qm_playground is a Python package, that enables simulation of simple 1D and 2D quantum mechanical bound state and scattering problems using a number of different adiabatic and nonadiabatic techniques.
- Analytical potentials in 1D and 2D
- Wavepacket propagation
- Classical molecular dynamics
- Ring Polymer molecular dynamics
- Nonadiabatic wavepacket propagation
- Interactive visualization
Execution of a qm_playground simulation involves initialising a model,
calling the model's run() or solve() methods, then visualising the
result.
Depending on the type of simulation being performed this step can vary
slightly. Creating an instance of the Model class requires it be provided
with all the necessary components for the simulation. The essential components
are a system, a potential, a mode. For a time-dependent simulation one must
also provide an integrator.
A brief look at the examples should make clear how this works.
Once the model is set up, this is the easy part. The run() method propagates
the system through time and writes the results to the model.data attribute.
The solve() method is used only for the solution of the time-independent
Schrodinger equation for wave models and similarly writes the output to
model.data.
After the simulation has finished, the contents of model.data are written to
a pickled output file with the .end extension. The name of this file defaults
to simulation.end but can be changed by providing the model with a name
keyword argument.
Provided in the
visualisation directory is a file called plot.py. This can be run from the
command line as bokeh serve --show plot.py. A browser window will open that
allows the user to provide the path to their .end file. By clicking the
button a visualisation of the trajectory will appear. This works currently for
wave, traj and rpmd modes only currently.
Another option is to use the matplotlib based visualisations available in
qmp.tools.visualizations.
This code is licensed under the GNU General Public License, version 3.