In situ and denuder-based measurements of elemental and reactive gaseous mercury with analysis by laser-induced fluorescence – results from the Reno Atmospheric Mercury Intercomparison Experiment

摘要

The University of Miami (UM) deployed a sequential two-photon laser-induced fluorescence (2P-LIF) instrument for the in situ measurement of gaseous elemental mercury, Hg(0), during the Reno Atmospheric Mercury Intercomparison Experiment (RAMIX) campaign. A number of extended sampling experiments, typically lasting 6–8?h but on one occasion extending to ～?24?h, were conducted, allowing the 2P-LIF measurements of Hg(0) concentrations to be compared with two independently operated instruments using gold amalgamation sampling coupled with cold vapor atomic fluorescence spectroscopic (CVAFS) analysis. At the highest temporal resolution, ～?5?min samples, the three instruments measured concentrations that agreed to within 10–25?%. Measurements of total mercury (TM) were made by using pyrolysis to convert total oxidized mercury (TOM) to Hg(0). TOM was then obtained by difference. Variability in the ambient Hg(0) concentration limited our sensitivity for measurement of ambient TOM using this approach. In addition, manually sampled KCl-coated annular denuders were deployed and analyzed using thermal dissociation coupled with single-photon LIF detection of Hg(0). The TOM measurements obtained were normally consistent with KCl denuder measurements obtained with two Tekran speciation systems and with the manual KCl denuder measurements but with very large uncertainty. They were typically lower than measurements reported by the University of Washington (UW) Detector for Oxidized Hg Species (DOHGS) system. The ability of the 2P-LIF pyrolysis system to measure TM was demonstrated during one of the manifold HgBrsub2/sub spikes but the results did not agree well with those reported by the DOHGS system. The limitations of the RAMIX experiment and potential improvements that should be implemented in any future mercury instrument intercomparison are discussed. We suggest that instrumental artifacts make a substantial contribution to the discrepancies in the reported measurements over the course of the RAMIX campaign. This suggests that caution should be used in drawing significant implications for the atmospheric cycling of mercury from the RAMIX results.