[en] The work in this thesis was motivated by the theoretical prediction that a pair of Majorana bound states (MBSs) will be formed in Josephson junctions (JJs), which consist of s-wave superconductors (SCs) and three-dimensional topological insulators (3D TIs). A fractional Josephson effect is expected in such JJs with a 4π-periodic current phase relation. Our JJs are based on the MBE-grwon (Bi${}_{1-<!--\; ???\; -->x}$Sb${}_x$)${}_2$Te${}_3$ (BST) film. Both planar JJs and sandwich JJs were fabricated and characterized by transport measurements. First, we discovered that SCs are formed by magnetron sputtering Pd onto BST film. The transition temperature (T${}_C$) and the upper critical field (H${}_{C2}$) of the SCs were characterized in Hall-bar devices based on Pd/BST bilayer films. To identify the SC phase in our samples, scanning transmission electron microscopy (STEM) Energy-dispersive x-ray spectroscopy (EDX) were performed. STEM and EDX analyses show the result that the self-formed PdTe${}_{1,2}$ SCs are epitaxially formed on the BST film. A planar SC-TI-SC JJ is built when two Pd electrodes are deposited on BST film with a small gap. A supercurrent and a Fraunhofer-like pattern were observed on this JJ, suggesting that BST film is proximitized by self-formed SCs successfully. To characterize the junction transparency, a superconducting quantum interference device (SQUID) was fabricated. A high transparency ∼ 0.96 is obtained based on the Octavio-Tinkham-Blonder-Klapwijk (OTBK) theory. For most of the planar JJs, the interface between SCs and TIs is on the surface of TI. The surface can be polluted by water and oxide layers during nanofabrication. While the interface is inside the TI film and the SCs are epitaxially grown in our device. These are possible reasons for the high junction transparency. However, there is a lack of features for the existence of multiple Andreev reflections, which is usually observed in high-transparency JJs. Epitaxial growth of a superconductor on the TI surface is still challenging. Our discovery of unexpected self-epitaxy to form SCs on the surface of BST makes the nanofabrication of devices with an epitaxial SC/TI interface much more straightforward. These devices provide a good platform to realize the Fu-Kane proposal in TIs to engineer a topological superconducting (TSC) state. Second, we found that SCs can also form when Pd is thermally-deposited onto BST films. The SCs of PdTe and PdTe${}_2$ with an "epitaxial" interface is confirmed by STEM and EDX analyses. It is different from the case of the sputter-deposited Pd. The Pd penetrates fully through the BST film when Pd was deposited by thermal evaporation. This finding opened the possibility to create a TI nanowire sandwiched by SCs, by converting the BST film into the SC and leaving only a narrow BST channel to remain as the pristine TI. TI nanowires are created in this way that is confirmed by STEM. The finding of the complete conversion of BST into SCs also opened the possibility to build SC-TI-SC sandwiched JJs. The TI nanowire was proximitized by self-formed SCs successfully. Standard Fraunhofer patterns were observed on such JJs, suggesting that the supercurrent is not due to a superconducting short-circuit and the distribution of the supercurrent is homogeneous. The junction transparency is about 0.8, and the resonance conductance peaks resulting from multiple Andreev reflection (MAR) were observed. The MAR feature observed with the index up to 9 and at temperatures up to 1.0 K which is close to the transition temperature of self-formed SCs, suggesting that the superconducting proximity effect in JJs is strong. Finally, we did the Shapiro measurements. The m = 1 Shapiro step is clearly missing at low frequencies and low power. Two possible topologically trivial origins of the first step missing are Landau-Zener transitions (LZTs) and hysteretic I - V characteristics. Temperature dependence of the Shapiro steps shows that the m = 1 step is still missing at a high enough temperature where the hysteretic I - V behavior is completely gone. Landau-Zener transition is considered when the junction transparency is close to 1, but our junction transparency is not that high.We proved that the origin of the missing first Shapiro step is neither hysteretic I - V characteristic nor Landau-Zener transitions. These phenomena are reproduced in another two devices. Hence, our Shapiro-step suggest that a 4π-periodic contribution in the sandwich JJs. This gives possible evidence for topological Majorana bound states.

[en] SnTe is a three-dimensional topological crystalline insulator. It is known that superconductivity can be induced in SnTe by indium doping, making In${}_x$Sn${}_{1-<!--\; ???\; -->x}$Te a promising platform for searching for topological superconductivity. Here we present our study on superconducting In${}_x$Sn${}_{1-<!--\; ???\; -->x}$Te nanostructures. In${}_x$Sn${}_{1-<!--\; ???\; -->x}$Te nano-plates and nanowires were grown via vapor-transport method. After the growth, they were fabricated into devices and then measured in a cryostat. A sharp superconducting transition was observed in all devices. Point contact spectroscopy was also taken at low temperatures. The preliminary results are presented and discussed.

[en] With indium doping, superconductivity can be induced in SnTe, a well-known topological crystalline insulator. Due to the possible co-existence of superconductivity and non-trivial band topology, In${}_x$Sn${}_{1-<!--\; ???\; -->x}$Te provides a promising platform for searching for topological superconductivity. In this talk, we present our study on the point-contact spectroscopy of superconducting In${}_x$Sn${}_{1-<!--\; ???\; -->x}$Te nanostructures. In${}_x$Sn${}_{1-<!--\; ???\; -->x}$Te nano-plates were grown on Si/SiO${}_2$ substrates via vapor-transport method. After the growth, Au contacts with naturally formed tunnel barrier were fabricated on selected nano-plates. The conductance spectra of Au contacts were measured in a dilution refrigerator, and non-trivial sub-gap features were observed in several Au contacts at the tunnel limit. The preliminary results are presented and discussed.

[en] Topological superconductor (TSC) has attracted extensive interest because of its potential application to topological quantum computation, which is based on Majorana fermion (MF). Recently, the long-existed quasi-one-dimensional superconductor, Tl${}_2$Mo${}_6$Se${}_6$, was predicted to be a new TSC. Here we report the growth and characterizations of high-quality superconducting Tl${}_2$Mo${}_6$Se${}_6$ crystals and nanowires. Fiber-shaped crystals with typical size of 0.3$\times <!--\; ??\; -->$0.3$\times <!--\; ??\; -->$10 mm${}^3$ were grown by reacting the starting raw materials in a quartz tube. The highest critical temperature (T${}_c$) was observed to be 6.7 K from both resistivity and magnetic susceptibility measurements. The highest shielding fraction was estimated to be 100%. The upper critical field (H${}_{c2}$) behaviour shows an upturn with lowering temperature, which is similar to the results reported before, suggesting unconventional superconductivity. Tl${}_2$Mo${}_6$Se${}_6$ nanowires with typical diameter of 100 nm were obtained by exfoliating Tl${}_2$Mo${}_6$Se${}_6$ crystals, which were used for fabricating devices by e-beam lithography for transport measurements. Superconductivity with T${}_c$ up to 6 K was observed.

[en] In recent years, possible existence of Majorana fermion (MF) in condensed matter has attracted extensive interest because of the potential application of MFs to the realization of topological quantum computation. So far, several strategies have been proposed to create and manipulate MFs based on nanowires of topological insulators (TI) and topological superconductors (TSC). In this contribution, we reported the nanomaterial growth and characterizations of one of the candidate TSCs, In${}_x$Sn${}_{1-<!--\; ???\; -->x}$Te (IST). Nanowires and nanoflakes of IST are grown on SiO${}_2$/Si wafer by using a vapour transport method in sealed quartz tubes with and without gold catalyst. Growth parameters such as the composition of source materials, source temperature, temperature gradient and growth time are optimized to obtain expected nanowires and nanoflakes. Scanning electron microscopy (SEM) and energy-dispersive X-ray spectroscopy (EDX) are used to investigate the morphology and composition of as-grown nanomaterials. For electronic transport characterization, devices based on nanowires and nanoflakes are fabricated by e-beam lithography and photo-lithography, respectively. Electronic transport properties are measured on helium-3 cryostat with the temperature down to 0.3 K.

[en] Josephson junctions (JJs) based on topological insulators (TIs) are interesting because of the possibility to generate Majorana fermions. Previous reports on TI-based JJs have employed deposited superconducting films such as Al and Nb for proximitization, which inevitably faces the challenge of the interface quality. In fact, the transparency of the JJs reported so far has been limited. Here we report that simply sputtering Pd on MBE-grown (Bi${}_{1-<!--\; ???\; -->x}$Sb${}_x$)${}_2$Te${}_3$ (BST) thin films naturally gives us a superconducting layer due to an epitaxial alloy formation at the interface. This superconducting layer allows us to fabricate JJ devices which show dissipationless supercurrents below a critical temperature of 0.9 K, confirming a strong proximity effect on the TI surface.