One material science area holding promise for spintronic applications is dilute magnetic semiconductors. These are regular semiconductors to which a small amount of magnetic atoms is added to make them ferromagnetic.
Spintronics is the processing of data using electron “spin” instead of charge; it could revolutionize the computing industry with more energy efficient, faster and smaller data storage and processing. But understanding the source of ferromagnetism in dilute magnetic semiconductors has been a big problem stopping their further development and use in spintronics.
An important step to removing this road-block has been taken by scientists at Lawrence Berkeley National Laboratory (Berkeley Lab), using a new technique called HARPES, an acronym for “Hard x-ray Angle-Resolved PhotoEmission Spectroscopy”.
Gallium Manganese Arsenide
HARPES data on GaMnAs indicate that the ferromagnetism of dilute magnetic semiconductors arise from two distinct mechanisms.
Credit: A Gray et al
In the HARPES technique, a beam of hard x-rays flashed on a sample causes photoelectrons from within the bulk to be emitted. Measuring the kinetic energy of these photoelectrons and the angles at which they are ejected reveals much about the sample’s electronic structure.
A team of researchers investigated the bulk electronic structure of the prototypical dilute magnetic semiconductor gallium manganese arsenide. Their findings show that the material’s ferromagnetism arises from both of the two different mechanisms that have been proposed to explain it.
“This study represents the first application of HARPES to a forefront problem in materials science, uncovering the origin of the ferromagnetism in the so-called dilute magnetic semiconductors. Our results also suggest that the HARPES technique should be broadly applicable to many new classes of materials in the future,”
says Charles Fadley, a physicist who led development of HARPES.
Nanoscale Spintronics
In semiconductors used in the current generation of computers, tablets and smartphones, once a device is fabricated it is the electronic structures below the surface, in the bulk of the material or in buried layers that determine its effectiveness.
HARPES, based on the photoelectric effect described in 1905 by Albert Einstein, enables scientists to study bulk electronic effects with minimum interference from surface reactions or contamination.
It furthermore lets them probe buried layers and interfaces that are omnipresent in nanoscale devices, and key to smaller logic elements in electronics, novel memory architectures in spintronics, and more efficient energy conversion in photovoltaic cells.
“The key to probing the bulk electronic structure is using hard x-rays, which are x-rays with sufficiently high photon energies to eject photoelectrons from deep beneath the surface of a solid material. High-energy photons impart high kinetic energies to the ejected photoelectrons, enabling them to travel longer distances within the solid. The result is that more of the signal originating from the bulk will be detected by the analyzer,”
says Alexander Gray, who formerly worked with Fadley’s group to develop the HARPES technique and is now with Stanford University and the SLAC National Accelerator Laboratory.
Dilute Magnetic Semiconductor
The study used HARPES spectroscopy to look into the electronic bulk structure of gallium manganese arsenide (GaMnAs). Gallium arsenide, as a semiconductor, is second only to silicon in prevalent use and importance.
If only a few percent of the gallium atoms in this semiconductor are replaced with atoms of manganese, the result is a dilute magnetic semiconductor, a material which would be particularly well suited for additional development into spintronic devices if mechanisms underlying their ferromagnetism were better understood.
“Right now the temperature at which gallium manganese arsenide operates as a dilute magnetic semiconductor is 170 Kelvin. Understanding the actual mechanism by which the magnetic moments of individual manganese atoms are coupled so as to become ferromagnetic is critical to being able to design future materials that would operate at room temperature,”
said Fadley.
Two Prevailing Models
The two existing theories addressing the origin of ferromagnetism in dilute magnetic semiconductors are the “p-d exchange model” and the “double exchange model.”
In the p-d exchange model, ferromagnetism is mediated by electrons in the valence bands of gallium arsenide whose influence extends through the material to other manganese atoms.
According to the double exchange model, the magnetism-mediating electrons lie in a separate impurity band created by doping the gallium arsenide with manganese. In effect, electrons jump back and forth between two manganese atoms so as to lower their energy when their ferromagnetic magnets are parallel.
“Our bulk-sensitive HARPES measurements revealed that the manganese-induced impurity band is located mostly between the gallium arsenide valence-band maximum and the Fermi level, but the manganese states are also merged with the gallium arsenide valence bands. This is evidence that the two mechanisms co-exist and both act to give rise to ferromagnetism,”
said Gray.
“We now have a better fundamental understanding of electronic interactions in dilute magnetic semiconductors that can suggest future materials with different parent semiconductors and different magnetic dopants. HARPES should provide an important tool for characterizing these future materials,”
Fadley adds.
Original study: Bulk electronic structure of the dilute magnetic semiconductor GaMnAs through hard X-ray angle-resolved photoemission
