Microwave photocurrent from the edge states of InAs/GaInSb bilayers

Rapid Communication

Microwave photocurrent from the edge states of InAs/GaInSb bilayers

Jie Zhang, Tingxin Li, Rui-Rui Du, and Gerard Sullivan

Phys. Rev. B 98, 241301(R) – Published 10 December 2018

Abstract

We measure microwave photocurrent in devices made from InAs/GaInSb bilayers where both the insulating bulk state and conducting edge state were observed in the inverted-band regime, consistent with the theoretical prediction for a quantum spin Hall (QSH) insulator. It has been theoretically proposed that microwave photocurrent could be a unique probe in studying the properties of QSH edge states. To distinguish a possible photoresponse between a bulk state and helical edge state, we prepare a Hall bar and Corbino disk from the same wafer. Results show that the Corbino-disk samples have a negligible photocurrent in the bulk gap, while clear photocurrent signals from the Hall-bar samples are observed. This finding suggests that the photocurrent may carry information concerning the electronic properties of the edge states.

Received 13 June 2018
Revised 18 November 2018

DOI:https://doi.org/10.1103/PhysRevB.98.241301

Physics Subject Headings (PhySH)

Edge states Optical & microwave phenomena Quantum spin Hall effect Topological insulators

Bilayer films III-V semiconductors Quantum wells

Hall bar

Condensed Matter, Materials & Applied Physics

Authors & Affiliations

Jie Zhang^1,*, Tingxin Li¹, Rui-Rui Du^1,2,†, and Gerard Sullivan³

¹Department of Physics and Astronomy, Rice University, Houston, Texas 77251, USA
²International Center for Quantum Material, Peking University, Beijing 100871, China
³Teledyne Scientific and Imaging, Thousand Oaks, California 91630, USA

^*jz38@rice.edu
^†rrd@rice.edu

Article Text (Subscription Required)

Click to Expand

References (Subscription Required)

Click to Expand

Issue

Vol. 98, Iss. 24 — 15 December 2018

Reuse & Permissions

Access Options

Article Available via CHORUS

Download Accepted Manuscript

Author publication services for translation and copyediting assistance advertisement

Images

Figure 1
(a) Structure of strained-layer InAs/GaSb wafer used in the experiment, which hosts robust quantum spin Hall edge states. (b) Resistance measured for Hall-bar sample and conductance measured for the Corbino sample. (c) Experimental setup of the differential photovoltage $V_{p h}^{f}$ and photocurrent $I_{p h}^{f}$ measurement on a typical Corbino disk with MW amplitude modulation provided by locked-in amplifier frequency $f$ . (d) Photoresponse measurement setup on a Hall-bar sample fabricated the same way as the Corbino sample shown in (c).
Reuse & Permissions
Figure 2
(a) Magnetoconductance measurement of the insulating bulk state on a Corbino disk with a strong external parallel magnetic field (up to 8 T). 2D bulk is accessed, sweeping the backgate voltage, which indicates a completely insulating behavior, giving no evidence of gap closing at increasing magnetic field. (b) Photoconductance measured on the same Corbino sample excited by a linearly polarized MW radiation. The MW amplitude is modulated by a lock-in amplifier. The miniscule photoconductance in the bulk gap demonstrates that the 2D bulk states make no contribution to the photocurrent. (c) Magnetoresistance measurements on a Hall-bar device. The plateau increases with parallel magnetic fields, indicating an enhancement of backscattering under broken TRS. (d) Resistance of the Hall-bar device in (c) measured by fixing the backgate at the plateau and sweeping a parallel magnetic field. A small bias current is used to avoid the heating effect. Resistance doubled at 1 T and showing saturation above 1 T.
Reuse & Permissions
Figure 3
(a) The MW power-dependent photocurrent traces measured in a Hall-bar device with no magnetic field application and a large external bias current. Enhanced photocurrent on both sides of the gap boundary is detected with increasing MW radiation. Miniscule photocurrent at the CNP is maintained up to 10 mW. This phenomenon starts to break down at 100 mW MW power due to heating. The conventional unit dbm is defined as $P (dbm) = 10 lo g_{10} \frac{P (W)}{1 mW}$ , where $P (dbm) / P (W)$ stands for power in the unit of dbm/Watt. (b) MW frequency-dependent photocurrent measurement is shown by tuning the MW power to keep temperature at a fixed value of 650 mK. This is chosen instead of using a fixed MW input power in order to rule out the possibility of different attenuations at various frequency ranges. Shifting of the CNP towards a lower backgate voltage is observed with increasing MW frequency.
Reuse & Permissions
Figure 4
(a) Photocurrent and (b) photovoltage measurement in a Hall-bar device with fixed MW frequency and power. The backgate voltage is swept at different perpendicular magnetic fields. A moderate magnetic field is employed to break TRS, thus causing Rashba-type spin flipping between the two branches of the helical edge states. This results in an enhancement of the photocurrent and photovoltage in the bulk gap. Comparison between the effect of parallel and perpendicular magnetic fields is displayed in the inset of (a), indicating a larger $g$ factor for perpendicular fields.
Reuse & Permissions

Physical Review B

covering condensed matter and materials physics