Rashba-type spin splittings in the surface electronic structure of heavy elements and topological insulators are a hot topic of today’s research in condensed matter physics. The interest is guided by possible applications of these materials in spintronic devices, in which the electron spin is used as an information carrier.

In this talk, I will present several experimental studies of surface states of different origin and with distinct spin configurations. At W(110), a spin-polarized Dirac-cone-like surface state with d character was identified, which appears below the Fermi level in a spin-orbit-induced symmetry gap of the projected bulk-band structure [1]. At Tl/Si(111), an unoccupied surface state with a spin-dependent energy splitting of more the 0.5 eV exhibits a distinct spin structure around the K point, leading to almost complete spin polarization at the Fermi level. Furthermore, the spin-dependent unoccupied electron states of the topological insulator Bi₂Se₃(111) were studied as a function of different preparation conditions. Making combined use of direct and inverse photoemission, we were able to characterize the electronic states below and beyond the Fermi level.

[1] K. Miyamoto et al., Phys. Rev. Lett. 108, 066808 (2012).

Three-dimensional (3D) topological insulators (TI) are a new state of matter with a bulk band gap but topologically protected gapless surface states. These protected surface states are massless helical Dirac fermions which are predicted to host many striking quantum phenomena [1]. Angle resolved photoemission spectroscopy (ARPES) and scanning tunneling microscopy (STM) measurements confirmed the existence of these surface states and their helical spin structure [1]. All the 3D TI materials to date have an initial doping level which places the Dirac point away from the Fermi level. Initial studies used chemical doping to align the surface states within the bulk band gap. A preferable method to realize the host of new phenomena in TI materials is to electrically tune the carrier density using the field effect from a gate electrode, as demonstrated in three terminal transport experiments [2]. However, the combination of local probe studies with samples containing low defect concentrations and a tunable carrier density remains a challenge, due to the chemical reactivity of the TI surfaces which precludes ex-situ fabrication and processing of the unprotected films.

In this talk we present new results on atomically flat Bi₂Se₃ and Sb₂Te₃ films grown on SrTiO₃ substrates using Molecular Beam Epitaxy (MBE). SrTiO₃ has a very large dielectric constant of ~10⁴ at 4 K [3], allowing tuning of the TI Dirac point and carrier density even with a relatively thick dielectric of 100 μm. The SrTiO₃ substrates were pre-patterned with platinum electrodes and mounted in specially designed sample holders, allowing us to in-situ control the carrier density with a back gate on in-situ grown films, avoiding any ex-situ post processing of the samples. As a result, we are able to continuously change the carrier density and observe the local electronic structure of pristine grown TI films. Initial measurements at 5 K are focused on very thin films of 2 to 10 quintuple layers. Scanning tunneling spectroscopy measurements of the thin film’s surface electronic structure allow us to study the gate's efficiency vs. local film thickness in a single sample. We find that the efficiency of gating the top surface state’s electronic structure depends on the film thickness, with a decreasing efficiency for thicker films. In addition, we observe substantial differences in gating between Bi₂Se₃ and Sb₂Te₃.We will discuss these results and models of the gating of carriers in the bottom and top surface states through the bulk films at different bulk carrier densities.

[1] Rev. Mod. Phys. 82 3045 (2010)

[2] Nano Lett., 10 (12), 5032 (2010)

[3] Phys. Rev. B 19, 3593–3602 (1979)

In this work, we use an ultra-low temperature scanning tunneling microscope [4] to investigate the superconducting properties of a cleaved Cu_xBi₂Se₃ bulk crystal. The crystal was synthesized by electrochemical intercalation of Cu atoms into a previously synthesized Bi₂Se₃ crystal. We observe a superconducting gap in scanning tunneling spectroscopy (STS) measurements. We estimate the size of the gap to be 0.35 meV from a preliminary BCS fit of the superconducting gap. STS measurements under a magnetic field show a complete suppression of the superconducting gap at a critical field of ≈ 1.5 T. Significant inhomogeneity is observed in the material with spatial variations of the superconducting gap. We will discuss these observations in the context of current theories of topological superconductors.

[1] Y. S. Hor et al., Phys. Rev. Lett. 104, 057001 (2010)

[2] L. A. Wray et al., Nat. Phys. 6, 855–859 (2010)

[3] S. Sasaki et al., Phys. Rev. Lett. 107, 217001 (2011)

[4] Y. J. Song et al., Rev. Sci. Instrum. 81, 121101 (2010)