Inverse Gas Chromatography

A lightweight, reproducible toolkit for parsing, analysing, and visualising Inverse Gas Chromatography (IGC) surface energy data.

GitHub — inverse-gas-chromatography

At a glance

  • Outcome: Functional toolkit for parsing IGC instrument exports and computing surface energy components
  • Tools: Python, Jupyter, Cirrus Plus v1.5 (validation reference)
  • Skills: scientific software development, data parsing, reproducible computation, AI-assisted development
  • Result highlight: calculations validated against Cirrus Plus v1.5 outputs; notebook prototypes available in the repository

Overview

Can IGC surface energy data be extracted, calculated, and visualised in a transparent, reproducible way outside of proprietary software? Cirrus Plus handles this well but operates as a black box. This toolkit reimplements the core calculations openly, so results can be inspected, reproduced, and extended.

Objectives

  • Build a reliable parser for multi‑table CSV exports from IGC instruments
  • Implement core surface energy calculations in a transparent, reproducible way
  • Provide clear visualisations (e.g., γsD, γsAB, acid/base components, isotherms)
  • Offer a simple interface for uploading CSVs and generating graphs
  • Maintain honest documentation of the development process, including AI‑assisted steps
  • Produce a user‑friendly tool suitable for early‑career researchers

Scope

Data handled: multi-table CSV exports from IGC-SEA instruments Calculations: dispersive surface energy (γsD), acid/base components (γsAB), total surface energy, adsorption isotherms

Approach

  • Developed a CSV parser to handle the irregular multi-table format exported by IGC instruments
  • Reimplemented surface energy calculations from the IGC-SEA methodology literature
  • Validated outputs against Cirrus Plus v1.5 reference calculations
  • Built Jupyter notebook prototypes to demonstrate parsing, calculation, and visualisation

Results

  • Parser successfully extracts retention data from instrument CSV exports
  • Calculated γsD, γsAB, and acid/base components match Cirrus Plus v1.5 outputs
  • Notebook prototypes produce plots for dispersive energy, isotherms, and component breakdown

Interpretation

What worked well

  • Validation against Cirrus Plus confirmed calculation accuracy
  • Jupyter notebook format makes each step transparent and reproducible
  • AI-assisted development accelerated prototyping without compromising scientific accuracy

Limitations

  • Currently prototype-level; not packaged for general distribution
  • Parser handles one instrument export format; other formats would need extension

Improvements

  • Package into a proper Python library with a CLI or lightweight web interface
  • Extend parser to handle additional instrument export formats
  • Add uncertainty propagation to surface energy outputs

What I learned

  • Practical experience structuring a scientific Python project
  • Working with messy, instrument-exported CSV formats
  • Using AI-assisted development responsibly within a technical domain
  • How to validate computational outputs against reference software