Objective Methane (CH4) is a critical greenhouse gas, ranking just behind carbon dioxide in terms of its impact on global climate change. For effective environmental management and to gain insights into emission characteristics, precise and continuous monitoring of atmospheric methane volume fractions is vital. Tunable laser absorption spectroscopy (TLAS) serves as a powerful technique for trace gas detection due to its high selectivity, rapid response speed, and relatively straightforward system design. However, in real-world measurement scenarios, spectral absorption features are significantly affected by environmental factors such as temperature and humidity, leading to complex nonlinear relationships between spectral characteristics and gas volume fractions. This creates challenges for traditional volume fraction inversion methods, so this study aims to develop a more accurate and stable methane volume fraction inversion model. Methods A methane volume fraction inversion model based on a genetic algorithm-optimized backpropagation neural network (GA- BP) is developed using spectral data from TLAS. The experimental setup uses a quantum cascade laser (QCL) operating at a central wavelength of 7.8 mu m as the light source, and high-resolution spectral scanning is conducted around the methane absorption line at 1281.6100 cm(-1). A multi-pass absorption cell is employed to increase the optical path length and enhance detection sensitivity in ambient air. Spectral data are continuously captured at a sampling rate of 1 Hz, forming a comprehensive dataset that includes methane volume fraction, the integrated area of absorption peaks, and temperature and humidity measurements-this dataset serves as the input for training and validating the GA-BP model. To minimize noise and environmental interference, wavelet denoising is applied to raw spectral signals, smoothing a few abnormal data points while preserving the overall data trend. Input features and output volume fractions are normalized using min-max scaling to ensure numerical stability during neural network training. The selected input variables include the absorption peak integrated area, temperature, and humidity, with experimentally measured methane volume fraction as the output label. A three-layer BP neural network (input layer, single hidden layer, output layer) is implemented, and systematic testing confirms that three neurons in the hidden layer balance modeling nonlinearity and computational efficiency. A genetic algorithm (GA) is used to globally optimize the initial weights and thresholds of the BP network (addressing its vulnerability to random initial parameters). Key GA parameters (initial weight range, population size, crossover rate, mutation rate) are analyzed, and the optimal configuration is determined as an initial weight range of [0, 1], a population size of 10, a crossover probability of 0.6, and a mutation probability of 0.1. The GA-BP model is evaluated against established inversion models (support vector machines (SVM), long short-term memory networks (LSTM), conventional BP neural networks, and particle swarm optimized BP neural networks (PSO-BP)). All models are trained and tested using the same dataset, with parameters optimized through accepted strategies to ensure fairness. Linear regression analysis and multiple error metrics are used for comprehensive performance evaluation. Results and Discussions The GA-BP model exhibits the highest overall inversion accuracy among the evaluated models. In the test set, it achieves a root mean square error (RMSE) of 0.517 %, a mean absolute error (MAE) of 0.382 %, a mean square error (MSE) of 0.003 %, and an exceptional correlation coefficient (R2) of 99.985 %. Additionally, the model shows a maximum relative error of 1.642 % and an average relative error of 0.181 %. Although its maximum relative error is slightly higher than that of the PSO-BP model, most data points demonstrate small relative errors, reflecting the GA-BP model's robust overall performance. The optimization of the BP neural network through GA significantly enhances its initial search capability and strengthens its effectiveness in modeling nonlinear relationships within TLAS spectral data. This confirms that the GA-BP approach effectively mitigates the impact of environmental factors (temperature and humidity) on spectral absorption features, addressing the limitations of traditional inversion methods. Conclusions The methane volume fraction inversion method based on GA-BP proposed in this study provides a reliable and efficient approach for high-precision gas volume fraction retrieval via laser absorption spectroscopy. Future work will focus on validating the model in more complex environmental settings and with larger datasets, as well as exploring the integration of advanced spectral feature extraction techniques and deep learning approaches to further enhance inversion performance.

A mid-infrared laser spectrometer with precise temperature and pressure control has been developed using a room-temperature continuous-wave quantum cascade laser (cw-QCL). This spectrometer was used to analyze 12 CO2 absorption lines near 2310 cm-1 and compare the results with the HITRAN 2022 database. The study evaluated eight line-shape models based on their residuals and computational efficiency. Notably, the Hartmann-Tran (HT) and speed-dependent Retina (SDR) profiles demonstrated the best fit, achieving mean deviations of 0.06%. These results affirm the accuracy and reliability of the integrated system and introduce an effective approach for integrating high-precision laser spectrometers. This work serves as a valuable reference for high-precision measurements of CO2 concentrations and isotope ratios.

A mid-infrared laser spectrometer with precise temperature and pressure control has been developed using a room-temperature continuous-wave quantum cascade laser (cw-QCL). This spectrometer was used to analyze 12 CO2 absorption lines near 2310 cm-1 and compare the results with the HITRAN 2022 database. The study evaluated eight line-shape models based on their residuals and computational efficiency. Notably, the Hartmann-Tran (HT) and speed-dependent Rautian (SDR) profiles demonstrated the best fit, achieving mean deviations of 0.06%. These results affirm the accuracy and reliability of the integrated system and introduce an effective approach for integrating high-precision laser spectrometers. This work serves as a valuable reference for high-precision measurements of CO2 concentrations and isotope ratios.

Gaseous nitrous acid (HONO) is an important source of hydroxyl radicals (OH) in the atmosphere, significantly influencing atmospheric oxidation capacity and the formation of secondary pollution. However, its extremely low environmental concentration, combined with considerable spatial and temporal variations, presents challenges for high-precision monitoring. This study employs a quantum cascade laser (QCL) with a central wavelength of 1280 cm-1, utilizing highly sensitive Tunable Laser Absorption Spectroscopy (TLAS) and Cavity Ring-Down Spectroscopy (CRDS) techniques to measure HONO. Initially, high-precision calibration measurements were conducted on the HONO absorption lines within this wavelength range. Subsequently, the acquired spectral line data were used to carry out highly sensitive measurements and noise reduction on the HONO spectral lines, employing CRDS technology alongside time convolutional neural networks. The findings of this study indicate that mid-infrared spectroscopy, in combination with deep learning analysis, provides an efficient and reliable new technological approach for real-time, high-precision monitoring of atmospheric trace HONO.

A new sensor system using tunable laser absorption spectroscopy (TLAS) technology and a quantum cascade laser (QCL) has been developed to rapidly and precisely measure atmospheric CO2 levels and its stable isotope ratios. The sensor can detect CO2 levels as low as 0.219 ppm within 116 ss. It achieved measurement precisions of 0.256 %o for 513C and 0.293 %o for 518O at 95 ss. This system used the weighted Kalman filtering algorithm for the first time, resulting in a 7-fold improvement in precision for 513C and 518O measurements within a 1-s integration time. This improvement is comparable to the precision obtained by the standard averaging technique after 95 ss. Multiple measurements near the laboratory showed average values of -9.19 +/- 0.29 %o for 513C and -1.46 +/- 0.34 %o for 518O in the atmosphere. The successful development of this gas sensor lays the foundation for its future application in atmospheric CO2 tracing.

Real-time monitoring of atmospheric water vapor concentrations is critical for combustion optimisation and emission control in high-temperature industrial processes. However, traditional spectroscopic techniques frequently encounter substantial noise interference, thereby impeding the precision of measurements when operating in extreme conditions. This study proposes a novel feedforward neural network (FNN)-assisted Savitzky-Golay collaborative filtering algorithm for noise suppression in tunable diode laser absorption spectroscopy (TDLAS). The proposed method integrates the nonlinear learning capability of FNN with the adaptive smoothing advantages of S-G filtering, achieving superior denoising performance compared to conventional single-filter approaches. Experimental validation using HITEMP-simulated spectra demonstrates a remarkable signal-to-noise ratio (SNR) enhancement from 20.93 dB to 46.60 dB, representing a 25.66 dB improvement 6.25 dB superior to optimal traditional filtering methods. Field tests conducted in a 1010 degrees C blackbody furnace environment revealed a concentration measurement standard deviation of 17.29 ppm, with Allan deviation analysis confirming a detection sensitivity of 2.2 ppm at 99 s integration time. The system exhibits excellent linear response (R2 = 0.97) across 600 continuous measurements, achieving a mean absolute error of 22.46 ppm compared to reference values. This breakthrough in spectral processing technology enables reliable water vapor monitoring under challenging high-temperature conditions, significantly improving process control capabilities in metallurgical and energy-intensive industries.

A mid-infrared quantum cascade laser was utilized to simultaneously scan three CO2 isotope absorption lines. The application of the weighted Kalman filtering algorithm has achieved rapid and precise inversion of exhaled CO2 isotope ratios. The Allan variance analysis found that the measurement precisions for 513C and 518O were 0.17 %o and 0.22 %o, respectively, with an optimal integration time of 110 seconds. When applying this system to measure CO2 isotopes in exhaled breath, the measurement precisions of 513C and 518O processed by the weighted Kalman filtering algorithm were improved by about 8 and 9 times within 1-second integration time, with average values of -26.55 +/- 0.27 %o and -27.29 +/- 0.35 %o, respectively. The standard average technique needs to take more than 71 seconds to reach the same level of precision. This study demonstrates the feasibility of using this system for precise and rapid detection of CO2 isotope ratios in exhaled breath, providing a powerful tool for respiratory detection and disease diagnosis.

A mid-infrared quantum cascade laser is utilized to simultaneously scan three noninterfering absorption lines of $\text{CO}_{2}$ gas. The weighted Kalman filter technique is employed to accurately invert the $\text{CO}_{2}$ gas concentration. Standard gas was used to evaluate the system's performance. According to Allan variance analysis, when the optimal integration time of the system is 102 seconds, the measurement accuracies for ${}^{12} \text{CO}_{2},{}^{16} \mathrm{O}^{12} \mathrm{C}^{18} \mathrm{O}$, and ${}^{13} \text{CO}_{2}$ are $0.119 \text{ppm}, 3.54 \text{ppb}$, and 15.8 ppb, respectively. When the optimal integration time is 99 seconds, the accuracies for $\delta^{13} \mathrm{C}$ and $\delta^{18} \mathrm{O}$ can reach 0.10 % and 0.49 %, respectively. The system's ability to achieve high-precision detection of atmospheric $\text{CO}_{2}$ concentration and isotope abundance has been demonstrated, providing a robust and invaluable tool for monitoring atmospheric conditions and advancing climate change research.

The spectral line positions and effective line strengths of dinitrogen tetroxide (N2O4) in the range of 1279.8-1282.5 cm-1 were measured using a continuous-wave quantum cascade laser spectrometer. A data processing program developed in MATLAB was used to fit the spectral data with Voigt, Rautian, and Galatry line profiles. The performance of different spectral line profiles was evaluated, and the Rautian profile achieved the best fit with faster calculations, making it ideal for real-time and long-term monitoring. The maximum uncertainties in the measured line positions and intensities are 1.95*10-3 cm-1 and 6.47 %, respectively. The line position uncertainty is mainly from reference spectral lines, while intensity uncertainty is due to N2O4 concentration calculations and fitting errors. These results aid in gas leak monitoring and spectral analysis of N2O4, as well as in studying nitrogen oxide reaction processes.

A new dual-gas sensor has been developed using a quantum cascade laser (QCL) that operates at 1255 cm −1 for fast and simultaneous measurements of nitrous acid (HONO) and methane (CH 4 ). The sensor uses direct absorption spectroscopy to detect HONO and CH 4 in the air, with a multi-pass absorption cell that has a path length of 173m. The sensor can detect HONO and CH 4 with minimum detection limits (1σ) of 198ppt and 2.95ppb, respectively, with an integration time of 1s. These detection limits can be improved to 39ppt and 0.31ppb, respectively, with an optimum integration time of 100s. The sensor can be used to monitor variations in the mixing ratios of HONO and CH 4 in the atmosphere, as well as basic meteorological parameters, to study the sources and sinks of HONO and monitor CH 4 pollution in the air.

A compact spectrometer based on a mid-infrared optical sensor has been developed for high-precision and real-time measurement of water isotope ratios. The instrument uses laser absorption spectroscopy and applies the weighted Kalman filtering method to determine water isotope ratios with high precision and fast time response. The precision of the measurements is 0.41‰ for δ18O and 0.29‰ for δ17O with a 1 s time. This is much faster than the standard running average technique, which takes over 90 s to achieve the same level of precision. The successful development of this compact mid-infrared optical sensor opens up new possibilities for its future applications in atmospheric and breath gas research.

The measurement of the line positions and effective line strengths of the & nu;3 fundamental band of trans-nitrous acid (trans-HONO) near 1280 cm-1 (7.8 & mu;m) by tunable laser absorption spectroscopy (TLAS) utilizing a room temperature continuous-wave quantum cascade laser (cw-QCL) was reported. The effective line strengths of 30 well-resolved trans-HONO absorption lines in the range of 1279.8-1282.2 cm- 1 were determined using the HONO line strength at 1280.3841 cm-1 as a scale. The maximum measurement uncertainty of 7.64% in the line strengths is mainly determined by the uncertainty of the referenced line strength, while the measurement precision of the line positions is better than 5.56 * 10-3 cm- 1. The line positions and strengths of the trans-HONO absorption lines obtained in this work provide a reference for continuous gas monitoring and analysis of the sources and sinks of atmospheric HONO.

Objective The OH radicals produced by HONO photolysis are important oxidants in atmospheric reactions. The sources and mixing ratios of OH radicals are closely related to the level of HONO. The sources of HONO are still unclear under different contaminated conditions, different meteorological conditions, and different reaction conditions. HONO also has a great impact on people's health. To better understand the photochemical cycle of atmospheric HONO and its sources, the levels of HONO need to be accurately measured. N2O4 plays an important role in liquid propellants and is also a toxic gas. It is necessary to accurately detect its levels to better understand its reaction mechanism and perform real-time monitoring. Accurate measurement of HONO and N2O4 levels requires precise absorption line parameters, such as line positions, line intensities, and spectral line broadening. Methods In this experiment, high-resolution quantum cascade laser absorption spectroscopy technology is used to measure HONO and N2O4 gas samples. A room temperature continuous wave quantum cascade laser (CW-QCL) combined with a 50 m path-length absorption cell are used to measure the absorption line frequencies of the two gases. Using the heterogeneous reactions of NO2 and H2O to prepare gas samples of HONO. The overall absorption line frequencies are calibrated by the two H2O absorption lines at the frequencies of 1280. 0475 cm(-1) and 1281. 1611 cm(-1). According to the known HONO line intensity at 1280. 4 cm(-1), the level of HONO in the gas sample as well as the signal-to-noise ratio and minimum detection limit of the system are calculated by the Beer-Lambert law and Voigt line shape fitting. Results and Discussions The absorption spectra of gas samples in the range of 1279.5-1282.5 cm(-1) are obtained as shown in Fig. 4. The gases that may exist in the absorption cell mainly include three types, i.e., exhaled gases, the gases in the air, and the gases generated by the chemical reaction. The HITRAN database and published papers are used to find the possible gases (CH4, N2O, H2O, CO2, NO2, HNO3, and HONO) in the absorption cell. Among them, CH4, N2O, and H2O are gases in the atmospheric environment. Since the absorption cell has been pumped into a vacuum before the gas sample to be measured is introduced, the absorption characteristics of these three gases will not be displayed in the measured absorption lines theoretically. The CO2 and H2O in the exhaled gas will inevitably enter the gas bag. Combining the simulated absorption spectra of NO2, HNO3, H2O, and CO2 under the same experimental conditions, the interferences of NO2, HNO3, and CO2 can be excluded and the gas species corresponding to each absorption line can be determined. The overall absorption line positions are calibrated by two H2O absorption lines with frequencies of 1280. 0475 cm(-1) and 1281. 1611 cm(-1). The specific absorption line positions of HONO and N2O4 obtained in this experiment are concluded in Table 1. Due to the instability, solubility, and photolysis of HONO, its absorbance intensities decrease with time in the absorption cell. In order to minimize the measurement error and avoid the reduction of HONO absorbance intensity, the absorption lines will be collected immediately (within 10 s) when the gas sample just entered the absorption cell. Finally, the level of HONO is calculated to be (0. 72 +/- 0. 04) x 10(-6) by a Voigt line shape fitting to the spectral line with a known line intensity of (3. 25 +/- 0. 17) x 10(-20) cm/(molecule.cm(-2)) at wave number of 1280. 4 cm(-1). The statistical calculation of the baseline part of Fig. 6 is performed, and the value is used as the noise value N of the output signal. The signal-to-noise ratio of the experimental system is about 64. 96 and the minimum detection limit is (11.15 +/- 0.50)x10(-9). Conclusions Trans-HONO and N2O4 gases are continuously measured at the same time, and the specific absorption line frequencies of the two gases in the range of 1279.5-1282.5 cm(-1) are obtained by using a 7.8 mu m room-temperature CW-QCL and a long path-length absorption cell. The decay time of HONO in a closed absorption cell made of quartz is obtained by fitting and analyzing the decay curve of HONO. According to the known absorption line intensity of trans-HONO at 1280.4 cm(-1), the level of trans-HONO in the sample to be measured is calculated to be (0.72 +/- 0.04)x10(-6), the corresponding minimum detection limit of the system is about (11.15 +/- 0.50) x 10(-9). As the absorption line intensity of N2O4 has not been reported in the spectral database and published articles, the level of N2O4 in the sample to be tested has not been calculated. The absorption line frequencies of HONO and N2O4 obtained in the experiment provide a reference for real-time continuous gas monitoring, sources and sinks analysis of atmospheric HONO, and analysis of the N2O4 chemical reaction process.

Simultaneous measurement of H217O/H216O, H218O/H216O, and HDO/H216O in air with a compact spectrometer based on a mid-infrared distributed feedback (DFB) laser was described. The obtained mixing ratios of H216O, H217O, and H218O agreed reasonably well with those measured by a hygrometer. The precision and repeatability of the spectrometer were analyzed. Indoor air tests demonstrated that its 220-s precision was 0.08 ‰, 0.06 ‰, and 0.14 ‰ for δ18O, δ17O, and δ2H respectively. The measured values of δ18O, δ17O, and δ2H in indoor air were highly correlated with the water vapor mixing ratios. The compact spectrometer provides in situ measurements of water vapor isotopes with high precision and fast time response, which opens new possibilities for its application in atmospheric and hydrological research in the future.

An improved Savitzky-Golay (S-G) filtering algorithm was developed to denoise the absorption spectroscopy of nitrogen oxide (NO2). A deep learning (DL) network was introduced to the traditional S-G filtering algorithm to adjust the window size and polynomial order in real time. The self-adjusting and follow-up actions of DL network can effectively solve the blindness of selecting the input filter parameters in digital signal processing. The developed adaptive S-G filter algorithm is compared with the multi-signal averaging filtering (MAF) algorithm to demonstrate its performance. The optimized S-G filtering algorithm is used to detect NO2 in a mid-quantum-cascade-laser (QCL) based gas sensor system. A sensitivity enhancement factor of 5 is obtained, indicating that the newly developed algorithm can generate a high-quality gas absorption spectrum for applications such as atmospheric environmental monitoring and exhaled breath detection.

Optical multi-pass cell (MPC) is widely used in laser absorption spectroscopy, a powerful gas analysis tool. In this letter, a miniature MPC with a high optical path length to volume ratio (OPL/V), consisting of trough mirrors, is proposed. It allows compact beam propagation without paths cross and spots overlap. The cell is optimized twice to reduce aberration and spot diffusion. An OPL/V of 309.03 is obtained for the proposed model. In addition, an OPL of 2.5 m with a volume of 16 cm(3) is achieved as the number of mirrors is increased to 30. To our best knowledge, it is the smallest MPC with the same OPL. The miniature MPC shows great potential for its application in portable gas sensor.

A compact laser spectrometry instrument was developed for high precision measurements of isotope ratio of CO 2 by tunable diode laser absorption spectroscopy in the mid-infrared at 2.7 μm. The experimental spectrum of carbon isotopologues in the gas phase near 3641 cm −1 is very suitable for real-time analysis of these isotopologues. Simultaneous measurements of the mixing ratio and the corresponding δ 13 C values of CO 2 in the atmosphere were performed. The achieved standard deviation (1σ) of δ 13 C was 1.8‰. The Allan analysis of the time series of the mixing ratio of CO 2 shows a measurement precision of 0.2‰ for δ 13 C with an optimum integration time of about 130 s. The spectrometer is capable of real-time measurements of stable carbon isotope ratios of CO 2 under ambient conditions.

Methane (CH4) and acetylene (C2H2) are important bioscience and chemical gases. The real-time monitoring and analysis of them have important research value in industrial process control. The time-sharing scanning assisted wavelength modulation spectroscopy (WMS) technique is developed for real-time and simultaneous detection of CH4 and C2H2. This system involves two near-infrared distributed feedback (DFB) lasers and a compact multipass cavity with an effective optical path of 52.2 m. The selected strong absorption lines of methane and acetylene are located at 6046.96 cm-1 and 6531.7 cm-1, respectively. The experiment environment is conducted at room temperature 23 °C and pressure 760 Torr. The sensor performance, including the minimum detection limit (MDL) and the stability, was improved by eliminating the influence of light intensity fluctuation using the WMS-2f/SAW technique. Allan deviation analysis indicates that a MDL of 0.1 ppm for CH4 and 0.2 ppm for C2H2 are achieved with 1-s integration time. And the instrument response time is about 44 s through the continuous analysis of standard gases. This sensitive, simple, reliable, and lowcost dual-gas sensor is very suitable for applications in the field environment, chemical process, and many other gas-phase analysis areas.

A compact isotope ratio sensor based on laser absorption spectroscopy at 2.7 μm was developed for high precision and simultaneous measurements of the D/H, 18O/16O and 17O/16O isotope ratios in glacier water. Measurements of the oxygen and hydrogen isotope ratios in glacier water demonstrate a 1σ precision of 0.3‰ for δ18O, 0.2‰ for δ17O, and 0.5‰ for δ2H, respectively. The δ values of the working standard glacier water obtained by the calibrated sensor system is basically identical to the IRMS measurement results with a very high calibration accuracy from 0.17‰ to 0.75‰. Preliminary results on the reproducibility measurements display a standard deviation of 0.13‰ for δ18O, 0.13‰ for δ17O, and 0.64‰ for δ2H, respectively.