sndust is a collection of python scripts that model the formation of dust grains in a cooling, expanding vapor. Any environment that has time-series temperature and density data can be used as input, but it was orignally designed to study dust production in core-collapse supernovae (CCSNe).

sndust was developed at Los Alamos National Lab (LANL C19146)

The top-level file runtime_settings.json includes several parameters that control the integration of the ODE that models the physics.

• abs_tol, rel_tol control the tolerance of the integration step
• max_dt is the maximum allowed timestep
• *_every are the output options. screen_every is the very much the slowest

sndust needs numpy, scipy, numba, and mpi4py. A conda package is being put together to make this explicit