In addition to charge, the electron has spin, which gives it a magnetic dipole moment. Until recently, applied physicists had overlooked the potential for using electron spin to store and process information. The rules of quantum mechanics mean that any attempt to observe the state of electron spin will result in either "up" or "down," which leads to the concept that electron spin could represent 0 (down) and 1 (up) in binary computing. Spintronics (the word dates from only 1996) exploits the spin, rather than the electrical charge, of the electron.

Giant magnetoresistance (GMR), discovered in 1988, is the result of subtle electron-spin effects in layered materials. In 1997 IBM launched read heads utilizing GMR technology, allowing the storage capacity of a hard drive to increase 20 times. This was the first industrial exploitation of electron spin. Currently all spin-based devices are memory devices involving metals.

Spin devices based on semiconductors offer the possibility of further miniaturization and increases in processor speed. It is no longer science fiction to imagine a fast compact computer with the following features: silent (no fans, no spinning hard drive), runs for days on a flashlight battery, no power outages, and instant on/off (no boot time). But to achieve this, physicists first must discover how to control spin currents in silicon.

For practical applications in semiconductors, physicists must find a means of generating a spin current, the flow of which then transfers information. One suggestion is to exploit spin-orbit coupling, the effect of which is to make the path of an electron in an electric field spin-dependent. But all methods for producing, manipulating, and detecting spin currents have eluded researchers. This lack of progress changes with Hot Paper #7, in which Shuichi Murakami and Naoto Nagaosa, both of Tokyo University, and Shou-Cheng Zhang, of Stanford University, report a major theoretical breakthrough. They suggest that the quest for spin current should concentrate on semiconductor holes rather than electrons.

The starting point of the new investigation is the two-dimensional quantum Hall effect (QHE), a type of superconductivity that occurs in semiconductors. Murakami and Zhang have a long-term collaboration to understand QHE, with the aim of generalizing the theory to higher dimensions. In the course of this work, the researchers made the serendipitous discovery of a spin current that is associated with holes in a semiconductor. Paper #7 is a theoretical demonstration that spin currents produced through the spin-orbit coupling route should arise more readily with holes than with electrons because of the strong spin-orbit coupling that naturally exists for holes, but not electrons, in many semiconductor fabrications.

The spin current described in #7 is completely reversible, and it is this property that is the driver behind an accelerating citation rate. Reversibility minimizes power dissipation, and requires neither liquid-helium temperatures nor a magnetic field. Spin currents therefore have a property in common with superconductors, but with the huge advantage of room-temperature operation. The hole-based spin current is also useful for injection because it is induced by an electric field that effectively creates the spin current inside the semiconductor and thereby avoids any tricky problems with an interface.

Spintronics theory has progressed rapidly since the publication of #7. Technological applications are on hold until experimentalists can confirm the prediction. And there will doubtless be further problems along the road to practical applications.