NO<sub>2</sub> gas detection based on standard sample regression algorithm and cavity enhanced spectroscopy

Abstract

Cavity-enhanced absorption spectroscopy is a highly sensitive trace gas measurement technology, and the algorithm for retrieving gas concentrations is critical. The absorption cross-section is traditionally used to retrieve the concentration. However, the absorption cross-section used in the fitting process is affected not only by the response function of the instrument and the light source, but also by temperature and pressure. The uncertainty of the absorption cross-section will influence the accuracy of the result. Therefore, in order to eliminate the measurement error introduced by the uncertainty of the absorption cross-section and the instrument response function, a concentration regression algorithm based on the absorption spectrum of the standard sample is proposed. The process of concentration inversion is optimized. The absorption spectrum of standard gas is used to fit the unknown spectrum. In order to verify the effectiveness of the algorithm, the incoherent cavity enhanced absorption spectroscopy (IBBCEAS) system based on a blue light-emitting diode (LED) operating at 440 nm is established to analyze the absorption spectrum of NO<sub>2</sub>; and the fitting effect, measurement accuracy and stability are compared with the counter parts from the traditional absorption cross-section method. In the experiment, the measured reflectance of the cavity mirror is 99.915%. Compared with the conventional absorption cross-section regression algorithm, the standard sample regression algorithm proposed in this paper shows good results, in which the measurement accuracy is increased by about quadruple. The Allan deviation shows that a detection limit of 5.3 ppb can be achieved at an integration time of 360 s. Finally, the performance of the experimental system is evaluated by measuring the NO<sub>2</sub> with different concentrations prepared by standard samples. The result shows good agreement with the theoretical value, which indicates that the improved spectral analysis algorithm is feasible and reliable for gas detection. This method can be used not only to measure NO<sub>2</sub>, but also to detect other gases, which shows great potential applications in monitoring the industrial emissions, atmospheric chemistry and exhaled breath analysis.