Polymer: transient model of solid polymer fuel regression

Python code for a one-dimensional, transient model of a solid polymer fuel (e.g., polyoxymethylene, POM) heated at its surface, as in a hybrid rocket combustion chamber. The model couples heat conduction, melting, depolymerization kinetics, evaporation of the decomposition products, and a moving (regressing) surface. It extends the steady-state analytical model in SS_Model and the counterflow regression-rate method in Counterflow_Polymer.

Developed by Wendi Dong in Prof. Hai Wang's group, Mechanical Engineering, Stanford University. This is research code; the cases in the scripts are the ones run during development.

Contents

File	Description
src/transient.py	Transient: conduction through the solid and melt layers with a beta-scission heat sink in the melt and a regressing surface. Central differences in space, classical RK4 in time (method of lines). cal_steady_state() evaluates the steady-state analytical solution (surface temperature, melt-layer thickness, regression rate) for comparison; main_TGA() and main_TGA_top_heat() simulate thermogravimetric analysis (TGA) heating.
src/evaporation.py	Evaporation: evaporation of multicomponent decomposition products (e.g., styrene oligomers) from a heated sample, with vapor pressures from the Nannoolal correlation and a Hertz-Knudsen evaporation flux.
src/integrated.py	Polymer: the integrated model. Conduction, melting, lumped depolymerization kinetics, multicomponent products with evaporation, and a time-dependent surface heat flux; checkpoints allow long runs to restart. case42() ... case60() are the development cases.
src/counterflow_SS.py	Couples the condensed phase to a Cantera counterflow diffusion flame through a surface energy balance (conduction, multicomponent Stefan-Maxwell diffusion, Soret effect, radiation) and solves for the surface temperature by bisection, giving the regression rate as a function of oxidizer velocity.

The solver was verified against the analytical solutions and compared with TGA and counterflow-flame measurements (the comparison data are not included here).

Data (src/data/)

• POM.json: POM properties used by case58() ... case60().
• polymer_evaporation.xlsx: boiling points (several estimation methods), Nannoolal parameters, molecular weights, and diffusion coefficients of decomposition products.
• FFCM2_CH2O.yaml: gas-phase kinetics, the formaldehyde sub-model of FFCM-2 (13 species, 47 reactions), used by counterflow_SS.py.
• TdependentStoicCoeff.xlsx: reference data, temperature-dependent product distribution of polystyrene pyrolysis (literature measurements and the CRECK model); not read by the scripts.

Usage

pip install -r requirements.txt
python src/integrated.py      # runs case60(); about 2 minutes on a laptop

To run another case, change the call in the if __name__ == "__main__": block at the end of the script. transient.py, evaporation.py, and counterflow_SS.py are run the same way.

Inputs are read from src/data/; results are written to src/output/<model>/<case>/ (git-ignored): case_dict.json with all inputs, NumPy arrays of the time history (e.g., T_mat.npy temperature field, phase_mat.npy phase field, dL_arr.npy regressed thickness), and log.txt. Saving animations (Transient.plot_box()) requires ffmpeg on the PATH.

Tested with Python 3.10, NumPy 1.24, SciPy 1.10, pandas 2.1, Matplotlib 3.7, and Cantera 2.6.