基于近红外胶体量子点阵列和光谱重构算法检测多种化学物质的方法研究

在众多化学物质检测技术手段中，红外检测技术由于具有非破坏性、灵敏度高、检测速度快、准确性好等特点，广泛应用于化工、生物医学、食品安全等领域。量子点光谱仪是使用量子点代替光栅作为分光器件，结合阵列探测器及光谱重构算法实现光谱检测的新型微型光谱仪，具有体积小、成本低等优点。为了进一步提升现有量子点光谱仪和量子点器件检测化学物质的普适性，为微小型近红外分光器件研制提供有效技术途径，以危险化学品乙醇、化学战剂模拟剂甲基膦酸二甲酯、二氯甲烷为目标物，通过将多种量子点材料与紫外固化胶混合后沉积在RGB点阵模块并固化，制备了发射光谱波段为900～1 600 nm的近红外胶体量子点阵列。采用经验模态分解方法提取输入光谱的高频信号以减小随机噪声干扰，并基于最小二乘法建立了相应光谱重构算法。实验结果表明，近红外胶体量子点阵列制备方法简单，成本低、稳定性较好。具有144条光谱通道的近红外胶体量子点阵列实现的重构光谱分辨率可以达到4.861 nm，与标准吸收光谱相比，其特征峰最小偏差仅为0.043%。因此，使用近红外胶体量子点阵列结合光谱重构算法可以实现气态、液态目标物的光谱重构和检测识别。未来，通过增加阵列数量可有效提升重构光谱的光谱分辨率；通过增加所选量子点材料，还可以实现从紫外到红外波段范围内的光谱检测；通过优化检测光路和重构算法参数提高目标物检测信噪比。

Methods of Detecting Multiple Chemical Substances Based on Near-Infrared Colloidal Quantum Dot Array and Spectral Reconstruction Algorithm

Infrared detection technology is widely used in the field of chemical engineering, bio-medicine, food safety, among the many chemical substance detection techniques due to its characteristics such as non-destructiveness, high sensitivity, fast detection speed, and good accuracy. Quantum dot (QD) spectrometer is a new type of micro spectrometer that uses QD instead of grating as a light splitting device and combines array detector with spectral reconstruction algorithm to realize spectrum detection. It has the advantages of small size and low cost. In order to improve the universality of existing QD spectrometers, QD devices for detecting chemical substances, and ultimately provide an effective technical approach for the development of micro-near infrared (NIR) spectroscopy devices. This article used hazardous chemicals Ethanol, simulants of chemical warfare agent sarin, mustard gas, including Dimethyl Methylphosphonate and Dichloromethane as the targets. A NIR colloidal quantum dot (CQD) array with an emission spectrum of 900~1 600 nm was prepared by mixing a variety of QD materials with UV curing glue and deposited on the RGB dot matrix. Extracted the high-frequency signal of the input spectrum and reduced the random noise interference with empirical mode decomposition method, established the corresponding spectrum reconstruction algorithm based on the least square method. The experimental results show that the preparation method of the NIR CQD array is simple, low-cost, and stable. The reconstructed spectral resolution achieved by the NIR CQD array with 144 spectral channels can reach 4.861nm. Compared with the standard absorption spectrum, the minimum deviation of its characteristic peak is only 0.043%. Therefore, detecting and identifying gas and liquid targets can be achieved by combing the NIR CQD arrays with spectral reconstruction algorithms. In the future, the spectral resolution of the reconstructed spectrum can be effectively improved by increasing the number of arrays; Spectral detection from UV to IR can also be achieved by increasing the QD materials selected; Target detection signal-to-noise ratio can be improved by optimizing the optical detection path and the reconstruction algorithm parameters.