Mid-infrared high-sensitive cavity-free in-situ CO gas sensing based on up-conversion detection

TL;DR

This work tackles precise, real-time CO detection by activating the mid-infrared fundamental vibrational band and converting the MIR absorption signal to the visible using upconversion, enabling detection with silicon detectors at room temperature. The authors implement a cavity-free, in situ scheme based on difference-frequency generation, achieving a record-like sensitivity with a detection limit of $79.6$ ppb over a $0.14$ m pathlength and validating the approach with Allan-deviation analysis that highlights 70 s integration for optimal precision. They demonstrate linear response with a slope of $0.0775$%/ppm and reproduce HITRAN-consistent results, while showing that single-photon-level detection with diffuse reflection (via SPAD) is feasible for non-line-of-sight sensing; MCT-based room-temperature detection, by comparison, exhibits larger fluctuations. The results establish room-temperature, high-sensitivity CO sensing via upconversion as a viable route for industrial monitoring and medical diagnostics, with a clear path to enhancements using higher-power MIR sources and wavelength-tunable pumps for multi-line spectroscopy.

Abstract

Carbon monoxide (CO) is a significant indicator gas with considerable application value in atmospheric monitoring, industrial production and medical diagnosis. Its fundamental vibrational band locates around 4.6 $\upmu$m and has larger absorption line strength than that of overtone band, which is more suitable for the precise identification and concentration detection of CO. In this paper, the up-conversion detection is employed to convert the mid-infrared absorption signal obtained by TDLAS to the visible light band, then a silicon-based detector is utilized for detection. By which, we can achieve the highest sensitivity of 79.6 ppb under the condition of cavity-free in-situ with an absorption range length of only 0.14 m. Furthermore, the single-photon level real-time detection of CO concentration after the diffuse reflection is realized by using SPAD. This work demonstrates the merits of the up-conversion detection in terms of its functionality at room temperature and capacity for sensitivity detection. Furthermore, it presents a design and optimization methodology that has the potential to underpin the advancement of the method towards more practical applications, like industrial process monitoring, medical diagnosis and so on.

Figures (6)

• Figure 1: (a) Schematic diagram of the experimental setup. L terms: lenses; DM terms: dichromatic mirrors; F terms: filters; PPLN terms: periodically poled lithium niobate crystals; R terms: Ag mirrors; HWP: half-wave plate; QWP: quarter-wave plate; Si-PD: silicon-based photodiode detector; MCT-PD: HgCdTe photodiode detector. (b) Spectral of pump and signal under different modulation voltages, along with their corresponding MIR spectral band and CO absorption peaks. (c) Energy level diagram corresponding to the DFG process.
• Figure 2: CO concentration measurement results using upconversion system under $\upmu$W MIR illumination. (a) Absorption line shapes corresponding to different CO concentrations. (b) Relationship between absorption rate and concentration at various CO levels, along with linear fitting results.
• Figure 3: Allan variance of absorption rate measured by upconversion detection at 5 ppm CO.
• Figure 4: Comparison of measurement results between MCT and upconversion detection of 300 ppm CO at room temperature.
• Figure 5: Single-photon-level measurement scenarios and experimental results. (a)Schematic diagram of experimental setup. D terms: diffuser. (b)Absorption line shapes of CO at different concentrations under a photon flux of approximately $1\times10^7$ cps.