The grain connectivity calculated using resistivity measurements has been investigated using in-situ PIT-processed tapes. The effective current area (1/F) of the tapes, in cases where commercial Mg powder was used, was only 4.3%. On the other hand, 1/F values of tapes using MgH2 and nano-Mg powders were larger than those using Mg powder. 1/F values of C-doped tapes decreased with increasing amounts of C in cases where MgH2 powder was used. Simultaneous addition of 3-ethyltoluene and SiC were effective in preventing the suppression of connectivity, because the packing density was increased using 3-ethyltoluene.
The critical current density and electrical resistivity of polycrystalline MgB2 samples vary largely among samples. In the present paper, the limiting mechanism of normal-state conductivity in polycrystalline MgB2 was studied by measuring the resistivity of a series of MgB2 bulk samples with systematically varied packing factors. The packing factor dependence of phonon term resistivity was found to be well explained using a three-dimensional site-percolation model. The porosities and the wetting oxide phases at grainboundaries are suggested to be the main causes of poor electrical connectivity in polycrystalline MgB2 samples. Our model provides quantitative evaluations of the influences of such secondary phases and anisotropy in randomly oriented polycrystalline samples, as well as the fraction of the active grains that can carry current. We argue that the suppressed critical current density in rather weak-link-free MgB2 can be understood under the scenario of a seriously percolating current path.