Sea ice covers about 7% of Earth’s surface at its maximum seasonal extent, representing one of the largest biomes on the planet. For decades, sea ice has been considered by the scientific community as an inert and impermeable barrier to air–sea gas exchanges. However, this assumption is not supported by studies on the ice permeability to gases and liquids, which show that sea ice is permeable at temperatures above –10°C. Recently, uptake of atmospheric CO₂ over sea-ice cover has been reported supporting the need to further investigate pCO₂ dynamics in the sea-ice realm and related CO₂ fluxes. The recent discovery of ikaite, a metastable phase of hydrated CaCO₃ (CaCO₃·6H₂O), has confirmed the presence and form of the carbonate mineral in sea ice. This precipitation of CaCO₃ can significantly affect pCO₂ dynamics and act as a sink for atmospheric CO₂. The INTERICE 4 project ran at the Arctic Environment Test basin in September 2009. It addressed fundamental questions on the biogeochemistry of carbon during ice formation with an emphasis on precipitation of CaCO₃. In the course of the experiment, we followed the pCO₂ changes within brines and underlying water of several ice mesocosms, together with physical parameters of the ice. In addition we measured air–ice CO₂ fluxes with the chamber method. Freezing started on 5 September. We observed a remarkably steady sea-ice growth until 13 September and then sea-ice growth was slightly reduced until the 19 September when it stopped due to the warming of the ice-tank room. The pCO₂ in the brines ranged from 444 to 706 ppm during the ice-growth phase due to concentration of brines but did not show significant increase after 9 September. During the ice-decay phase, pCO₂ showed a dramatic decrease and ranged from 200 to 335 ppm. At the beginning of the ice growth all bags released CO₂ to the atmosphere. Then as the ice temperature decreased, permeability of sea ice decreased and reduced the release of CO₂ to the atmosphere. During most of the ice-growth phase, air–ice CO₂ fluxes were low or below the detection limit of the measurement. At the beginning of the ice-decay phase (21 September onward), we observed a significant uptake of atmospheric CO₂ by the ice in accordance to the step decrease of the pCO₂ of brines, below the mean pCO₂ of the atmosphere of the ice-tank room. However, this uptake decreased significantly afterwards.