Why sensors got there first
A quantum computer needs many qubits to stay coherent together and to interact on command. A quantum sensor needs one well-understood quantum system to respond to its environment as precisely as possible. That is a far easier engineering problem, and it explains why atomic clocks have defined the second since 1967 while the first error-corrected quantum computers are only now appearing. The recent breakthroughs in sensing are therefore refinements of working instruments rather than proofs of principle, which changes how we read them.
Clocks: from caesium to optical transitions
The SI second is defined by the hyperfine transition of caesium-133, fixed at exactly 9 192 631 770 hertz. Caesium fountain clocks at national metrology institutes realise this definition and are the reference against which everything else is judged. Optical clocks, which interrogate transitions at frequencies roughly a hundred thousand times higher, have for some years outperformed caesium in both stability and reproducibility. Two designs lead: optical lattice clocks holding thousands of strontium or ytterbium atoms in a standing wave of light, developed at laboratories including JILA, NIST, RIKEN and the National Physical Laboratory, and single-ion clocks based on species such as aluminium or ytterbium ions.
The lattice approach averages over many atoms and so reaches a given precision faster; the ion approach is easier to shield from perturbations but must average longer. Neither has a decisive edge, which is one reason the international committee responsible for the second has set out a roadmap towards redefinition rather than a decision. We would expect the eventual definition to rest on an optical transition or a weighted combination, but the timetable is the committee's to set and we would check its published documents rather than trust any secondary summary, including this one.
The unexpected application is geodesy. General relativity says a clock higher in a gravitational field runs faster, and optical clocks can now resolve the difference between two heights a centimetre or so apart. Comparing clocks across a country therefore measures the shape of the gravitational potential directly, a method called chronometric levelling that national mapping agencies have begun to explore.
Magnetometers: diamond, vapour and the old guard
Superconducting quantum interference devices, or SQUIDs, have measured the magnetic fields of the brain in hospital magnetoencephalography systems since the 1980s. They are exquisitely sensitive and they need liquid helium, which fixes the sensor array in a rigid helmet that does not fit children well or tolerate movement.
Two newer sensors relax that constraint. Optically pumped magnetometers use a warm vapour of alkali atoms whose spin precession is read out with a laser; because they run near room temperature they can be mounted in a wearable helmet, and groups at the University of Nottingham and University College London have published on this for brain imaging. Nitrogen-vacancy centres in diamond use a defect whose spin state can be read optically; they trade some sensitivity for spatial resolution down to the nanometre scale and for robustness, and are being developed for applications from battery inspection to imaging currents in chips. The trade-off between the three is roughly sensitivity against cryogenics against resolution, and there is no single winner.
Gravimeters that see underground
An atom interferometer drops a cloud of cold atoms and uses laser pulses to split, redirect and recombine their matter waves. The phase at the output depends on the acceleration the atoms felt, which is a direct measurement of gravity. In 2022 a group at the University of Birmingham published in Nature the detection of a buried utility tunnel using a gravity gradient sensor of this kind operated outdoors, in real conditions rather than in a laboratory. Measuring a gradient rather than the field itself cancels much of the vibration noise that has limited earlier instruments, and that design choice is the real advance.
The obvious uses are surveying, civil engineering and archaeology, where the current alternative is drilling. The obstacles are size, laser complexity and the time each measurement takes. Commercial cold-atom gravimeters exist, but the instruments that can map a site quickly are still in development, and the field-ready gradiometer remains a research instrument.
Squeezed light at LIGO
The clearest case of quantum measurement doing frontier science is gravitational-wave detection. Since the observing run that began in 2019, the LIGO detectors have injected squeezed vacuum states into their interferometers, reducing quantum noise in one variable at the expense of another. The gain in sensitivity is modest in isolation, but it applies at frequencies where nothing else helps, and it has been credited with increasing the rate of detected events. Frequency-dependent squeezing, added in the following run, extends the benefit across the detection band.
Reading claims in this area
Ask which noise limit a sensor is actually near. Many instruments described as quantum are limited by ordinary technical noise, and the quantum advantage is potential rather than realised. Ask whether the result came from a laboratory or a field deployment. And ask what the incumbent technology is, because in sensing there almost always is one, and it is usually cheaper.