Low-dimensional Bismuth-induced structures on III-V semiconductor surfaces

Controlling the surface morphology and chemical interaction is necessary for the development of advanced quantum
materials, particularly in low-dimensional systems. As the research increasingly focuses on miniature and nanoscale
technologies, a fundamental understanding of localized or interface states becomes essential for successful integration
into functional devices. This thesis investigates atomic bismuth (Bi)-induced surface structure and chemical
composition, enabling electronic band engineering in the material system. Bi is of particular attraction due to its strong
spin–orbit coupling and inherent topological properties, making it a promising element in the design of next-generation
quantum materials.
In the course of this doctoral research, I have explored atomic-scale Bi deposition on III–V semiconductor surfaces,
specifically on InSb and InAs planar surfaces and GaAs nanowires (NWs). Bi incorporation was studied using a
combination of surface science techniques, including scanning tunneling microscopy/spectroscopy (STM/S), X-ray
photoemission spectroscopy (XPS), angle-resolved XPS (ARPES), and X-ray photoemission electron microscopy
(PEEM).
Surface termination plays a critical role in Bi incorporation into InSb and InAs by influencing nucleation and the resulting
atomic structures. On the InSb(111)A surface, Bi deposition results in a large-scale periodic structure and which grows
as an epitaxial film. STM/S measurements reveal localized trimer states, which induce Rashba-type splitting in the
electronic bands. In contrast, Bi incorporation into InSb(111)B surface shows a locally ordered structure (lacking largescale
periodicity) but results in a Bi-Sb interface layer of self-limiting thickness at elevated temperatures. Unlike the
continuous thickness increase seen in the Bi/InSb(111)A, the Bi/InSb(111)B interface layer thickness remains constant
under subsequent deposition cycles. A similar type of spin-polarized splitting and two-dimensional (2D) metallic surface
states are observed when a Bi-As layer grows on the InAs(111)B surface. The Bi-As layer inhibits the island growth and
induces structural asymmetry, leading to giant Rashba splitting.
Besides 2D structures, Bi incorporation also leads to periodic one-dimensional (1D) atomic chains, which are
investigated here on the non-polar InSb(110) surface. The Bi/InSb(110) surface has self-assembled and periodic 1D
atomic chains of Bi. The band dispersion along the chains shows giant Rashba splitting, while a quasi-1D state exists
across the chains. The Bi chains form stable In-Bi and Bi-Sb bonds with the underlying substrate. The In-Bi epitaxial
growth has been unsuccessful in the past; thus, this work offers an alternative path for the growth of III–Bi compounds.
Beyond the planar surface, this work examines atomic Bi incorporation into zinc blende (Zb) and wurtzite (Wz) segments
of GaAs heterostructure NWs. Bi deposition on Zb and Wz facets has been reported to form self-selective growth and
different incorporation mechanisms. Using PEEM, I confirmed the difference in chemical interaction of Bi within the Zb
and Wz phases. The core level analysis reveals site-selective Bi and Ga signal variation, as well as differing
incorporated mechanisms.
This research highlights the critical role of the underlying substrate structure in the Bi incorporation mechanism and the
resulting low-dimensional electronic states, ranging from localized states to 1D/quasi-1D states. Understanding these
states at the atomic level paves the way for controllable quantum architecture. Overall, it demonstrates that Bi
incorporation is a powerful tool for electronic band engineering, and the Bi/III-V material systems are a suitable
candidate for future quantum and spintronic devices.