Integrated-circuit imaging devices of a proposed type would contain planar arrays of microscopic sensors that would exploit giant magnetoresistance (GMR). Each GMR sensor in such a device would define one pixel in an image with a pixel pitch of the order of tens of microns. With the difference that the GMR sensors would respond to local magnetic fields instead of to locally incident light, they would perform essentially the same role as that played by photodetectors in familiar optoelectronic imaging devices. Indeed, the arrays of GMR sensors could be deposited on readout integrated circuits similar or identical to those on which, heretofore, visible and infrared photodetectors have been deposited. The proposed devices could be used, for example, as conventional magnetometers and gradiometers, magnetic microscopes for examining small ferromagnetic particles or cracks in ferromagnetic materials, imagers for mineralogical exploration, and readers of magnetic inks and magnetic cards.

Figure 1. A Typical GMR Sensor would be constructed in either the current-in-plane (CIP) or current-perpendicular-to-plane (CPP) configuration. The electrical resistance of the device would vary with the applied magnetic field.

A typical GMR sensor according to the proposal (see Figure 1) would comprise several layers of magnetic materials separated by layers of nonmagnetic materials, all supported by a silicon substrate. With the exception of the substrate, the thicknesses of the layers would be of the order of tens of nanometers or less. The layers could be grown on the substrate by vacuum deposition, sputtering, or even electroplating. The growth and other fabrication steps would involve delicate nanotechnological techniques.

The antiferromagnetic FeMn layer would serve to pin the orientation of the magnetization of the Co layer. The Co layer would be regarded as magnetically hard in that the applied magnetic field that one seeks to sense would not be strong enough to change the direction of magnetization of the Co layer. The NiFe layer, designated as the sensing layer, would be regarded as magnetically soft in that its coercivity would be less than that of the pinned (Co) layer and its direction of magnetization could be affected by the applied magnetic field that one seeks to sense.

Figure 2. The Orientation of a Magnetic Field relative to the axis of sensitivity of a GMR sensor would affect the sensor response.

GMR is a large change in electrical resistivity in response to a magnetic field applied along an axis of sensitivity that would lie in a plane parallel to the layer planes and that would be determined by the direction of magnetization in the pinned layer. GMR is the macroscopic electrical manifestation of a phenomenon that is known as spin-valve action and that involves differential scattering, by the magnetic fields of the various layers, of electrons with different spin polarizations. In order to obtain GMR in the proposed device, the separated magnetic layers would have to be antiferromagnetically coupled. Inasmuch as the coupling between two magnetic layers oscillates from ferromagnetic to antiferromagnetic as the thickness of the intervening nonmagnetic layer is varied, it would be necessary to tailor the thicknesses of the nonmagnetic layers precisely in order to obtain the antiferromagnetic coupling needed to realize a highly sensitive GMR sensor.

Figure 2 shows the geometric relationship between an applied magnetic field, B, and the components of the magnetic field parallel and perpendicular to the axis of sensitivity of a GMR sensor, which could be one of many such sensors arrayed in the x-y plane. The sensor would respond only to the component By because the axis of sensitivity as depicted here is oriented parallel to the y axis. In order to sense the component Bx, it would be necessary to rotate the sensor 90° about the z axis to make the axis of sensitivity lie parallel to the x axis. Similarly, in order to sense Bz, it would be necessary to rotate the sensor 90° about the x axis to make the axis of sensitivity lie parallel to the z axis. Alternatively, it should be possible to construct an imaging device containing a two- or three-dimensional array of GMR sensors with orthogonal axes of sensitivity, so that it would be possible to sense Bx and By or even Bx, By, and Bz, simultaneously.

This work was done by Sir B. Rafol of Caltech for NASA's Jet Propulsion Laboratory. NPO-20925

Photonics Tech Briefs Magazine

This article first appeared in the March, 2001 issue of Photonics Tech Briefs Magazine.

Read more articles from the archives here.