The edge currents of a topological insulator serve as a source of spin-polarized electrons. Graphics: Luis Maier

Electrons have an intrinsic angular momentum, called spin. As a consequence, not only do they carry charge, but they also behave like tiny magnets, which can be aligned. In our everyday use of computers, however, so many electron magnets point randomly in all directions as to cancel out as a whole. But if the spin were to be controlled, conventional computers might suddenly become a lot faster: In the field of so-called spintronics, the magnetic orientation of the electrons is used for information transfer, which generates much less heat than is produced by continually switching the current on and off as is required in conventional electronics.

Metal and insulator at the same time: Topological insulators

Topological insulators represent a very promising class of materials for the implementation of spintronic devices. They conduct electricity only on their surface, but not in their interior. In the thin layers of some of these materials, the edge current consists of exactly two channels in which the individual electrons flow. The flow direction in the two channels is opposite to each other as is the spin orientation. This behavior is called the quantum spin Hall effect due to its analogy to the quantum Hall effect. The QSH effect was discovered in 2007 by the research group of Professor Laurens Molenkamp at the University of Würzburg.

Physicists at the department of Laurens Molenkamp and the research group of Professor Ewelina Hankiewicz now demonstrate—together with researchers of Stanford University in California—how the spin polarization of the channels can be experimentally verified. They also present an electronic device that can generate and measure spin-polarized currents and thus possesses some basic qualities required for spintronics. The results are published in the prestigious journal Nature Physics ("Spin polarization of the quantum spin Hall edge states").

From theory to experiment: Successful with an H-shaped nanostructure

Until recently, the spin-polarization of the channels was just mathematically described; experimentally, it could only be indirectly inferred. "However, the quantum spin Hall effect requires an actual spin-polarized transport as a condition for its existence," says research group leader Hartmut Buhmann of Molenkamp's department.

Würzburg physicist Christoph Brüne managed to furnish the desired experimental proof with an ingenious experimental set-up. Critical to the success was an H-shaped nanostructure, consisting of mercury telluride and fitted with an additional gold electrode at each leg.

With this configuration, it is possible to induce a quantum spin Hall state in one leg of the H-structure by means of an applied gate voltage. The other leg causes an imbalance between the two spin currents at the connection point, the cross bar of the H. As a consequence, only electrons with magnetic alignment can be extracted and measured. This also works in the reverse direction so that you can inject a spin-polarized current and measure the induced voltage in the QSH material.


Electron microscopic image of the circuit: The semiconductor H is shown in red, the gate contacts in yellow. The picture shows a section of about three by three micrometers. Photo: Luis Maier

The theory required for the clear identification of the measured values as spin-currents, including some sophisticated simulations, comes from the group of Ewelina Hankiewicz and her colleagues in the research group of Professor Shou-Cheng Zhang in Stanford: "It wasn't easy to calculate how the spin edge currents get into the metal of the second leg," Professor Hankiewicz says.

However, all the hard work paid off in the end. The editors of Nature Physics even dedicated a "News & Views" review article to the Würzburg research ("Quantum spin Hall effect: Left up right down"). "This is equivalent to a high distinction, classifying our results as particularly important," explains Laurens Molenkamp.

Next research steps: Development of the concept

So far, the configuration presented by the Würzburg physicists only works at extremely low temperatures of typically -271 C. To make it work at room temperature, the scientists still need to find suitable materials. In the future, the Würzburg researchers intend as a first step to develop the concept into a spin transistor, thus providing all the basic elements required for application in spintronics.

In addition, topological insulators have even more potential: They are a safe bet for further exotic discoveries, such as Majorana fermions, i.e. particles that are their own anti-particles. So it doesn't come as a surprise that the German Research Foundation (DFG) intends to establish a new priority program for "topological insulators" this year.

Source: Julius-Maximilians-Universität Würzburg