What Quantum Sensors Are Used For

Four quantities account for almost every quantum sensor now on sale or close to it: magnetic field, acceleration and gravity, time, and light. Most overviews sort by industry instead, and that sort reads well but travels badly. A hospital physicist and a mine surveyor land in separate chapters while evaluating instruments built from the same physics, and a reader who finishes the article still can’t place the next datasheet that arrives.

Sorting by measured quantity holds up better. Each family has its own classical baseline, its own reason for existing, its own way of failing outside the lab, and its own honest maturity level. Learn the four and the industry applications arrange themselves.

One definition first, because the term gets stretched in marketing. A quantum sensor is an instrument whose measurement depends on a deliberately prepared, fragile quantum state, such as an atom held in superposition, an electron spin in a diamond, or a beam of squeezed light. An MRI scanner does not qualify, and neither does a semiconductor thermometer, even though both rest on quantum physics in the broad sense. The distinction matters for procurement, because it tells the buyer which existing instrument to benchmark the new one against.

Magnetic field

The workhorse here is the optically pumped magnetometer, or OPM. A small glass cell holds a vapour of alkali atoms, usually rubidium or caesium. A laser aligns the atomic spins, an external magnetic field makes them precess, and the amount of laser light the cell transmits changes as they do. Read the light, get the field. No liquid helium anywhere in the instrument.

Brain imaging without cryogenics

Magnetoencephalography, or MEG, measures the magnetic fields produced by electrical activity in the brain. Those fields are extremely weak, and the conventional instrument uses SQUID (superconducting quantum interference device) sensors, which need liquid helium and therefore sit inside a rigid insulating vessel. The vessel wall imposes a gap of a few centimetres between sensor and scalp, the helmet comes in one size, and the patient has to hold still.

Field strength from a current source falls off steeply with distance, so closing that gap is worth a great deal. OPMs run warm rather than cold, so they can sit directly on the scalp in a light helmet. A team at the University of Nottingham led by Matthew Brookes demonstrated a wearable OPM-MEG system in 2018, reported by Boto and colleagues in Nature, and participants could move their heads during recording. For paediatric neurology and for studies of movement, that changes what is measurable at all, not just what is comfortable.

The constraint is worth teaching alongside the capability. OPMs work only in a very low background field, so the scanner still needs a shielded room and active coils that cancel the residual field as the head moves. The cryogenics went away. The magnetic environment did not.

Magnetic imaging of chips and materials

A nitrogen-vacancy centre, or NV centre, is a defect in diamond where a nitrogen atom sits beside a missing carbon atom. Its electron spin can be prepared and read out with light at room temperature, and its resonance shifts with the local magnetic field. Mount one near the tip of a scanning probe and the instrument maps magnetic fields with spatial resolution in the tens of nanometres. Qnami builds commercial scanning NV instruments, used for failure analysis on semiconductor devices and for research on magnetic materials.

Both examples share one property, and it drives every purchasing decision in this family. A magnetometer measures the field at a point. Resolution comes from getting close and from adding channels.

Acceleration and gravity

The dominant technique is atom interferometry. Cool a cloud of atoms with lasers, then use laser pulses as beam splitters and mirrors for the atoms’ matter waves. The two paths pick up a phase difference set by the acceleration along the beam, and reading that phase gives the acceleration. The atoms are the test mass, and unlike a mechanical proof mass they don’t age, drift with temperature, or wear.

Finding what is underground

An absolute quantum gravimeter measures local gravity without needing calibration against a reference instrument, and it holds its accuracy over months. That suits long deployments where nobody wants to revisit the site: volcano monitoring, aquifer depletion, reservoir management, subsidence.

The harder problem is measuring gravity somewhere noisy. Vibration swamps a single gravimeter, which is why the gradiometer matters. It runs two atom clouds separated vertically and compares them, so vibration common to both cancels while the gravity difference survives. A University of Birmingham team, writing as Stray and colleagues in Nature, used a gradiometer to detect a buried utility tunnel from the surface, outdoors, in 2022. That result is the field’s real threshold, because the difficulty in gravity surveying has never been sensitivity in a basement.

Navigating without GNSS

For navigation, the figure of merit is drift. A classical inertial measurement unit accumulates position error steadily once GNSS, the satellite positioning signal, is jammed, spoofed, or simply unavailable underwater. Cold-atom accelerometers have very low bias drift, so the near-term architecture is hybrid: the quantum sensor periodically corrects the classical unit rather than replacing it.

Two honest limits shape where this lands first. Atom interferometers have limited bandwidth and a narrow dynamic range, so violent motion is a problem, and the instruments remain large and power-hungry. Submarines and ships come well before aircraft, and aircraft come well before anything handheld.

Time

An atomic clock counts oscillations of light or microwaves tuned to a transition between energy levels in an atom. Optical clocks, which use optical rather than microwave transitions, now reach fractional uncertainties near one part in ten to the eighteenth.

At that precision, a clock becomes a gravity sensor. General relativity says a clock higher in a gravitational field ticks faster, and a group at JILA (the Joint Institute for Laboratory Astrophysics), reporting as Bothwell and colleagues in Nature, resolved that shift across a strontium sample about a millimetre tall in 2022. Clocks compared across a network can therefore map gravitational potential, an approach called relativistic geodesy.

The near-term commercial use is much less exotic and much more immediate. Telecommunications networks, power grids, and financial trading venues all depend on GNSS for time, and GNSS jamming is now routine in several regions. Compact optical and chip-scale atomic clocks give those networks holdover, meaning they keep accurate time through an outage instead of drifting until services fail.

Light

Fair warning: this is the family where marketing runs furthest ahead of the physics, and the gap is wide enough to be worth naming.

Squeezed light is the mature part. Quantum mechanics sets a floor on the combined noise in two conjugate properties of a light beam, phase and amplitude, but the noise can be redistributed between them. Squeeze the property you are measuring, accept more noise in the other, and the measurement improves. LIGO injects squeezed vacuum into its interferometers and detects fainter gravitational-wave events as a result, which extended the volume of space the observatory can see. This works, it is in production, and it is confined to instruments where quantum noise is genuinely the limiting factor.

Quantum radar is the counterexample. The scheme, quantum illumination, sends one photon of an entangled pair toward the target and keeps its twin, then correlates the returning signal against it to pull a reflection out of noise. Barzanjeh and colleagues, writing in Science Advances, demonstrated the effect with entangled microwaves over short laboratory distances in 2020. The advantage was modest, it applies in a narrow noise regime, and the entanglement does not survive the amplification a real radar system requires. Treat vendor claims in this corner with more scepticism than anywhere else in the field.

Three questions to ask of any sensing claim

What quantity, over what range and bandwidth? A supplier who cannot answer this in one sentence is selling a demonstration, not an instrument.

What does the quantum part buy against the best classical option? Not against nothing. Against a SQUID array, a spring gravimeter, a ring-laser gyroscope, a rubidium clock. Several quantum sensors win on operational grounds rather than raw sensitivity, and that is a legitimate answer as long as somebody says it out loud.

Does the advantage survive leaving the lab? This is where the OPM shielding requirement and the gradiometer sit. The same gravimeter that resolves a centimetre of water table change on a concrete pier is limited by vibration on the back of a truck. The instrument didn’t get worse. The question changed.

Where to go next

Run those three questions against any product page and you’ll separate the deployable from the promising quickly. Then go deeper on whichever of the four families touches your work, because the engineering detail is where the buying decisions actually live.

Quantum Academy’s current quantum sensing programs, with enrolment details and prices, are listed at Quantum Academy. For a broader survey of applications across sectors, PostQuantum.com covers the same ground from the industry angle.