Peking University scientists unveil levitating magnetometer for femtotesla-level brain sensing
This digest was compiled by AI from multiple sources — links to the originals are below.

Peking University physicists report today in Science a levitating-magnet magnetometer that detects magnetic fields as weak as a few femtotesla. The device, which needs no cryogenics or elaborate shielding, can measure brain activity with a sensor-sample spacing of a few hundred micrometres. The approach is significantly simpler than existing SQUID or SERF technologies.
The Levitating Magnetometer
Wei Ji and colleagues at Peking University suspended a sub-millimeter magnet between a lifting magnet and a diamagnetic material, which stabilizes the sensor via repulsion. A laser reflects off the magnet onto a detector, tracking orientation changes when a sample approaches. The entire setup fits inside a vacuum chamber about the size of a Tupperware box. Magnetic coils and damping systems suppress external noise, enabling detection of fields as weak as a few femtotesla. Ji notes that even a thin aluminium film can introduce 100 femtotesla of interference.
Performance Advantages Over Existing Sensors
The new magnetometer operates at room temperature and without the near-perfect shielding required by spin-exchange relaxation-free (SERF) devices. It is orders of magnitude more sensitive than diamond-based magnetometers and reduces the sample-to-sensor distance to a few hundred micrometres, compared with larger gaps in SQUID-based systems that need cryogenic cooling. The sensor’s sensitivity reaches the femtotesla scale, ten billion times weaker than Earth’s magnetic field.
What's Next
The team is investigating the magnetometer’s use for recording ultrafaint brain signals, potentially enabling non-invasive neural imaging. Whether the device can transition from laboratory demonstration to a practical clinical tool remains an open question.
1 source
Peking University scientists unveil levitating magnetometer for femtotesla-level brain sensing



