Bulk Ice Classification Using Topological Network Criteria

The trajectory file for this example is here on figshare. The trajectory file details a portion of a long run of 4096 molecules of mW water, which have undergone crystallization. In this example, DDCs (Double-diamond Cages), HCs (Hexagonal Cages), and mixed rings are identified inside the largest ice cluster. On running the example, an output top-level directory named runOne is created.

Steps to Run the Example

In order to run this example, without making any changes to the example files, please follow the steps below.

  • Download the LAMMPS trajectory file from here on figshare. Copy the downloaded trajectory file, entitled nucleation.lammpstrj, into the traj folder inside the top-level directory input. Alternatively, you could change the path to the trajectory file in the conf.yaml file:
    trajectory: "path/to/trajectory/file"
  • You can obtain the other input files required from example_lua/bulkTopologicalCriterion folder. Copy the contents of the bulkTopologicalCriterion into the top-level lua_inputs directory.
  • You can change the frames to be analyzed by updating the options in the vars.lua file. The starting and ending frames are inclusive, starting from 1 onwards, irrespective of the timestep number.
  • A custom volume slice can also be defined in the vars.lua file.
  • The functions.lua file actually contains the Lua functions which interface with the C++ backend.

Analyzing the Output

Inside the output directory, a file called clusterStats.dat contains the cluster statistics for each frame. Inside the bulkTopo directory, cageData.dat contains the number of HCs, DDCs and mixed rings for each frame. Inside runOne/bulkTopo/dataFiles, LAMMPS data files which are numbered according to the frame number are created. These data files can be visualized in OVITO or VMD, although OVITO is recommended for optimal type visualization.

References

  1. Haji-Akbari, A., & Debenedetti, P. G. (2015). Direct calculation of ice homogeneous nucleation rate for a molecular model of water. Proceedings of the National Academy of Sciences, 112(34), 10582-10588. doi:10.1073/pnas.1509267112