Mercury is a persistent, toxic and bio-accumulative pollutant of global interest. Its main mass in the troposphere is in the form of elemental gas-phase mercury. Rapid, near-complete depletion of mercury has been observed during spring in the atmospheric boundary layer of frozen marine areas in Arctic, sub-Arctic and Antarctic locations. It is strongly correlated with ozone depletion. To date, evidence has indicated strongly that chemistry involving halogen gases from surface sea-salt is the mechanism of this destruction. Precisely which halogen gases are the main players has remained unresolved. Our novel kinetic data and multiscale modelling show that Br atoms and BrO radicals are the most effective halogens driving mercury oxidation. The reduction of oxidized mercury deposited in the snow pack back to Hg⁰ and subsequent diffusion to the atmosphere is observed. However, it cannot compensate for the total deposition, and a net accumulation occurs. We use a unique global atmospheric mercury model to estimate that halogen-driven mercury depletion events result in a 44% increase in the net deposition of mercury to the Arctic. Over a 1-yr cycle, we estimate an accumulation of 325 tons of mercury in the Arctic

We investigated the springtime temporal dynamics of both total mercury (Hg) and gaseous Hg in snowpacks from the High Arctic. In situ incubation experiments of snow samples indicated that the production of volatile mercury in snow (VMS) was photomediated and occurred in the first 3 cm of snow. The newly produced VMS (consisting mainly of elemental Hg) was partly oxidized back to Hg(II) when light intensity declined or in the absence of UV radiation, probably through a chain of reactions involving photo-induced radicals and organic compounds in the surface snow. During a 2 week monitoring of total Hg in surface snow, we observed a sharp increase in total Hg concentrations, reaching levels 11 times higher than background concentrations, likely as a result of an atmospheric mercury depletion event. Stratigraphic depth profiles indicated that this increase was restricted to the first 2 cm of the snowpack. Total Hg levels subsequently decreased by 92%, reaching background concentrations within 2 days after this event. The photoproduction rate of VMS calculated on the basis of this episode could account for subsequent daily loss of total Hg from the surface of the snowpack.

A B S T R A C TMercury is a persistent, toxic and bio-accumulative pollutant of global interest. Its main mass in the troposphere is in the form of elemental gas-phase mercury. Rapid, near-complete depletion of mercury has been observed during spring in the atmospheric boundary layer of frozen marine areas in Arctic, sub-Arctic and Antarctic locations. It is strongly correlated with ozone depletion. To date, evidence has indicated strongly that chemistry involving halogen gases from surface sea-salt is the mechanism of this destruction. Precisely which halogen gases are the main players has remained unresolved. Our novel kinetic data and multiscale modelling show that Br atoms and BrO radicals are the most effective halogens driving mercury oxidation. The reduction of oxidized mercury deposited in the snow pack back to Hg 0 and subsequent diffusion to the atmosphere is observed. However, it cannot compensate for the total deposition, and a net accumulation occurs. We use a unique global atmospheric mercury model to estimate that halogen-driven mercury depletion events result in a 44% increase in the net deposition of mercury to the Arctic. Over a 1-yr cycle, we estimate an accumulation of 325 tons of mercury in the Arctic.

No abstract

Ozone is assumed to be the predominant tropospheric oxidant of gaseous elemental mercury (Hg0(g)), defining mercury global atmospheric lifetime. In this study we have examined the effects of two atmospherically relevant polar compounds, H2O(g) and CO(g), on the absolute rate coefficient of the O3-initiated oxidation of Hg0(g), at 296 +/- 2 K using gas chromatography coupled to mass spectrometry (GC-MS). In CO-added experiments, we observed a significant increase in the reaction rate that could be explained by pure gas-phase chemistry. In contrast, we found the apparent rate constant, k(net), varied with the surface-to-volume ratio (0.6 to 5.5 L flasks) in water-added experiments. We have observed small increases in k(net) for nonzero relative humidity, RH < 100%, but substantial increase at RH > or = 100%. Product studies were performed using mass spectrometry and high resolution transmission electron microscopy coupled to an electron dispersive spectrometer (HRTEM-EDS). Our results give evidence for enhanced chain growth of HgO(s) on a carbon grid at RH = 50%. A water/surface/ozone independent ozone oxidation rate is estimated to be (6.2 +/- (1.1; tsigma/ radicaln) x 10(-19) cm3 molecule(-1) s(-1). The total uncertainty associated with the ensemble of experiments amount to approximately < or = 20%. The atmospheric implications of our results and the effect of an added reaction partner in homogeneous and heterogeneous atmospheric chemistry will be discussed.