Calculates electrical conductivity from molecular simulations using root-mean square displacements and optionally splits it into contributions from the individual species, and/or decomposes it spatially.
This is similar to what has been described here: http://aip.scitation.org/doi/abs/10.1063/1.4890741
Those guys were, to the best of my knowledge, the first ones who did a spatial decomposition analysis of the electrical conductivity.
There is not much documentation. The directory data contains three example inputs that cover what is possible with canal at the moment
All necessary files are in the src subdirectory, including a makefile that might need changing depending on what compiler you prefer For gcc you should be fine by simple typing make.
You need to create an input file that contains a couple of keywords and values in the style:
keyword = value1 value2
The location of the input file has to be passed as first command-line argument.
All keywords only take one value, except for 'spatial', which takes one to three values. The keywords are:
mandatory:
- xcom/ycom/zcom: x,y,z coordinate files for all fragments' centers of mass (in Angstrom)
the format is one line per snapshot, with values for each fragment separated by spaces
- chgs: charges of the fragments
the format is one line per fragment
optional:
- cell: cell dimensions (if a spatial decomposition is what you're looking for)
- nrestart: number of restart points along the trajectory
- fitoffset: where to start linear fit (fraction of data, default 0.1)
- fitlength: length of data for linear fit (fraction of data, default 0.9)
fitoffset + fitlength should not exceed 1.0, otherwise accessing unallocated memory
needed for correct conductivty calculation:
- temp: temperature (in K)
- timestep: timestep (in fs)
- avvol: average volume (in Angstrom^3)
needed for splitting conductivity into parts from anion-anion, anion-cation, cation-cation:
- split 1 number, turned on if != 0
needed for spatial decomposition:
- spatial: 3 numbers:
number of distance bins <-- if > 1 analysis turned on
starting distance
bin width