Gas phase multicomponent detection and analysis combining broadband dual-frequency comb absorption spectroscopy and deep learning

Abstract

In absorption spectroscopy, analysis of multicomponent gas mixtures becomes challenging when absorption features overlap (blended spectra). Here we propose a gas sensor which can accurately identify the species and retrieve the concentrations of components in a gaseous mixture in a broad spectrum. The sensor integrates a mid-infrared dual-frequency comb laser source for spectrum acquisition and a deep learning algorithm for spectral analysis. The sensor was tested on gas phase mixtures of methane, acetone and water vapor. A prototype sensor was assessed in realistic scenarios in real time. We also systematically analyzed and presented explicit visualizations to explain the underlying working mechanism of the algorithms.

Dielectric Waveguide-Based Sensors with Enhanced Evanescent Field: Unveiling the Dynamic Interaction with the Ambient Medium for Biosensing and Gas-Sensing Applications—A Review

“…This technique provides precise measurements of gas concentrations and can identify gases with a high sensitivity [86]. However, frequency comb-based gas detection typically requires complex and expensive equipment, limiting its accessibility in some applications [87]. On the other hand, evanescent field absorptionbased techniques are often simpler and more cost-effective than frequency comb-based methods.…”

Deep Learning for Gas Sensing via Infrared Spectroscopy

“…Deep learning has recently been applied to multi-component gas speciation and concentration predictions [ 17 , 18 ]. The emergence of powerful deep learning pattern recognition in gas speciation applications, as explored here, is expected to benefit IR absorption gas sensing by replacing manual feature selection and spectral interpretation with automated processes [ 19 , 20 ]. Using these learning methods, sensors can learn on large quantities of spectral information (features and their shapes) in frequency bands that are arbitrarily selected and/or imposed by hardware or other constraints and make useful predictions on constituent species and their concentrations [ 17 , 18 , 21 , 22 , 23 , 24 , 25 ].…”