Atomistic Nodal Approach for Obtaining Thermal Conductance

Python Code for Processing LAMMPS Data generated for two data types: LAMMPS trajectory file and LAMMPS dump files. Subsequently, the code involves post-processing of the data to give interfacial heat flux, temperature discontinuity and ultimately interfacial conductance using the atomistic nodal approach.

LAMMPS Package used to generate Data:

• LAMMPS 19MAR20

Python Pacakages Needed:

• Scipy 1.6.0
• Dask 2.3.0
• Numpy 1.6.1
• SKlearn 0.24.1
• Pandas 1.1.3

Data Visualisation

• Ovito 3.5.2

Part 1: Processing the Data

Two data types are generated from the sample LAMMPS script given, which are processed using the DASK and Pandas in the raw_data_treatment folder. The first is the cuboid grid-based DUMPS data format, which is used to process the fluid's thermodynamics and positional information. This is done through dump_file_averaging.py. The second is the per-atom averaging conducted for the relatively stationary atoms of the NP, which are treated as nodesm whose physical properties are averaged on a per atom basis for the simulation. To process the data from the simulation, traj_averaging.py can be used

Part 2: Interpreting the Data

To find the thermal conductance of a NP, 2 quantities are needed: Interfacial heat flux and temperature disconituity, shown in the data_processing folder

1. Interfacial Heat Flux

• Volume: Volume can be calculated at a per-atom level using the Voronoi volume calculated by LAMMPS
• Direction of Heat Flux. Heat flux is projected along the temperature gradient.
• This Temperature gradient can be calculated using temperature_gradient_interpolation.py, using interpolation method in Scipy LinearNDInterpolator library

2. Temperature Discontinutiy

• The Temperature discontinuity can be obtained from constructing a C-KD tree for rapid querying of fluid grid information within a radius of interest (normally 1st peak in g(r)). This is done using cdktree_dT_discontinuity.py

Copyright and Terms of Use

This code and workflow can be used for strictly non-commercial purposes only. When using this code, do cite this Github page in the references section.