Fraunhofer ENAS shrinks 2D compasses
Spin valves (SV) consist of two ferromagnetic layers separated by a non-magnetic conducting spacer – see figure 1 and their field sensing principle relies on a so-called giant magneto-resistance effect (GMR).
Fig. 1: Spin-valve multilayer stack with a coupling to an antiferromagnet.
The magnetization of one of the ferromagnets is fixed and serves as a reference layer via exchange coupling with an anti-ferromagnetic layer, while the other ferromagnet is allowed to respond to an external magnetic field (free layer).
A giant magneto-resistance effect (GMR) originates from the asymmetry in the spin-dependent scattering of electrons at the magnetic/non-magnetic interfaces for spin-up and spin-down electrons. This effect leads to a state of maximum resistance for an antiparallel alignment of the two ferromagnetic layer magnetizations, and to a state of minimum resistance in the case of parallel alignment.
For a maximum signal-to-noise ratio, the researchers have crafted, layer by layer a double full-bridge layout with an antiparallel alignment of the pinned layer magnetization for neighbouring meander structures – see figure 2.
Even producing the GMR sensors at wafer level, they were able to precisely tune the magnetic alignment of the ferromagnetic layers of these multi-layered meander resistors through localized laser heating and in-field cooling (at a microscopic resolution).
Fig. 2. Optical microscopy image of the double full-bridge sensor showing the alignment of the exchange bias defined by laser heating and in-field cooling for each individual meander (yellow arrows).
The full design only measures 0.8×0.5mm, and because the sensing material layers are only a few nanometres thick and the laser heating so shallow, the 2D monolithic compass is fully CMOS-compatible, that is, the sub-micrometre thick sensing meanders could be manufactured as a post-processing step on top of another functional IC, claims Ueberschär. In effect, you could probably add magnetic sensing on top of the chip processing the signal for gyroscope and accelerometer MEMS.
The design enables full 360° sinusoidal signal resolution in the geomagnetic field, with an output signal of around 250μV for 50μT and a very high spatial and temporal resolution (1mm and 1ms respectively).
“This 2D monolithic compass integration could not only be very cost-effective, low power and better performing for navigation or industrial sensing applications”, told us Dr. Olaf Ueberschär during a demo at SEMICON Europa, “but it could yield other interesting applications such as the magnetic field camera for which we have a patent pending”, he added.
“You could build a large array of 2D spin-valve sensors to operate as pixels in a magnetic field camera, to check the homogeneity of magnets, or to determine the magnetic field of a machine part and highlight invisible defects”, Ueberschär continued. What’s more, this non-destructive observation technique is completely biocompatible for use in medical applications.
Ueberschär and his team are hard-working on developing such a camera and hope to have a first prototype ready in the spring of 2015.
Visit Fraunhofer ENAS at www.enas.fraunhofer.de/en.html