BOMC is a Monte Carlo based generator of defect-free amorphous samples of Si and SiO2, written by Sebastian von Alfthan (galfthan at iki dot fi).
If you publish research where Bomc has been utilized then you are kindly asked to cite: "Realistic models of amorphous silica: A comparative study of different potentials", S. von Alfthan, A. Kuronen, and K. Kaski, Phys. Rev. B 68, 073203 (2003). (pdf)
The code was developed as a personal research code in 2002 - 2003, and thus there is little documentation. It is best to look at the code, and the two example cases, to understand how it operates. The article mentioned above, and my thesis also describes in some details the method. A more detailed discussion of the optimized algorithms implemented in the code is available from epaps: Optimization of the WWW-method
On a standard unix machine the code should compile by running make in the src folder. One can edit the Makefile if it does not compile.
There is also a parallel version that one can enable in the makefile. It performs in parallel multiple trial steps, and picks randomly among the accepted ones. At higher temperatures some accepted steps are thus not taken, and you may wish to slightly increase anneal length in main.par to balance this out if a larger numbers of processes are in use.
When you run the program you should have three .par files in your folder where all settings are defined:
- main.par General settings, the testcases have some comments in them which should explain most options
- cg.par Options for the optimization algorithms. There is a number of heuristics, some of which are explained in the article. One probably do not need to touch this too much.
- wwwPot.par Potential model parameters
When the code is run, no parameters are given on the command line:
./bomc