We report on the observation of a negative magnetoresistance (NMR) regime in boron-doped nanocrystalline diamond films at low temperatures. A comparative analysis of our experimental results and those reported for systems composed of superconducting granules embedded in an insulating matrix (also referred as granular films) suggest the presence of superconducting regions inside the insulating films as causing the NMR. By considering the latter scenario, the experimental observations are explained by modeling the systems as consisting of a distribution of superconducting granules whose global properties are tuned by the intergrain distance.
Bibliographical noteFunding Information:
This work was supported by the European Science Foundation (ESF), NES programme, the Methusalem Funding of the Flemish Government, the IAP-P6/42 project “Quantum Effects in Clusters and Nanowires,” and GOA and FWO (Contract Nos. G.0430.07 and G.0068.07) projects. FIG. 1. The temperature dependence of the resistance of the boron-doped diamond films grown on quartz plotted on a log-log scale. The samples differ from each other in their boron concentration, which increases when going from q 1 to q 5 . The T C of q 4 and q 5 are 1.8 and 2.2 K, respectively. FIG. 2. The log-linear plot of the R ( T ) of samples q 1 and q 2 . Inset: The W = d ln σ / d ln T vs temperature ( T ) plot of q 1 on a log-log scale. Here, σ is the sample conductivity. The dashed line corresponds to the linear fit. FIG. 3. The Arrhenius plots of samples q 1 and q 2 (inset). Good linear fits were obtained in the temperature range between 20 and 1.2 K. FIG. 4. The magnetoresistance of sample q 1 for temperatures between (a) 0.4 and 1.4 K and (b) 1.6 and 3 K. NMR stands for negative magnetoresistance. FIG. 5. The magnetoresistance for sample q 2 measured between 0.5 and 2.5 K. NMR stands for negative magnetoresistance. FIG. 6. The temperature dependence of the fields at which the NMR → PMR 2 crossover happens for sample q 1 . The dashed line corresponds to the fit of the B 1 values by using a temperature dependence similar to that empirically found for the superconducting critical field. FIG. 7. The R ( B ) data for sample q 2 , taken at 0.5 K, plotted as a function of B on a semilogarithmic scale.