Abstract
Although a field‐reversed configuration (FRC) has closed magnetic field lines, its particle confinement time is strongly influenced by the flow in the edge layer located outside the separatrix. Previous simple analyses of the edge layer have used fluid models with freely streaming end loss. However, a review of experimental results uncovers major inconsistencies between the fluid picture and actual measurements. In particular, the edge layer is unusually broad; the global particle confinement does not show the expected scaling with FRC length; the magnetic flux imbedded in the edge layer is inconsistent with that in the jet region; and the outflow speed in the jet is anomalously high. The most promising explanation for these anomalies is electrostatic confinement, which is well known in mirror and magnetic cusp plasmas. A tentative description of the electrostatic structure and plasma flow in the edge layer is presented.