Composite magnetoelectric sensors, comprised of thin films of magnetostrictive and piezoelectric materials assembled on fork-like silicon cantilever structures, can detect magnetic field changes as small as 500 fT/Hz^0.5 as reported by researchers from the University of Kiel in Germany in a recent issue of the Journal of Materials Research. This magnetic field detection limit is very close to that required in challenging biomedical applications like magnetocardiography, magnetoencephalography, and deep brain stimulation. 

Magnetoelectric (ME) sensors based on these composite systems can be very sensitive at room temperature and, unlike the currently used superconducting quantum interference device magnetometers, they do not require expensive liquid helium cooling to operate. This makes ME sensors promising low-cost candidates for the detection of the feeble magnetic fields that arise by electrical activity of the heart or the brain.

The mechanical coupling of a magnetostrictive layer to a piezoelectric one is the basic principal behind these heterostructures. Magnetostrictive materials are ferromagnets, the magnetic dipoles of which acquire a parallel alignment in the bulk in a changing magnetic field. Macroscopically, this translates into a change in the length of the material. In a composite sensor, this elongation results in mechanical stress and electrical polarization in the piezoelectric layer, measured as an electrical voltage proportional to the magnetic field.

“A pronounced sensitivity to external excitation is a characteristic property of any sensing material. The challenge is to make use of this sensitivity in the most efficient way without evoking unwanted cross sensitivities to other influences,” says Dirk Meyners, a researcher at the Institute of Materials Science at Kiel University and and one of the principal investigators. “For example, the sensor fabrication must be tuned in such a way that mechanical stress does not interfere with the magnetic properties of the magnetostrictor,” he says.

For the fabrication of piezoelectric thin films the researchers used either PZT (lead zirconate titanate) or AlN (aluminium nitride). PZT, a widely used piezoelectric ceramic material, has a low dielectric loss factor, while AlN is characterized by a large piezoelectric coefficient. The magnetostrictive layer was an amorphous alloy of iron and cobalt—FeCoSiB or (Fe₉₀Co₁₀)₇₈Si₁₂B₁₀—with soft magnetic properties and significant magnetostriction.

Thin films of the piezoelectric and magnetostrictive materials were deposited by various deposition processes (low-temperature pulsed-DC sputter deposition for AlN, chemical solution deposition for PZT) on Si cantilever structures, with thicknesses of a few micrometers.

The findings show that one of the measured limits of detection at 180 pT/Hz^0.5 at 10 Hz, for ME sensors with exchange biased magnetostrictive multilayers, is only a small factor above the upper magnetocardiography signal limit. For resonant ME cantilever sensors with a single FeCoSiB layer, the limit of detection is even lower, at 1 pT/Hz^0.5, reaching the upper frequency limit of magnetoencephalography signals.

Recording of brain signals is even more challenging, with amplitudes between 10 fT and 10 pT at maximum. But the best limit of detection achieved by the team so far at 500 fT/Hz^0.5 at 958 Hz, shows that measurement of current density distributions in the human brain induced by deep brain stimulation are potentially within reach.

According to Nian Sun, a professor at Northeastern University, “This ultra-low [detection] limit achieved by Meyners’ team means that they could be used in a range of applications, from biomagnetic sensing to magnetic anomaly detection, navigation, etc.” Sun sees a great potential in these microelectromechanical systems (MEMS) magnetoelectric sensors. “I am hopeful that devices which make use of these ME sensors, like nanoelectromechanical systems (NEMS) ME sensors, ME antennas, voltage tunable ME devices etc., will lead to a bright future for these materials and devices,” he says. 

According to Meyners, the team is currently investigating more complex resonators to improve the sensor’s sensitivity and bandwidth and to even further lower the detection limit.

Read the abstract in the Journal of Materials Research.