This paper reviews recent advances in the research and development of sorbents used to capture mercury from coal-fired utility boiler flue gas. Mercury emissions are the source of serious health concerns. Worldwide mercury emissions from human activities are estimated to be 1000 to 6000 t/annum. Mercury emissions from coal-fired power plants are believed to be the largest source of anthropogenic mercury emissions. Mercury emissions from coal-fired utility boilers vary in total amount and speciation, depending on coal types, boiler operating conditions, and configurations of air pollution control devices (APCDs). The APCDs, such as fabric filter (FF) bag house, electrostatic precipitator (ESP), and wet flue gas desulfurization (FGD), can remove some particulate-bound and oxidized forms of mercury. Elemental mercury often escapes from these devices. Activated carbon injection upstream of a particulate control device has been shown to have the best potential to remove both elemental and oxidized mercury from the flue gas. For this paper, NORIT FGD activated carbon was extensively studied for its mercury adsorption behavior. Results from bench-, pilot- and field-scale studies, mercury adsorption by coal chars, and a case of lignite-burned mercury control were reviewed. Studies of brominated carbon, sulfur-impregnated carbon and chloride-impregnated carbon were also reviewed. Carbon substitutes, such as calcium sorbents, petroleum coke, zeolites and fly ash were analyzed for their mercury-adsorption performance. At this time, brominated activated carbon appears to be the best-performing mercury sorbent. A non-injection regenerable sorbent technology is briefly introduced herein, and the issue of mercury leachability is briefly covered. Future research directions are suggested. (C) 2007 Published by Elsevier B.V.