Quasi-One-Dimensional Ice Classification¶
The trajectory file for this example is here on figshare. The trajectory file details a portion of a long run of 750 molecules of TIP4P/2005 water, inside a smooth featureless (13,0) nanotube. This nanotube approximates a zigzag carbon nanotube. At 240 K, the temperature of the simulation, the ice nanotube is primarily composed of tetragonal blocks. In this example, prism blocks are identified according to a topological network criterion for confined ice [1]. 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 dump-240-square.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/iceNanotube folder. Copy the contents of the iceNanotube 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 directory called topoINT is created. Inside the topoINT directory, a file called nPrisms.dat contains the number of prism blocks and the corresponding height% [2], for each frame. Inside runOne/topoINT/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¶
- Goswami, A., & Singh, J. K. (2020). A general topological network criterion for exploring the structure of icy nanoribbons and monolayers. Physical Chemistry Chemical Physics. doi:10.1039/c9cp04902a
- Goswami, R.; Goswami, A.; Singh, J. K. (2019). "d-SEAMS: Deferred Structural Elucidation Analysis for Molecular Simulations". arXiv:1909.09830 [physics.comp-ph].