Surface and Bulk Transport of Samarium Hexaboride

Abstract

There has been a renewed interest in SmB6 during the past several years after the theoretical prediction that it is a topological Kondo insulator (TKI). Soon after, the conducting surface was experimentally discovered, which is the key initial step for the TKI verication. Motivated by this work, this dissertation further studies both the surface and the bulk properties of SmB6 using electrical transport methods. To study the surface transport of SmB6, choosing the appropriate transport geometry is extremely important. A Corbino disk geometry, which confines the current path to a single surface, was used in this study. The measurements from the Corbino disk resulted in more physically acceptable values of carrier density than the results that were obtained from other conventional transport geometries. During this study, we also found that subsurface cracks that are created during surface preparation and domain boundaries of a polycrystal can provide unwanted conduction paths. After careful surface preparation, the magnetotransport was measured with an applied magnetic field of 34.5 T and a temperature of 0.3 K. A higher magnetic field was also applied up to 93 T, using the pulsed magnetic field, but Shubnikov de-Haas oscillations were not observed. Therefore, the strongest signatures needed to verify the nontrivial topology, such as the half-integer Landau index in the fan diagram have not yet been seen, nor were we able to find signatures of the three Fermi pockets predicted by theory. Instead, the Corbino disk magnetotransport results were consistent with the case when only one Fermi pocket exists. This can happen when the mobilities for each of the pockets are too small. From analyzing our magnetotransport, possible ranges of carrier density and mobility of each of the three Fermi pockets could be constructed in a 2D parameter space. When comparing with these Fermi pocket ranges that were constructed by magnetotransport to other experimental reports, only the angle-resolved photoemission spectroscopy reports are within the X-pocket range. To study the bulk of SmB6, a new transport method was invented, called the inverted resistance measurement. This inverted resistance measurement can be used for studying the bulk of SmB6 when the surface conduction dominates. By comparing the numerical simulation and the experimental measurement, the bulk resistivity was successfully extracted in the low-temperature range, where previously this was impossible due to the overwhelming surface conduction. Using this new method, we investigated the bulk of pure and disorder-induced SmB6 samples. We find that the bulk of SmB6 is an ideal insulator. The thermally activated behavior, with an activation energy of 4.01 meV, continues for ten orders of magnitude in resistivity. This thermally activated bulk behavior is found to be almost identical in the highly disordered SmB6 samples. These results suggest that the bulk of SmB6 has an energy gap that is free of impurity states, making the bulk transport remarkable. In addition to the thermally activated bulk behavior, in our disorder-induced samples, our measurements revealed a mysterious bulk resistivity plateau that has been buried under the surface conduction. This bulk plateau cannot be understood by normal impurity states. Instead, a new understanding of impurity and disorder may be required for the case of SmB6.