希特勒
材料科学
激光器
可调谐激光吸收光谱技术
频率调制
计量系统
光谱学
电子工程
可调谐激光器
光学
光电子学
带宽(计算)
计算机科学
吸收光谱法
波长
工程类
物理
电信
天文
量子力学
作者
Walter Johnstone,Andrew McGettrick,Kevin L. Duffin,Amy Cheung,George Stewart
出处
期刊:IEEE Sensors Journal
[Institute of Electrical and Electronics Engineers]
日期:2008-07-01
卷期号:8 (7): 1079-1088
被引量:40
标识
DOI:10.1109/jsen.2008.926168
摘要
Tunable diode laser spectroscopy (TDLS) can only be successfully implemented if a number of system characterization procedures and critical parameter measurements can be made accurately. These include: application of a wavelength/frequency scale to the signals recovered in time; measurement of the frequency dither applied to the laser; measurement of the relative phase between the laser power modulation and frequency modulation; determination of the background amplitude modulation for normalization purposes and measurement of required cross broadening coefficients for the host/target gas mixtures. Easy to implement, accurate and low-cost systems and procedures for achieving these are described and validated below. They were developed for two new approaches to TDLS measurements, viz the residual amplitude modulation (RAM) technique and the phasor decomposition (PD) method, but are equally applicable to all forms of TDLS. Following full system characterization using the new techniques, measurements of the absolute transmission function of the 1650.96 nm absorption line of methane over a wide range of concentration and pressure were made using the RAM technique. The close agreement with theoretical traces derived from HITRAN data validated the entire approach taken, including the system characterization procedures. In addition, measurements of a wide range of gas concentration and pressure were made by curve fitting theoretical traces to the measured transmission functions obtained using a variety of operating conditions. Again, the low errors confirmed the validity of the new methods and the system characterization/measurement procedures described here.
科研通智能强力驱动
Strongly Powered by AbleSci AI