Study of Bi-containing III-V alloys as potential candidates for a new 3-dimentional topological insulator

Joint research project

Project leaders: Ernesto Placidi, Jan Honolka
Agreement: REPUBBLICA CECA - CAS (ex AVCR) - Czech Academy of Sciences
Call: CNR-CAS (ex AVCR) 2016-2018
Department: Physical sciences and technologies of matter
Thematic area: Physical sciences and technologies of matter
Status of the project: New

Research proposal

In the past decade a new class of materials called topological insulators (TI) has caught the interest of material scientists, triggered by their fascinating fundamental physics but also by their potential for future spintronic applications [Rev.Mod.Phys. 82, 3045 (2010) and 83, 1057 (2011)]. The physics of TI were first observed in 2D HgTe/HgCdTe quantum well (QW) systems, possessing spin-polarized helical 1D edge states [Science 318, 766 (2007) and 314, 1757 (2006)]. The corresponding class of 3D TI was found in melt-grown chalcogenide-type materials Bi_(1-x)Sb_x, Bi_2Te_3 and Bi_2Se_3 with large spin-orbit coupling.
3D TI systems host 2D topological surface states (TSS), which follow a Dirac-like dispersion relation thereby forming a Dirac point (DP) singularity. Close to the DP solutions +k and -k are strictly locked to opposite spin currents. While spin-polarized currents are usually only achieved in low-dimensional electronic systems under high magnetic fields, chalcogenide-type TI intrinsically develop spin-polarization as a consequence of their large spin-orbit and time reversal symmetry. For spintronic concepts, time reversal symmetry is especially interesting since 180° backscattering is forbidden, allowing electrons to travel free over longer distances along the surface [Nat.Mater. 11, 409 (2012)], and furthermore scattering channels can be opened in the presence of magnetic degrees of freedom which break time reversal symmetry. In the latter case, transport should be inhibited and energy gap is predicted to appear at the DP [Phys.Rev.Lett. 102, 156603 (2009) and 109, 076801 (2012)]. In (Bi, Mn)_2Se_3, a system which is believed to have similar magnetic mechanisms like the ferromagnetic semiconductor (Ga,Mn)As, such effects were intensively studied. In the important field of 2D TSS transport progress is strongly hampered by unwanted contributions of non-topological 3D bulk states due to rather small band gaps in chalcogenides and high density of growth defects. Although molecular beam epitaxy (MBE) growth succeeded in decreasing bulk contribution, defect density is still high. In search for new TI materials, many materials have been proposed by theory ranging from Heusler alloys, oxides, semiconductor heterostructures, and also homogeneous semiconductors with a much wider band gap compared to chalcogenide.
For the bilateral Italian-Czech collaboration we propose a research project based on a recent concept [Phys.Rev.B 90, 195105 (2014)] that predicts a non-trivial topological phase in Bi-based III-V semiconductor. Starting at Bi concentrations higher than 19% the necessary band inversion occurs in GaAsBi alloy, triggered by the large spin-orbit coupling of Bi. At low concentrations the incorporation of Bi in GaAs is known to drastically reduce the energy gap and increase thermal stability, making the alloy one of the most promising material system for light emitters, optical fiber detectors and solar cells. A successful demonstration of a topological insulating phase of GaAsBi at high Bi contents would launch III-V based systems also as a new TI platform adding new functionalities and promising perspectives of the material for spintronics, low-power electronic applications, and integration into various current semiconductor devices.
Thereafter, for x>19%, GaAs_(1-x)Bi_x can be readily driven into the TI phase by external strain or by doping which reduces the symmetry [Phys.Rev.B 90, 195105 (2014)]. Compared to chalcogenide TI, such class of alloys promises a larger potential for purely TSS dominated transport, due to suppressed bulk contributions because of the large energy gap of the host material.
In this project we intend: (i) to perform intensive MBE growth studies; (ii) to apply various material characterization techniques to study structural, chemical, and electronic properties, including the direct search for TI properties.
To improve Bi incorporation we rely on the long-time expertise acquired by the proponents [Phys.Rev.B 78, 155310 (2008), and 81, 094412 (2010) and 83, 094420 (2011), New J.Phys. 12, 093022 (2010), Appl.Phys.Lett. 98, 022503 (2011)] in particular tackling similar problems with the growth of GaMnAs alloys. Indeed, like Mn, the Bi solubility in GaAs is rather low and its stoichiometric incorporation in the crystal is achieved only for a narrow range of growth parameters. Density Functional Theory (DFT) will be performed (in Rome) to understand the Bi incorporation in III-V alloys.
The fine tuning of the GaAsBi/GaAs growth will benefit from a detailed structural and electronic investigation of the first stage of the interface formation to clarify important issues, such as the role of substrate reconstruction, of Bi segregation, and of the strain to increase the Bi content in the III-V alloy toward the concentration of 20-40%, where the TI phase is expected. In order to reduce the strain we will also explore the use of other III-V substrates with higher lattice parameter than GaAs (InAs, GaSb, InP).
The quality of films grown will be checked in situ by surface diffractions (RHEED, LEED) and X-ray Photoemission Spectroscopy (XPS)(all Italian partner), and ex situ by X-Ray Diffraction, and Scanning Probe Microscopy (AFM, STM) and Electron Backscatter Diffraction (EBSD)(all Czech partner).
XPS and Wave Vector Resolved Photo-Emission Electron Microscopy (k-PEEM) measurements will allow for the surface sensitive determination of chemical bonding configurations and stoichiometry of samples as well as the mapping of TI surface state and bulk band dispersion relations up to high binding energies. The important technique k-PEEM (NanoESCA Omicron) is permanently available at the Czech partner institute. This new instrument is one of the few facilities to do fast 3D band mapping [Rev.Sci.Instr. 79, 053702 (2008)] and ensures the necessary rapid feedback of electronic band structure properties for a fast and efficient optimization of the MBE growth in Italy.

Research goals

The main goal of the project is the study of Bi-containing III-V alloys as potential candidates for a new 3D TI. Thus, this research focuses on the progress toward TI properties, or precursors of TI properties, of the electronic band structure of GaAsBi and other Bi-based III-V alloys, with high Bi content.
Besides, the improvement of Bi incorporation and quality of GaAsBi beyond the state of the art is an important result "per se", giving their high interest as new laser source for optical fiber communication. After the optimization of the growth process, we will map the surface states and bulk dispersion and their evolution toward the occurrence of the TI phase. Here, the observation of a band inversion is an important benchmark.
Since Bi and Mn show similar growth problems of low solubility in GaAs, the experience gained by the proponents on GaMnAs diluted magnet semiconductor will be useful.
A main benefit for the project partners will be the combination of sophisticated techniques and unique materials available in collaborating laboratories. Moreover, we intend to submit proposals at large scale facilities such as ESRF, ELETTRA or ASTRID-III if certain material properties have to be studied with higher resolution.
We expect that the work carried out in this proposal will form a solid base for application in EU funding within the Horizon 2020 program.
It should be mentioned the longstanding tradition of cooperation of the two groups, which resulted in 5 joint papers.

Last update: 03/08/2025