The effect of segregated transition metals on the grain boundary resistivity of 1 mol% Gd-doped ceria has been investigated. The main focus of interest is whether the space charge potential that causes the blocking effect of the grain boundaries of the ceria can be extrinsically modified. The introduction of a small amount (< 0.5 mol%) of transition metals (Fe, Co, Mn and Cu) to 1 mol% Gd-doped ceria results in significant reductions in only the grain boundary resistivities of the samples attributed to exclusive segregation of the transition metals into the grain boundary core. In the case of Co- and Fe-doped samples, the grain boundary resistivity is lowered by an order of magnitude. EELS line scans across the grain boundaries of the Fe-doped sample have indicated that the grain boundaries are free of a secondary phase of transition metal oxide and that the Fe in the grain boundary likely exists as point defects. These results strongly suggest that it is indeed possible to reduce the excess positive charge in the grain boundary core, and thus the grain boundary resistivity in a ceria electrolyte, extrinsically as initially postulated. A defect chemistry model which explains partial counterbalance to the positive grain boundary core charge has been suggested. The resistivity minimum shown for the samples with different Fe concentration indicates that there is an optimum concentration of transition metal in the grain boundary core of the ceria necessary for such a countbalance. (c) 2006 Elsevier B.V. All rights reserved.