Skarstrom
Skarstrom is an open-source library for adsorption process simulations including pressure, vacuum and temperature swing adsorption (PSA, VSA and TSA). Written in C++, the engine solves a 1D finite-volume model of a packed-bed adsorption column with multiple layers. Skarstrom is designed for engineers and scientists to rapidly develop new adsorption-based gas separation systems, with modules for accelerating convergence to cyclic steady state and for optimising purity, recovery, productivity and energy consumption.
Why Skarstrom
Several commercial simulation packages are available, but to date there has been no good open-source tool for adsorption modelling. It has become common practice for people to write their own models from scratch, resulting in a lot of repeated work and a lack of transparency and reproducibility.
Skarstrom aims to bridge this gap by being the first comprehensive open project for adsorption researchers to collaborate on. We also showcase exceptional solver speed and performance, and hope this can be improved further by collaboration with the community.
Performance
Cycle optimisation and parameter fitting are important design tasks that require running to cyclic steady state (CSS) hundreds or thousands of times, so Skarstrom has been optimised for performance to make these tasks practical on common hardware. There are no well-established performance benchmarks; however, the table below can be used as a guide to the expected time to reach CSS. You are encouraged to run the examples yourself and benchmark them against your own code.
CSS is reached when every component's mass balance over a cycle closes within 0.5 %. Median of 3 runs on one core of an Apple M4, Release build; reproduce with python scripts/benchmark_css.py.
Graphical user interface
We host a user interface on the Skarstrom website. It is a browser-based interface that runs the C++ engine and is the best starting point for users who don't want to write code. From the interface you can run examples, visualise the results, and run optimisation and parameterisation studies.
Quick start
Skarstrom is published on PyPI, so you can use it in your project with:
pip install skarstrom
The best place to start is one of the 33 bundled examples, each of which reproduces a published study.
Documentation
The full documentation is at https://samaffleck.github.io/skarstrom/: getting started, the Python API and command line, tutorials, the model equations, the adsorbent library and the validation results.
Citing
If you use Skarstrom in published work, please cite it using CITATION.cff.
Get involved
If you would like to be a part of the steering committee to guide Skarstrom's direction, please contact Sam Affleck on LinkedIn. People from academia and industry are both welcome.
Contributing
Contributions are welcome — see CONTRIBUTING.md (DCO sign-off, tests, data-integrity rules) and the Code of Conduct. Security issues: SECURITY.md. Coding agents: AGENTS.md.
Licence
Skarstrom is licensed under the Apache License 2.0; see also NOTICE. It builds on third-party components under their own permissive licences, listed in THIRD_PARTY_NOTICES.md.
