Abstract: The Arctic sea-ice cover is declining with global warming, which bears significant impacts on global thermohaline circulation, biogeochemistry process and climate changes. To comprehensively understand the Arctic environment and to predict its changes in the future, it is important to reconstruct the paleo-sea ice variability for the region. In the recent decade, a newly developed sea-ice proxy IP₂₅(an Ice Proxy with 25 carbon atoms), monounsaturated highly branched isoprenoid (HBI) alkene biosynthesized specifically by sea-ice associated diatoms only found in Arctic and sub-Arctic marine sediments, has been universally used to reconstruct the sea-ice variability. Since the first use of IP₂₅ as a proxy for paleo-sea ice, more and more laboratories have measured this biomarker in Arctic and subarctic sediments to verify the application of IP₂₅ to sea ice reconstruction. In this review, we firstly summarized the traditional indicators for sea ice reconstruction and their limitations, and then described is the scientific basis for IP₂₅ proxy and its development from qualitative description to quantitative calculation as well as it limitations. Secondly, we summarized the case studies of using IP₂₅ to reconstruct the distribution and variation of the sea ice. These studies spatially cross the Central Arctic Ocean, Arctic marginal seas, Arctic estuaries and the subarctic regions, and temporally cover the time scales of the modem times, the Holocene, the Quaternary and the Miocene. A good linear correlation between reconstructed modern sea-ice concentrations by using IP₂₅ and satellite-derived spring sea ice concentrations has been observed, providing the basis for paleo-sea ice reconstruction, which may provide important evidence and insight for numerical simulation of paleoclimate and future sea ice prediction.