Atomic structure of thin films and heterostructure of Bi2Te3 and Bi2Se3 topological insulators

• Main Author: Ghasemi, Arsham
• Other Authors: Lazarov, Vlado
• Published: University of York 2017
• Subjects: 530
• Online Access: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.729542

The atomic structure of transition metals doped three dimensional (3D) topological insulators (TIs) and the bonding nature of Bi2Te3 with FeSe layers and Ge(111) substrate were studied. Motivation behind transition metal doping of 3D TIs is driven by achieving long range ferromagnetism of Bi2Se3 and Bi2Te3, which is expected to give rise to different spintronic effects that can be utilise in device applications. The nature of this magnetisation depends on the location of the dopants in the Bi chalcogenide matrix. Dopants in Bi based TIs can substitute for Bi, Te, or incorporate between the quintuple layers in the van der Waals gap. Long range ferromagnetism is observed in both Cr doped Bi2Se3 and Mn doped Bi2Te3; however, the main goal of achieving room-temperature ferromagnetism in homogeneously doped TIs has proven to be difficult. In this thesis it is shown that 4.6 at-% of Cr is incorporated substitutionally on Bi sites with no phase segregation. The presences of grain boundaries can cause Cr segregation; hence by controlling the defect density a homogeneous Cr distribution could in principle be achieved even at higher concentrations. In case of Mn as a dopant, we show that the local environment of Mn in Bi2Te3 is heterogeneous. The first principal calculations revealed that the Mn dopants ferromagnetically couple in Bi2Te3 lattice. In addition, we have shown that doping of Bi2Te3 with Mn should be limited to low concentrations (< 6 at-%), higher dopants concentrations results in the formation of secondary phases. Next we have demonstrated that epitaxial growth of FexCu1-xSe on Bi2Te3 is possible regardless of their different lattice symmetries and large lattice mismatch of 19%. First-principles energy calculations revealed that this is realised through van der Waals-like bonding between the Se and Te atomic planes at the interface. Finally, we have shown that the weak van der Waals bonding between the Bi2Te3 and Ge(111) substrate can be strengthen by formation of a Te monolayer at the interface. The electronic band structure calculations revealed that this is due to the stronger atomic p-type orbital hybridization at the interface.