Effects of Particles on Trace-Gas Measurement Using Open-Path Cavity Ring-Down Spectroscopy

Abstract:

Open-path Cavity Ring-down Spectroscopy offers many potential advantages over traditional closed-path configurations for the measurement of atmospheric trace gasses. Removal of the vacuum pump and flow system may enable more compact instruments suitable for remote and mobile deployments as well as real time measurement of ‘sticky’ gases. However, open path operation introduces new challenges including exposure of high reflectivity mirrors to ambient air and aerosols, the need to measure wider (pressure broadened) spectral peaks and possible signal interferences due to optical extinction by aerosol particles in the cavity laser beam. The present submission focuses on the effects of aerosol particles on open-path CRDS using a near-infrared (1742 nm) methane gas measurement system as a test bed. A simple purge enclosure system was developed to prevent aerosol deposition on the cavity high-reflectors. The purge uses ambient air pulled in with a micro-pump through a hepa filter and maintained mirror reflectivity R>0.99996 over 100 hours of use in the presence of high aerosol loading. Optical extinction due to ambient aerosols can change the cavity loss and influence the recorded ring-down times. We observed relatively large fluctuations due to supermicron particles and a near-constant baseline shift due to smaller submicron particles. The fluctuations correspond to absorption on the order of 10^-8–10^-7 cm^-1, comparable to the amplitude of the targeted methane absorption features, causing significant interference. Simple software filter approaches were developed to counter these fluctuations without a priori knowledge of the ambient aerosols. The filters exploit the statistical distribution of signals as well as the expected absorption lineshape. Using these filters, noise-equivalent sensitivities within a factor of ~3 of closed-path systems were obtained (4x10^-10cm^-1Hz^-1/2). Outdoor open-path measurements were validated with side-by-side measurements with a commercial closed path system and showed agreement to <0.3 ppmv for measurement times of 10 s. We also present computations based on Mie theory to examine similar effects for our newer open-path CRDS sensors for detection of methane and ammonia in the mid-infrared.