Professor, Research Institute of Electrical Communication, Tohoku University
Crossing of magnetism and semiconductor physics
— Ferromagnetic semiconductors expand the possibilities of spintronics —
The charge of electron enables semiconductors to process information, and its spin allows us to realize magnetic information storage devices. Even though these properties are normally utilized separately using magnetic and semiconductor materials, respectively, spintronics uses both properties in the same material. There are two approaches for designing advanced spintronic devices. One is to use both degrees of freedom in metal-based magnetic materials, and the other is in semiconductors. Prof. Ohno has expanded the possibilities of utilizing semiconductors with various structures and functions by making III-V semiconductors ferromagnetic.
The ferromagnetic semiconductor that Prof. Ohno developed was (In, Mn)As in which part of the indium atoms in InAs is replaced with manganese, which is a magnetic element. He said, “When I went to the IBM T. J. Watson Laboratory in 1988, joining Leo Esaki's group, I thought of conducting two kinds of research if I was given the chance. One was research which seemed too difficult to accomplish, and the other was research through which I could publish papers.” It was widely known then that if the Mn concentration was too high, a MnAs phase would nucleate before making InAs magnetic. However, he and his coworkers were able to synthesize single-phase (In, Mn)As by decreasing the crystal growth temperature to 250°C.
Several years later, he discovered that single-phase (In, Mn)As was ferromagnetic at low temperatures. Although he had an idea of causing semiconductors to have magnetism, he never expected that semiconductors would be ferromagnetic. A ferromagnetic material transforms to a paramagnetic material when it exceeds a transition temperature. He produced a field effect transistor using (In, Mn)As and was able to change the transition temperature of (In, Mn)As by applying an electric field. He showed that the transition temperature of (In, Mn)As increases as its carrier concentration increases in an applied electric field . This occurs because the Mn spins align in the same direction due to their interaction with charge carriers in the semiconductor. He says, “A semiconductor has the capability of changing the number of electrons. Although it has been about 26 to 27 centuries since human beings found magnetic materials, this was the first time that we prepared magnetic materials of which the properties could be altered after fabrication was complete.” However, unless the transition temperature drastically increases, these materials cannot be commercialized. As well as increasing the Mn concentration, he has been searching for other alloying elements with a stronger interaction between charge carriers and Mn spins because the interaction depends on the type of semiconductors.
The next generation magnetic device, MRAM (Magnetic Random Access Memory), where its memory bits are highly integrated, will require a high magnetic field or a large electric current to record data. The ferromagnetic semiconductor, (In, Mn)As, meets this requirement when it is used in a field effect transistor. When an electric field of about 1.5MV/cm is applied to (In, Mn)As, its coercive force decreases, and the magnetic field required for magnetization reversal falls to one fifth of that currently used. When data is recorded by a low magnetic field while applying an external electric field, and then the field is turned off, the coercive force is recovered, which means that a higher magnetic field will be required to rewrite data.
Prof. Ohno thinks the key to a scientific breakthrough is curiosity. “However, it takes more than curiosity. You have to decide what you are going to do and do it quickly. You need to design things in such a way that you can go through a number of trials without spending too much time. Then, quantity turns to quality at a certain point,” said Prof. Ohno. His research is fundamental, but he is always thinking about its applications. He says, “You can publish your paper no matter what your research is. There are lots of research areas, which others have not studied yet. However, time and resources are limited. The first thing you have to do is to choose what you will do or will not do, and then figure out how to deepen your research. After all, what you choose to do first will show your taste.
