This is an implementation of the Gaussian patchy disc model. The code was built based on the VMMC library of Lester Hedges.
- cmake >= 3.11
- nlohmann json library
git clone git@github.com:mosayebi/PatchyDisc.git
cd PatchyDisc
mkdir build
cd build
cmake ..
make
A successfull compilation will create the executable file build/PatchyDisc.
cd examples
../build/PatchyDisc input_mixture.json
This will run a simulation using the parameters configureed in the input_mixture.json.
Below is the contents of a sample input file for simulating a mixture of tri-valent and bi-valent patchy discs.
{
"initialisation" :
{
"mode": "from_init_conf",
"from_random_conf":
{
"types": 2,
"particle_numbers" : [{"type": 0, "N":300}, {"type": 1, "N":100}],
"box": [40.0, 40.0]
},
"from_init_conf":
{
"init_conf": "last_conf.xyz",
"restart_step_counter": false
}
},
"topology" :
{
"patches": [
{"type": 0, "nPatches": 3, "angles" : [0.0, 120, 240.0]},
{"type": 1, "nPatches": 2, "angles" : [0.0, 120.0]}
],
"pair_coeff" : [
{"type1":0, "type2":0, "epsilon":15.0, "delta":1.0, "sigma":1.0, "sigma_p": 0.3, "rcut": 1.5},
{"type1":0, "type2":1, "epsilon":5.0, "delta":1.0, "sigma":1.0, "sigma_p": 0.3, "rcut": 1.5},
{"type1":1, "type2":0, "epsilon":5.0, "delta":1.0, "sigma":1.0, "sigma_p": 0.3, "rcut": 1.5},
{"type1":1, "type2":1, "epsilon":15.0, "delta":1.0, "sigma":1.0, "sigma_p": 0.3, "rcut": 1.5}
]
},
"simulation parameters":
{
"seed": 42,
"interaction": "GaussianPatchyDisc",
"trajectory": "trajectory.xyz",
"last_conf": "last_conf.xyz",
"log_file" : "log.dat",
"output_every": 1000,
"sweeps": 1e5
}
}Currently, the following interactions are implemented
-
GaussianPatchyDisc-- The model is described in this reference. -
GaussianPatchyDiscHR-- Same as theGaussianPatchyDisc, but with a hard core repulsion when particles overlap (i.e. when$r_{ij} \lt \sigma_{ij}$ ). -
GaussianPatchyDiscHRSW-- Same as theGaussianPatchyDiscHR, but with a Square-Well poential when particles are within the interaction range (i.e. when$r_{ij} \le {r_{cut}}_{ij}$ ).
The trajectory is saved in XYZ format. You can vissualise it in the visualaisation package of your choice. However, I recommend using ovito.
When importing the trajectory in ovito, please choose the follwing order for the column mappings: Particle type, Position X, Position Y, Orientation Z, Orientation W. You can then load particle shape STL files saved in the UTILS/particle_shapes/ if you are simulating a standard patchy disc, or make your own particle shape using the python script provided.
To load a particle shape in ovito under the "Data source" panel click on "Particle Types" and choose a particle type. The "Load shape..." button can then be found in the bottom of the panel.