It is well known that low-frequency (0.01–10 Hz), 10–100 nanotesla (nT) magnetic fluctuations occur hours or days before an earthquake. This article reports the development of ultrasensitive, low-cost, room-temperature, magnetic sensors that may enable the deployment of such sensors and sensor networks for recording these magnetic precursors to earthquakes. The sensors are based on magnetoelectric (ME) composites that produce a voltage response to the magnetic field fluctuations. The composites have two components: a ferromagnetic layer that responds to the magnetic field by producing a mechanical strain and a piezoelectric layer that converts the mechanical strain to a voltage. Several ferromagnetic-piezoelectric composites were prepared and characterized, and we found the best response in terms of low noise and sensitivity from sensors constructed of Metglas lead zirconate titanate (PZT). The data indicate that the samples can measure 0.1 nT at 0.1–10 Hz, a sensitivity ideal for the detection of 10–100 nT field fluctuations that precede an earthquake. Suggested follow-up efforts include a large-scale deployment of the sensors and sensor networks in earthquake-prone regions for field tests and ultimately the development of a sensor-based early warning system.

…the most reliable warning system today can provide only four to five seconds of lead-time, and that is only for places kilometers away from the epicenter of the earthquake. I started my research by scouring online journals looking for any ideas that I might be able to develop. I came across the theory that fluctuations in the Earth’s magnetic field might occur hours or even days before an earthquake. Recently, scientists have begun using magnetometers, devices that can measure the strength and direction of a magnetic field, to test this theory.

Existing magnetometers are capable of measuring changes in magnetic fields on the scale of 10 to 100 nanoTesla (nT), which is not precise enough to rule out environmental noise. The only existing device precise enough to measure what would be a minute change in the Earth’s magnetic field is called a SQUID magnetometer; however, because it uses liquid nitrogen, it operates only at extremely low temperatures and would thus not be practical for use in the field.
I decided to create a device that could precisely sense extremely small changes in magnetic field and that could operate at room temperature. I was going to need some help.

I developed a sensor made up of two components: a magnetostrictive part and a piezoelectric part. A magnetostrictive material is one that changes dimension when exposed to a changing magnetic field, and a piezoelectric material is one that produces an electrical charge when it is deformed (twisted or stretched). The magnetostrictive and piezoelectric materials available to me—Metglas and PZT, respectively—were not the most effective materials of their kind, but they were the only ones the laboratory was able to give me, so I had to try to make them work. The Metglas and the PZT components were connected with an epoxy binder. The idea was that when the device was exposed to a changing magnetic field, the Metglas portion would change its dimensions, causing the PZT portion to change shape and thus produce a measurable voltage.