A physical effect called superinjection lies beneath advanced light-emitting diodes (LEDs) and lasers. For several years, this effect was thought to happen only in semiconductor heterostructures — that is, structures made up of two or more semiconductor materials. Scientists from the Moscow Institute of Physics and Technology have found superinjection to be possible in homostructures, which are composed of a single material. This facilitates completely new opportunities for the progress of light sources. The paper was published on February 21st in the journal Semiconductor Science and Technology.
Semiconductor light sources, such as LEDs and lasers, are at the center of modern technology. They facilitate laser printers and high-speed internet. But just 60 years ago, no one would have predicted semiconductors would be used as materials for bright light sources. The problem was that to produce light, such devices need electrons and holes — the free charge carriers in any semiconductor — to rejoin. The greater the concentration of electrons and holes, the more repeatedly they recombine, rendering the light source brighter. However, for a long time, no semiconductor device could be made to deliver an adequately high concentration of both electrons and holes.
In the 1960s, the solution was discovered by Zhores Alferov and Herbert Kroemer. They suggested the use of heterostructures, or “sandwich” structures, comprising of two or more complementary semiconductors rather than just one. If one puts a semiconductor between two semiconductors with broader bandgaps and applies a forward bias voltage, the concentration of electrons and holes in the middle layer can realize values that are orders of magnitude greater than those in the outer layers. This effect, called superinjection, lies beneath modern semiconductor lasers and LEDs. Alferov and Kroemer’s discovery got them the Nobel Prize in physics in 2000.
However, two arbitrary semiconductors cannot make a feasible heterostructure. The semiconductors need to possess the same period of the crystal lattice. Otherwise, the number of flaws at the interface between the two materials will be very high, and no light will be produced. In a way, this would be quite like trying to screw a nut on a bolt whose thread pitch does not fit that of the nut. Since homostructures are made up of only one material, one portion of the device is a natural extension of the other. Although homostructures are easier to fabricate, it was supposed that homostructures could not handle superinjection and thus are not a feasible basis for practical light sources.
Igor Khramtsov and Dmitry Fedyanin from the Moscow Institute of Physics and Technology made a discovery that radically alters the viewpoint on how light-emitting devices can be engineered. The physicists learned that it is possible to realize superinjection with merely one material. Furthermore, most of the existing semiconductors can be used.
“In the case of silicon and germanium, superinjection requires cryogenic temperatures, and this casts doubt on the utility of the effect. But in diamond or gallium nitride, strong superinjection can occur even at room temperature,” Dr. Fedyanin said. This means that the effect can be used to build mass market devices. According to the new paper, superinjection can create electron concentrations in a diamond diode that are 10,000 times higher than those formerly believed to be eventually possible. Consequently, diamond can act as the foundation for ultraviolet LEDs thousands of times brighter than what the most optimistic theoretical calculations projected.
Surprisingly, the effect of superinjection in diamond is 50 to 100 times stronger than that used in most mass market semiconductor LEDs and lasers based on heterostructures.
Igor Khramtsov, PhD Student, Moscow Institute of Physics and Technology.
The physicists stressed that superinjection should be viable in a wide variety of semiconductors, from conventional wide-bandgap semiconductors to novel 2D materials. This paves the way for new opportunities for designing extremely efficient ultraviolet, blue, violet, and white LEDs, as well as light sources for optical wireless communication (Li-Fi), transmitters for the quantum internet, new types of lasers, and optical devices for early disease diagnostics.
The research was aided by the Russian Science Foundation (17-79-20421).