1. Experimental Physics VI, Center for Electronic Correlations and Magnetism, Institute of Physics, University of Augsburg, D-86135 Augsburg, Germany
2. Low Temperature Division and MESA ⁺ Institute, University of Twente, Box 217, 7500 AE Enschede, The Netherlands

Correspondence to: J. Mannhart¹ Correspondence and requests for materials should be addressed to J.M. (e-mail: Email: jochen.mannhart@physik.uni-augsburg.de).

With the discovery of high-temperature superconductivity¹, it seemed that the vision of superconducting power cables operating at the boiling point of liquid nitrogen (77 K) was close to realization. But it was soon found that the critical current density J _c of the supercurrents that can pass through these polycrystalline materials without destroying superconductivity is remarkably small^{1, 2}. In many materials, J _c is suppressed at grain boundaries^{2, 3, 4}, by phenomena such as interface charging and bending of the electronic band structure^{5, 6, 7, 8, 9}. Partial replacement ('doping') of the yttrium in YBa₂Cu₃O_7- with calcium has been used to increase grain-boundary J _c values substantially, but only at temperatures much lower than 77 K (ref. 9). Here we show that preferentially overdoping the grain boundaries, relative to the grains themselves, yields values of J _c at 77 K that far exceed previously published values. Our results indicate that grain-boundary doping is a viable approach for producing a practical, cost-effective superconducting power cable operating at liquid-nitrogen temperatures.