Electrons on the surface of a topological insulator can flow with little resistance. Their spin and direction are intimately related; the direction of the electron determines its spin and in turn is determined by it. Credit: LBNL.
Last week several press releases came out about new work on topological insulators. A topological insulator—more accurately, a “strong 3D topological insulator”—are semiconductor materials that are bulk insulators and surface conductors. A press release from Lawrence Berkeley National Laboratory compares it to copper-coated bowling balls.
Experimental work in the field of TIs is new, only about five years old, but it has captured the attention and imaginations of condensed physicists and materials scientists because of possible applications as spintronics devices that exploit the spin properties of materials and, eventually, for quantum computers.
The surface of TI materials behaves differently than a conductive metal like a copper coating would, though. There are two mechanisms by which the electrical conductivity of a material is limited: at low temperatures (in the 20 K range), electrons scatter off crystal defects; at high temperatures (room temperature range), electrons scatter off phonons, or the crystal’s lattice vibrations.
TI materials are “topologically protected” from losing conductivity via scattering from defects because of the spin of the electrons, an intrinsic property like mass or charge. A press release from Brookhaven National Laboratory explains,
“On the surface of a topological insulator, the electrons moving in one direction have the opposite spin from the electrons moving in the opposite direction. If they hit a defect, they cannot just bounce back, as that would require also flipping the spin to match the spin of the other electrons flowing in the opposite direction. Flipping the spin would require a change in the magnetic moment at the barrier; without a magnetic change, there’s no flip of spin, and a U-turn is forbidden.”
In short, scattering off defects does not happen because the rules of quantum mechanics do not allow it.
But, what about scattering at higher temperatures? Any realistic device applications would require that the material perform at warmer temperatures, like room temperature, which is plenty high to excite lattice vibrations. BNL researcher, Tonica (Tony) Valla said in the press release, “If the bulk materials behaved as its name implies it should—that is, as an insulator—we’d be able to make very efficient room-temperature electronic devices with these materials.” Valla is coauthor of a paper describing this work published in May in Physical Review Letters.
The collaborative team of researchers from BNL, LBL and MIT used an instrument at LBL to investigate the high-temperature scattering properties of Bi2Se3, a well-known TI material. The trick is to separate the electron transport on the surface from the total conductivity, including the bulk. Angle-resolved photoemission spectrometry is able to do this by shining bright light on the sample and capturing the electrons that the light photons knock loose. By recording the angle and energy of the photoemitted electrons, the instrument slowly builds up a graphic representation of the sample’s electronic structure.
Cone-shaped electronic structure of Bi2Se3 at two temperatures. The two points that correspond to E=0 correspond to electrons at the Fermi surface moving in opposite directions. Credit: BNL.
The LBL press release explains the physics by comparing Bi2Se3 to graphene, which has a similar band structure, but is not a TI. The fundamental difference is that the band structure diagrams generated by ARPES look like slices through cones with an X centered on the Dirac point. In graphene, the Fermi surface lies at the Dirac point. However, the Fermi surface of Bi2Se3 was found to lie high above the conical conduction band and maps out a perfect circle. From the press release, “It’s as if the circular Fermi surface were drawn right on the surface of the topological insulator, showing how spin-locked surface electrons must change their spin orientation as they follow this continually curving path.”
The ARPES work on Bi2Se3 shows that the coupling between electrons and phonons is very weak and that, effectively, surface electrons are also “protected” from scattering off lattice vibrations.
Alexis Fedorov, LBL’s ARPES staff scientist and paper coauthor said, “Although there’s still a long way to go, the experimental confirmation that electron-phonon coupling is very small underlines Bi2Se3‘s practical potential.”
Duke University professor, Stefano Curtarolo, is also working on TIs, but is starting with the electronic band structures of constituent elements and developing a “genetic” approach to finding promising compounds.
Curtarolo established the Center for Materials Genomics at Duke, which includes a “materials genome repository,” a property database encompassing structure, electronic and thermoelectronic properties and a scintillator database.
However, Curtarolo says in a press release, a database is not enough. “While extremely helpful and important, a database is intrinsically a sterile repository of information, without a soul and without life. We need to find the materials’ ‘genes.'” To do that, his group has developed a “topological descriptor,” that can be applied to the database to “provide the directions for producing crystals with desired properties.”
A paper published in Nature Materials describes how the descriptor can be used to evaluate and classify combinations of elements. At one end of the spectrum, are compounds Curtarolo classifies as “fragile.” He says, “We can rule those combinations out because, what good is a new type of crystal if it would be too difficult to grow, or if grown, would not likely survive?” A middle group of compounds is classified as “feasible,” with the obvious potential. The final, and most exciting group, are those Curtarolo classifies as “robust.” These compounds are stable and easily produced. Also, the crystals can be grown in different directions, which means electrical properties can be tailored during the crystal growth process.
His paper reports finding 28 TI materials (some known, some new), including some very unusual compounds that are unlikely to have been anticipated, such as Cs{Sn,Pb,Ge}{Cl,Br,I}3.
The BNL/LBL/MIT paper is “Measurement of an Exceptionally Weak Electron-Phonon Coupling on the Surface of the Topological Insulator Bi2Se3 Using Angle-Resolved Photoemission Spectroscopy,” Z.-H. Pan, et. al., Phys. Rev. Letters, doi: 10.1103/PhysRevLett.108.187001.
The Duke paper is “A Search Model for Topological Insulators with High-Throughput Robustness Descriptors,” Kesong Yang, et. al., Nature Materials, doi: 10.1038/NMAT3332.