Quantum cascade laser for gas sensing in mid-infrared region

• Main Author: Gao, Yuying
• Other Authors: Wang Qijie
• Format: Thesis-Master by Coursework
• Language: English
• Published: Nanyang Technological University 2021
• Subjects: Engineering::Electrical and electronic engineering::Optics, optoelectronics, photonics
• Online Access: https://hdl.handle.net/10356/152418
• Institution: Nanyang Technological University

Volatile Organic Compounds (VOCs) come in a variety of forms, some of which are hazardous to human health or harmful to the environment. As a result, VOC detection and tracking has piqued interest for quite some time. In the Mid-infrared (MIR) spectral area, most VOCs display simple vibrational absorption bands (2.5-25 μ m). The absorption of light by these basic bands provides an almost universal detection method. Most molecular species' rotational vibration band can be directly accessed by Quantum Cascade Laser (QCL), which work in the mid-infrared region, so it is very suitable for high specificity and high sensitivity trace gas detection. These sensors have a variety of applications, for instance, atmospheric pollution monitoring, medical diagnosis, identification of dangerous substances, homeland security, as well as process control in industry. More importantly, tunable External Cavity Quantum Cascade Lasers (EC-QCL) especially provide narrow line width, a large range of tuning options and a consistent power output, which opens up new opportunities for the development of sensors. In this thesis, we use the Mid-infrared (MIR) tunable External Cavity Quantum Cascade Laser (EC-QCL) with the detection method of Tunable Diode Laser Absorption Spectroscopy (TDLAS) to focus on the absorption spectra of two kinds of VOC gases, acetylene and acetone, as well as a dangerous gas, ammonia. More importantly, we find out the relationship between absorption spectra and gas concentration by using Matlab. At the same time, we also predict the concentration of each gas in binary gas mixture by using Deep Extreme Learning Machine (Deep-ELM) networks.