A team of researchers from the U.S. Department of Energy’s Lawrence Berkeley National Laboratory (Berkeley Lab) and the University of California at Berkeley has just widened the vista of possibilities with an unexpected discovery about TIs: when hit with a laser beam, the spin polarization of the electrons they emit (in a process called photoemission) can be completely controlled in three dimensions, simply by tuning the polarization of the incident light.
In energy-momentum diagrams of graphene and TIs, the conduction bands (where energetic electrons move freely) and valence bands (where lower-energy electrons are confined to atoms) don’t overlap as they do in metals, nor is there an energy gap between the bands, as in insulators and semiconductors. Instead the “bands” appear as cones that meet at a point, called the Dirac point, across which energy varies continuously.
The experimental technique that directly maps these states is ARPES, angle-resolved photoemission spectroscopy. When energetic photons from a synchrotron light source or laser strike a material, it emits electrons whose own energy and momentum are determined by the material’s distribution of electronic states. Steered by the spectrometer onto a detector, these photoelectrons provide a picture of the momentum-space diagram of the material’s electronic structure.
Similar as their Dirac-cone diagrams may appear, the electronic states on the surface of TIs and in graphene are fundamentally different: those in graphene are not spin polarized, while those of TIs are completely spin polarized, and in a peculiar way.
A slice through the Dirac-cone diagram produces a circular contour. In TIs, spin orientation changes continuously around the circle, from up to down and back again, and the locked-in spin of surface electrons is determined by where they lie on the circle. Scientists call this relation of momentum and spin the “helical spin texture” of a TI’s surface electrons. (Electron spin isn’t like that of a spinning top, however; it’s a quantum number representing an intrinsic amount of angular momentum.)
Directly measuring the electrons’ spin as well as their energy and momentum requires an addition to ARPES instrumentation. Spin polarization is hard to detect and in the past has been established by firing high-energy electrons at gold foil and counting which way a few of them bounce; collecting the data takes a long time.
The new instrument was first used at the ALS to study the well-known topological insulator bismuth selenide. While the results confirmed that bismuth selenide’s helical spin texture persists even at room temperature, they raised a perplexing question.
In the first experiments, the incident light was p-polarized, which means the electric part of the light wave was parallel to a plane that was perpendicular to the TI surface and oriented according to the path of the emitted photoelectrons. Since studies of topological insulators typically use p-polarized light in this geometry, sure enough, the spin-ARPES measurements showed the photoelectrons were indeed spin polarized in directions consistent with the expected spin texture of the surface electrons.
Repeated careful experiments with a range of laser polarizations showed, however, that the spin polarization of the photons in the laser beam controlled the polarization of the emitted photoelectrons. When the laser polarization was smoothly varied – and even when it was circularly polarized right or left – the photoelectron spin polarization followed suit.
Why had no results counter to the expected surface textures been reported before? Probably because the most common kind of spin-ARPES experiment makes a few measurements in a typical geometry using p-polarized light. With other arrangements, however, photoelectron spin polarization departs markedly from expectations.
The ability to hit a topological insulator with a tuned laser and excite polarization-tailored electrons has great potential for the field of spintronics – electronics that exploit spin as well as charge. Devices that optically control electron distribution and flow would constitute a significant advance.
Optical control of TI photoemission has more immediate practical possibilities as well. Bismuth selenide could provide just the right kind of photocathode source for experimental techniques that require electron beams whose spin polarization can be exquisitely and conveniently controlled.
This story is reprinted from material from Berkeley Lab, with editorial changes made by Materials Today. The views expressed in this article do not necessarily represent those of Elsevier. Link to original source.