Abstract:Wavelength modulation-direct absorption spectroscopy (WM-DAS) integrates the advantages of measuring absolute absorbance profile from calibration-free direct absorption spectrum (DAS) with the enhanced noise rejection and high sensitivity of wavelength modulation spectrum (WMS). This method can be used to precisely recover the crucial absorbance profile via the extraction of the characteristic frequency of the modulated transmitted light. In this paper, the WM-DAS method with non-calibration and high signal-to-noise ratio is integrated with a Herriott cell (about 128 m). Under the condition of atmospheric pressure and room temperature, the absorptance functions of two spectral lines of CO2 (6330.821 cm–1) and CH4 (6046.964 cm–1) in air are measured, and their standard deviations of spectral fitting residual are 5.6 × 10–5 and 7 × 10–5, respectively. Subsequently, the concentration of CO2 and CH4 in air are monitored on-line by the WM-DAS method integrated with the Herriott cell, and compared with those by the highly sensitive continuous wave cavity ring down spectroscopy (CW-CRDS). The experimental results show that the measured results of the long optical path WM-DAS method are consistent with those by the CW-CRDS method, and the linear correlation between the two methods is above 0.99. The detection limit of CO2 and CH4 by the WM-DAS method are 170 ppb and 1.5 ppb respectively, which are slightly higher than those by the CW-CRDS. However, the measurement speed by WM-DAS is much higher than that by CW-CRDS, and possesses the advantages of simpler operation, lower environmental requirements, long-term stability, etc. Keywords:wavelength modulation-direct absorption spectroscopy/ cavity ring down spectroscopy/ trace gas monitoring/ calibration free/ CH4
其中c为光速; P为气体总压; S(T)为线强, T为气体温度; X为待测气体摩尔分数; φ(v)是线型函数; τ0为空腔衰荡时间, 其值取决于镜面反射率、散射、吸收等导致的损耗. 在较窄的波长范围内, τ0可认为是常数, 因此吸收光谱的拟合实际只需对(cτ)–1拟合即可. 实验中采取连续扫描激光电流的方式来改变激光波长, 同时高速扫描腔长以使任意波长的激光均能与腔模式耦合[27], 从而得到周期性的、蕴含气体吸收的衰荡时间信号, 对该信号采用平均或者傅里叶滤波的方式以提高信噪比, 最终可得到衰荡时间τ与激光电流i的关系, 如图4(a)所示. 由于采用连续扫描激光电流的方式, 因此激光相对波长v可采用标准具进行标定, 从而得到v与i的关系. 最后, 根据电流i可建立起衰荡时间τ与相对波长v的关系τ(v), 再结合(7)式, 即可得到吸收系数κ(v), 对其拟合[23,24]可得到气体的光谱参数(如温度、浓度等). 图 4 采用CW-CRDS在298 K, 100.9 kPa下测量的CO2 (红色)和CH4 (蓝色)两条谱线(用时约24 min) (a) 衰荡时间与电流的关系; (b) 吸收率函数及其Voigt拟合 Figure4. The absorption spectra of CO2 (red) and CH4 (blue) measured by CRDS in about 24 min at 298 K and 100.9 kPa: (a) The relationship between the ring down time and the current; (b) the absorption function and the best fits of Voigt profile.
如图5(a)所示, 随着气体浓度的升高, 在中心波长处, 衰荡时间减小导致衰荡信号的采样点减少, 以及腔的出射光强减弱[27], 均会降低CW-CRDS信噪比, 增大测量的离散度(如图5(b)所示). 长光程的WM-DAS出射光信号强, 不受光强减弱影响, 信噪比与气体浓度正相关, 且浓度上限更高, 量程范围大. 这表明CW-CRDS方法最适合低浓度的区域A, 在较低浓度的区域B, 两种方法测量结果一致. 大气中CH4的含量(约1.7 ppm)处于区域B, 两种方法均可测量, 但长光程WM-DAS测量速度(0.1 s)更快、系统简单, 更适合在线监测. 图 5 (a) 两种方法的CH4量程对比; (b)两种方法在不同CH4浓度下的直方图 Figure5. (a) Comparison of measuring range of CH4 between the two methods; (b) histograms of two methods at different concentrations of CH4.
-->
4.1.大气痕量气体CH4和CO2在线监测方案
如图6所示, 两种方法(WM-DAS和CW-CRDS)采取分时的方式实现大气痕量气体的在线监测. 一次完整的测量用时为T0, 约60 s. WM-DAS采用正弦波扫描激光电流(黑色)得到透射光It (蓝色), 其中, 正弦波周期Td为1 ms, 采集个数为1000. 实际中由于数据量较大, 因而每次采集100个并处理, 处理时间为td (约1.5 s), 采集10次共耗时约15 s, 浓度取10次测量的平均值. CW-CRDS采用三角波扫描激光电流的方式来改变激光波长, 得到周期性的、蕴含气体吸收的衰荡时间信号(红色), 周期为Tr (约14.5 s), 采集个数为3, 处理时间为tr (约1.5 s), 共计约45 s. 通过一个完整测量过程, 即可得到两种方法测得的气体浓度, 重复该测量过程即可实现两种方法在线监测气体浓度. 图 6 WM-DAS与CW-CRDS方法联合测量的时序图、激光电流(黑色实线)、透射光强It (蓝色实线)、衰荡时间(红色实线) Figure6. Time sequence diagram of WM-DAS and CW-CRDS, laser current (black solid line), transmitted light It (blue solid line), ring down time (red solid line).
24.2.大气中CO2和CH4连续监测结果 -->
4.2.大气中CO2和CH4连续监测结果
如图7(a)所示, 室外CO2平均浓度约420 ppm, 白天CO2浓度升高, 这可能与测量点附近机动车尾气排放、人类活动有关, 这与文献[28]测得的结果一致. 从图7(b)可看出, WM-DAS和CW-CRDS方法测得的浓度一致性较好, 两组数据的线性拟合相关度达到0.99. 室内CO2平均值高于室外且浓度变化快, 与人类活动关系更大. 两种方法的室内测量数据有一定差异, 这主要是因为分时测量的方式使两种方法存在一定的测量延迟, 浓度快速变化时延迟更明显, 这种差异可以通过优化硬件和程序进一步减小. 图 7 (a) 两种方法测量的大气中CO2浓度; (b) 两种方法测量数据的线性拟合 Figure7. (a) CO2 in atmosphere measured by the two methods; (b) linear fitting of the data measured by the two methods.
如图8(a)所示, 室外CH4平均浓度在1.7 ppm左右, 这与文献[29]测得的结果一致. 两次CH4浓度升高均在0点以后, 这可能与土壤微生物活动有关[30], 而2019年9月17日18点到22点也有升高, 这可能与周围机动车排放有关. 从图8(b)可看出, WM-DAS与CW-CRDS方法测得的浓度一致性较好, 两组数据的线性拟合相关度达到0.994. 图 8 (a) 两种方法测量的大气中CH4浓度; (b) 两种方法测量数据的线性拟合 Figure8. (a) CH4 in atmosphere measured by the two methods; (b) linear fitting of the data measured by the two methods.