The chemistry of reactive halogens in the polar atmosphere plays important roles in ozone and mercury depletion events, oxidizing capacity, and dimethylsulfide oxidation to form cloud-condensation nuclei. Among halogen species, the sources and emission mechanisms of inorganic iodine compounds in the polar boundary layer remain unknown. Here, we demonstrate that the production of tri-iodide (I 3 − ) via iodide oxidation, which is negligible in aqueous solution, is significantly accelerated in frozen solution, both in the presence and the absence of solar irradiation. Field experiments carried out in the Antarctic region (King George Island, 62°13′S, 58°47′W) also showed that the generation of tri-iodide via solar photo-oxidation was enhanced when iodide was added to various ice media. The emission of gaseous I 2 from the irradiated frozen solution of iodide to the gas phase was detected by using cavity ring-down spectroscopy, which was observed both in the frozen state at 253 K and after thawing the ice at 298 K. The accelerated (photo-)oxidation of iodide and the subsequent formation of tri-iodide and I 2 in ice appear to be related with the freeze concentration of iodide and dissolved O 2 trapped in the ice crystal grain boundaries. We propose that an accelerated abiotic transformation of iodide to gaseous I 2 in ice media provides a previously unrecognized formation pathway of active iodine species in the polar atmosphere.

Current chemistry climate models do not include polar emissions and chemistry of halogens.This work presents the first implementation of an interactive polar module into the very short-lived (VSL) halogen version of the Community Atmosphere Model with Chemistry (CAM-Chem) model. The polar module includes photochemical release of molecular bromine, chlorine, and interhalogens from the sea-ice surface, and brine diffusion of iodine biologically produced underneath and within porous sea-ice. It also includes heterogeneous recycling of inorganic halogen reservoirs deposited over fresh sea-ice surfaces and snow-covered regions. The polar emission of chlorine, bromine, and iodine reach approximately 32, 250, and 39 Gg/year for Antarctica and 33, 271, and 4 Gg/year for the Arctic, respectively, with a marked seasonal cycle mainly driven by sunlight and sea-ice coverage. Model results are validated against polar boundary layer measurements of ClO, BrO, and IO, and satellite BrO and IO columns. This validation includes satellite observations of IO over inner Antarctica for which an iodine "leapfrog" mechanism is proposed to transport active iodine from coastal source regions to the interior of the continent. The modeled chlorine and bromine polar sources represent up to 45% and 80% of the global biogenic VSL Cl and VSL Br emissions, respectively, while the Antarctic sea-ice iodine flux is~10 times larger than that from the Southern Ocean. We present the first estimate of the contribution of polar halogen emissions to the global tropospheric halogen budget. CAM-Chem includes now a complete representation of halogen sources and chemistry from pole-to-pole and from the Earth's surface up to the stratopause. Key Points:• The contribution of polar sea-ice sources to the global tropospheric halogen budget has been computed using the CAM-Chem model • The annual halogen flux from the sea-ice surface represents between 45% and 80% of the global chloro-and bromo-carbon oceanic VSL emissions • In Antarctica, the annual iodine sea-ice flux is~10 times larger than the oceanic inorganic iodine sourceSupporting Information:• Supporting Information S1