From Fortran to Python: A Conversation Across Generations of Scientific Computing

This project demonstrates how to integrate legacy Fortran code with Python to solve a system of ordinary differential equations (ODEs) modeling the thermal cracking of Ethylene Dichloride (EDC). We compare the performance of a compiled Fortran implementation with a pure Python version using the Runge-Kutta-Gill method.

This project accompanies a talk at PyOhio 2025.

Problem Overview

We model a two-step reaction system:

EDC → MVC + HCl         (rate constant k1 = 1.0)
MVC → HCl               (rate constant k2 = 0.5)

Project Structure

.
├── cracker.cpython-312-darwin.so       # Compiled Fortran module (Python 3.12)
├── cracker.cpython-313-darwin.so       # Compiled Fortran module (Python 3.13)
├── SummaryResults-FortranvsPython.pdf  # PDF report for performance results (summarized)
├── fortran-code.f90                    # Fortran source code
├── ftp.py                              # Run and plot the Fortran-based simulation
├── pyvsft-basic_example.py             # Benchmark Fortran vs Python (single value of n)
├── pyvsft-with_steps.py                # Benchmark across multiple values of n
├── requirements.txt                    # Python dependencies

Installation

1. Clone the Repository

git clone https://github.com/rodrigosf672/pyohio2025-from_fortran_to_python.git
cd pyohio2025-from_fortran_to_python

2. Create a Virtual Environment

python -m venv venv
source venv/bin/activate

3. Install Dependencies

pip install -r requirements.txt

4. Compile the Fortran Code

To compile the fortran code, you need to have f2py installed, which comes with NumPy. However, the backend Fortran compiler for Python versions 3.12 and above has changed, hence being necessary to install meson and ninja:

brew install meson ninja  # macOS (use equivalent commands for Ubuntu/Windows)

Why do we need this? --> Meson reads the Fortran file and generates a Ninja build script. Then, Ninja runs the compilation commands (using gfortran) to build the .so file.

Then, you can compile the Fortran code using f2py:

f2py -c fortran-code.f90 -m cracker

This generates a .so file that Python can import.

In case you have problems with this step, you can find two already compiled .so files for Python 3.12 and 3.13 in this repo.

Usage

• Plot chemical concentration profile:

  python ftp.py

• Run basic benchmark (n=1000):

  python pyvsft-basic_example.py

• Run benchmark over multiple step sizes - takes a long time!:

  python pyvsft-with_steps.py

Key Takeaways

• Fortran provides significant performance advantages for numerical solvers.
• Python offers ease of prototyping, visualization, and benchmarking.
• Together, they demonstrate the power of using the right tool for the right job.
• Our Fortran implementation runs up to 700× faster than the pure Python version.

License

This project is licensed under the MIT License.

References

This project was inspired by prior studies that modeled the cracking of Ethylene Dichloride (EDC), particularly:

• Ferreira, H. S. (2003).
  Métodos matemáticos em modelagem e simulação do craqueamento térmico do 1,2-Dicloroetano.
  Doctoral dissertation, University of Campinas, Brazil. https://repositorio.unicamp.br/Busca/Download?codigoArquivo=477111

• Schirmeister, R., Kahsnitz, J., & Träger, M. (2009).
  Influence of EDC cracking severity on the marginal costs of vinyl chloride production.
  Industrial & Engineering Chemistry Research, 48(6), 2801–2809.
  https://doi.org/10.1021/ie8006903

• Fahiminezhad, A., Peyghambarzadeh, S. M., & Rezaeimanesh, M. (2020).
  Numerical Modelling and Industrial Verification of Ethylene Dichloride Cracking Furnace.
  Journal of Chemical and Petroleum Engineering, 54(2), 165–185.
  https://doi.org/10.22059/jchpe.2020.286558.1291

These works provided a foundation for understanding the chemical and engineering significance of EDC cracking.
What we present here is a highly simplified model, focusing on the reaction kinetics and solver performance. Hence, this work does not explore industrial-scale reactor behavior, fluid dynamics, or heat transfer, which are crucial in the real-world systems studied in the references above.

Authors

• Rodrigo Silva Ferreira
• Prof. Helianildes Silva Ferreira