It is only recently that there has been growing awareness of the potential impact of biogeochemical processes within the sea ice itself on the climate system. At the foreground lies the concept that, in a doubling anthropogenic CO₂ scenario, the polar oceans might remain the only significant sink at the Earth’s surface. There is therefore an increasing interest for understanding better what the role of sea ice might be in controlling both ‘passively’ (transfer) and ‘actively’ (production/consumption) the cycles of climatically significant gases between the ocean and the atmosphere. A whole suite of techniques has been set up and tested to measure pCO₂ values and gradients in ice, brines, atmospheric and oceanic boundary layers as well as snow/ice–air CO₂ fluxes at the local and regional scale. These show that, although modest in intensity, the air–ice CO₂ sink might amount to nearly 50% of the new air–ocean estimates in the Southern Ocean, south of 50° S. Understanding fluxes requires knowledge of what drives pCO₂ gradients within the sea ice: besides the purely physical concentration/dilution of the brine inclusions, two other processes were shown to be involved, namely CaCO₃ precipitation/dissolution and algal photosynthesis (i.e. primary production). This explains the interest in: (1) actually finding CaCO₃ crystals in sea ice and demonstrating the feasibility of the ‘calcium carbonate sea-ice pump’; and (2) understanding better the controls on primary production of sympagic organisms. In this regard, recent progress on reliable micronutrients analyses (e.g. iron), stables isotopes measurements (C, Si, N, Fe), microprobes (O₂), speciation of organic matter, have brought knowledge forward. Dimethylsulfide (DMS) is another climatically active compound that has been shown to reach concentrations two to three orders of magnitude higher in sea ice than in surface waters. Process and experimental studies are still attempting to decipher the complexity of this sulfur cycle in sea ice. Other climatically significant gas compounds such as N₂O and CH₄ still remain nearly unexplored in sea ice. It is also only very recently that the first attempts have been made to develop models integrating the new physical aspects of brine and impurity transfer in sea ice during growth and decay (mushy layer convection vs diffusion) with biological processes to reconstruct the time evolution of biogeochemical profiles (e.g. nutrients) within the sea-ice cover. This keynote address aims at summarizing those recent developments in sea-ice biogeochemistry and discussing directions to go in the near future.