A miniscule magnetic sensor can map out the currents coursing through a microchip, physicists report. The gizmo might someday inspect chips as they roll off the production line.
Tiny magnetic fields emanate from the electric currents in a microchip, and, in principle, those fields could be used to map the precise flow of electricity through the chip's circuits. Researchers have struggled, however, to detect such fields. The most sensitive magnetic detectors must be immersed in liquid nitrogen or some other chilly liquid, so they cannot get close enough to the chip to resolve the magnetic variations just above the chip's surface. On the other hand, a nanometer-sized magnetized finger called a magnetic force microscope gets close enough to the chip, but is less sensitive and can be tricky to control.
A new sensor can detect the fields, report Gang Xiao and Ben Schrag of Brown University in Providence, Rhode Island. Like the read head in a computer's hard drive, the sensor consists of a tiny high-tech layer cake known as a spin valve. One layer remains magnetized in a fixed direction, while the magnetization of the other layer can flip back and forth in response to a magnetic field. When the two layers are magnetized in the same direction, current flows more easily through the junction than when the layers point in opposite directions. Xiao and Schrag passed the device across a simple chip's surface and monitored the sensor's electrical resistance, thus mapping out the magnetic fields and deducing the currents, they report in the 12 May issue of Applied Physics Letters. In particular, the researchers tracked the changes in current flow as an overloaded wire burned out.
The new sensor could help chipmakers ferret out costly design flaws and control chip quality, says C. L. Chien, a physicist at Johns Hopkins University in Baltimore. "They've taken advantage of all the technology for disk drives and have made a microscope out of it," Chien says. "It's a very clever idea." In the same issue of Applied Physics Letters, physicist Ellen Williams and colleagues at the University of Maryland, College Park, reported similar results using a magnetic force microscope. But, Williams says, Xiao and Schrag's approach "may turn out to be a more robust technique."