putting a spin on topological matter

Joris Voerman is a PhD student in the research group Interfaces and Correlated Electrons. His supervisor is prof.dr.ir. A. Brinkman from the Faculty of Science and Technology.

One of the most promising ways of creating computational devices beyond the limits of the current semiconductor technology, is being able to utilize the properties of the electron spin. This field of technology has been named spintronics. In this light, topological matter comes to mind since it has a fundamental property called spin-momentum locking. The fact that the spin orientation is directly coupled to the momentum direction of the charge carrier, could be the key to unlocking the field of spintronics. This thesis explores the role that the electron spin has on the electronic properties of a few recently discovered topological materials: BiSbTeSe2, a topological insulator designed to have very little contribution of the bulk to the electronic transport. Secondly, I study ZrSiS, a nodal-line semimetal that shows hints of a trivial to topological transition under a certain magnetic field. And finally, I investigate PdTe2, a material known to be both a superconductor and a Dirac semimetal, which makes it a good candidate for being a highly sought-after topological superconductor.

In chapter 3 the spin-momentum locking in BiSbTeSe2 is shown experimentally. To this end, I have placed topological insulator flakes on a strip of graphene, that can transport the spin-polarization because of its long spin-relaxation length. These devices allow for the creation of spin-polarization and its detection in a non-local configuration. By changing the magnetization of ferromagnetic leads with a magnetic field, the spin-polarization underneath the ferromagnetic lead was measured through the observed tunnel magnetoresistance. The experiments show that the spin-polarization changes linearly with the applied bias current, until increased scattering due to Joule heating diminishes the polarization. This experiment shows that spin-momentum locking indeed generates a discernible spin-polarization. Furthermore, the tuning of the Fermi level through backgating shows that the spin-polarization remains qualitatively unchanged when the carrier type is changed from electrons to holes.

Because of spin-momentum locking elastic scattering is suppressed in a topological insulator. The hyperfine interaction then plays a more important role in scattering processes. Effectively, this type of scattering transfers the electron spin to the bismuth nuclei. When the nuclear spins relax they again polarize the electrons, making the system act like a capacitor, or a battery. A square wave with a minimum voltage of zero is applied to the device while the voltage response of the device is measured. The extracted RC-time is a measure for the capacitance that is present in the set-up. The experiments described in chapter 4 show no capacitance beyond the calibration, providing an upper limit to the performance of a topological insulator flake-based spin battery. In our measurements this upper limit was determined to be 0.1 nF.

Chapter 5 turns its attention to the nodal-line semimetal ZrSiS. Several research groups have reported on its curious angle-dependent magnetoresistance, called the butterfly magnetoresistance. By studying the magnetoresistance of several ZrSiS flakes from two different crystals, near-perfect electron hole compensation has been identified in this chapter as the origin of the buttery magnetoresistance. For certain angles of the magnetic field the Zeeman shift slightly distorts the bands, which produces these equal, but opposite, carrier densities. In the two-band Drude model this leads to a large magnetoresistance. The beating pattern in the Shubnikov-de Haas oscillations supports this conclusion. Additionally, the Berry phase that is extracted from the Lifshitz-Kosevich fitting of quantum oscillations, increases with angle and hints towards a link between the buttery magnetoresistance and the topological properties of ZrSiS.

In the final chapter the order parameter symmetry of the superconductivity in PdTe2 is investigated by tunneling spectroscopy. Even though PdTe2 has been firmly established as a superconductor and a Dirac semimetal, it remains uncertain whether it also exhibits topological superconductivity. The experiments performed on Josepshon junctions in the ab-plane of PdTe2, combined with a theoretical model of the experiments, point to an important conclusion: in transparent Josephson junctions the conductance spectrum of conventional superconductivity can be nearly indistinguishable from unconventional superconductivity. The addition of effects related to the critical current in the disordered region near the interface to the model, is required to properly interpret the conductance spectra. In the conductance spectrum obtained from most opaque junction, there is a small feature that could not be explained without the addition of 4% helical p-wave symmetry to the model. This hints to the presence of unconventional superconductivity in PdTe2.