined cdos.dat file for the density of state;

cd Wang-Landau-data-processing
python combine-dos. py

Then use dos.py to test the convergence and calculate the thermodynamic properties, such as specific heat Cv.

python dos.py

We can also plot the density of states (dos) with

python plot_dos.py

Other related codes/tools

The drivers of the Metroplis Monte Carlo and Wang-Landau Monte Carlo can be found in other repositories.

Metropolis Monte-Carlo code

For Metropolis, please refer to https://github.com/peizong/Simulated-Annealing-Metropolis-MC-.

Wang-Landau Monte-Carlo code

For Wang-Landau, please refer to https://github.com/peizong/WL-CE.