Automated Quantitative Determination of Xe, CH4 and SF6 in Gas Mixtures Using a Residual Gas Analyser

Accurate identification and quantification of gaseous components used in radiation detectors are essential to guarantee their optimal performance and long-term reliability. In this study, we introduce a powerful computational tool developed to automatically analyse complex gas mixtures containing xenon (Xe), methane (CH4), and sulphur hexafluoride (SF6), utilising data collected from a Residual Gas Analyser (RGA). Built using the Python programming language, the algorithm expertly processes mass spectra by performing detailed statistical analyses on raw data and applying precise instrument-specific sensitivity corrections. It successfully pinpoints the key characteristic peaks of each gas, estimates their cracking patterns, and accurately calculates their relative concentrations within binary mixtures. The initial validation of the tool with pure gas spectra confirmed a precise match between the detected peaks and established literature values. Further testing with mixed gas samples demonstrated its robust capability to reliably identify each component and quantify their concentrations with high accuracy. For example, when analysing a mixture of 20% CH4 and 80% Xe, the system not only identified both gases but also provided concentration estimates closely aligned with the known composition. This innovative approach offers a reliable, efficient, and cost-effective solution for continuous monitoring of gas mixture composition and purity, effectively automating the analysis process and significantly reducing the need for manual intervention.