A dual-comb spectrometer for trace gas analysis in the mid-infrared

Abstract

A dual frequency comb spectrometer is realized by electro-optic modulation of a 1550 nm laser and subsequent conversion to the mid-infrared by difference-frequency generation (DFG). Using an optical parametric oscillator for the DFG the combs can be tuned from 3 µm to 4.7 µm with 440 comb modes covering 220 GHz (> 6 cm -1 ). Trace gas detection of nitrous oxide, carbon dioxide and methane is demonstrated with a 7.2-m-multi-pass cell while a sufficiently low noise-equivalent absorbance is reached in already 1 s. The bandwidth normalized noise-equivalent-absorption coefficient is consistently below 2.8 x 10 -6 Hz -1/2 cm -1 while the precision of the determined concentrations is better 2 % Hz -1/2 .

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

“…A variety of broadband MIR laser sources have been realized, such as frequency combs (FC) 6 , 13 , 15 – 17 , super-continuum sources (SC) 12 , quantum cascade lasers (QCLs) 1 , 10 that were combined with high sensitivity optical methods to further improve the detection limit and robustness of spectroscopic systems. Among the various methods, the dual-frequency comb (DFC) spectrometers integrate the broadband sources and comb teeth resolution into one platform, which exhibits high signal-to-noise ratio in a relatively short acquisition time and shows the potential of multicomponent gas identification in real-time applications such as biomass pyrolysis and combustion diagnostics 18 , 19 . The DFC system provides a unique set of characteristics that are especially useful for multicomponent gas mixture sensing, combining the broad spectral span, high spectral power, and high spectral resolution with a high spatial coherence, stability and ease of use 20 – 23 .…”

Analysis of Gas Mixtures with Broadband Dual Frequency Comb Spectroscopy and Unsupervised Learning Neural Network

“…[15] Modeling and fitting of the blended spectra is a feasible way to overcome such challenges. To determine the concentrations of gases of interest, fitting algorithms based on nonlinear least squares, such as partial least squares (PLS), are often employed, [16,17] relying on absorption profiles provided by spectroscopic databases such as HITRAN [18] or PNNL. [19] The extended spectral bandwidth leads, however, to an increased number of absorption profiles, which have to be computed for every iteration during the fitting.…”