Methane detection to 1 ppm using machine learning analysis of atmospheric pressure plasma optical emission spectra

Abstract

Optical emission spectroscopy from a small-volume, 5 μl, atmospheric pressure RF-driven helium plasma was used in conjunction with partial least squares-discriminant analysis for the detection of trace concentrations of methane gas. A limit of detection of 1 ppm was obtained and sample concentrations up to 100 ppm CH4 were classified using a nine-category model. A range of algorithm enhancements were investigated including regularization, simple data segmentation and subset selection, feature selection via Variable Importance in Projection and wavelength variable compression in order to address the high dimensionality and collinearity of spectral emission data. These approaches showed the potential for significant reduction in the number of wavelength variables and the spectral resolution/bandwidth. Wavelength variable compression exhibited reliable predictive performance, with accuracy values >97%, under more challenging multi-session train—test scenarios. Simple modelling of plasma electron energy distribution functions highlights the complex cross-sensitivities between the target methane, its dissociation products and atmospheric impurities and their impact on excitation and emission.

Identification of Different Classes of VOCs Based on Optical Emission Spectra Using a Dielectric Barrier Helium Plasma Coupled with a Mini Spectrometer

“…19−23 Additionally, some studies have tried to use machine learning techniques to identify the type and concentration of VOCs. 16,24 However, the reported DBD- OES systems usually have a relatively large size and need to be coupled with a bulky spectrometer, which makes them unsuitable for μGC systems.…”

“…Dielectric barrier discharge (DBD) plasma has the advantages of simple configuration, wide operating pressure range, low temperature and power consumption, zero electrode contamination, and high ionization ability . Therefore, DBD (argon- or helium-based) is often used as the plasma excitation source of OES, and several DBD-OES systems have been reported for detection of VOCs. − Additionally, some studies have tried to use machine learning techniques to identify the type and concentration of VOCs. , However, the reported DBD-OES systems usually have a relatively large size and need to be coupled with a bulky spectrometer, which makes them unsuitable for μGC systems.…”

A compact AC-glow discharge-optical emission spectrometer for real-time detection of N2 and Ar in O2 with sub-ppm-level sensitivity

“…There have been some important achievements in gas analysis by optical emission spectroscopy of plasma recently. [7][8][9][10][11][12] Due to pioneering works, real-time, ppb-level detection of multiple gases in the air is close. Among them, we have promoted the sensitive detection of N 2 and Ar in O 2 to sub-ppm level in realtime by employing a highly efficient signal collection optical system and the micro-glow discharge driven by high DC voltage.…”

Distinguishing methane from other hydrocarbons using machine learning and atmospheric pressure plasma optical emission spectroscopy

“…The application of supervised learning models based on partial least squares discriminant analysis (PLS-DA) was investigated for the identification of trace gases using data obtained from plasma optical emission spectroscopy [1]. As a result, the parts-per-million (ppm) classification of a single type of * Author to whom any correspondence should be addressed.…”