Scientists Found an Ion That Makes Atomic Clocks 20× More Efficient

Here's a fun thing to picture: atomic clocks are already so ridiculously accurate that they could run for the entire age of the universe and only drift by about a second. They're basically the most precise things humans have ever built. So you'd think we'd be done, right? Not quite. Scientists have been bumping into a frustrating wall whenever they try to make these clocks even better — and a team from PTB in Germany just found a surprisingly elegant way around it, buried inside one of the most peculiar atomic nuclei in the periodic table.

The team — led by Jialiang Yu and Tanja Mehlstäubler, with collaborators from Leibniz University Hannover and Thailand's National Institute of Metrology — spent years poking at a rare ion called ¹⁷³Yb⁺ (that's ytterbium-173, if you want to impress someone at a dinner party). What they found, published in Physical Review Letters in January 2026, turned out to be genuinely exciting news for everyone who cares about clocks, quantum computers, or just cool physics.

Wait — What's the Problem With Atomic Clocks Right Now?

Let's back up for a second. The way an optical atomic clock works is actually pretty elegant. You trap a single ion — a charged atom — in mid-air using electric fields, cool it down to almost absolute zero, then shine a laser at it. That laser nudges one of the electrons to jump between two energy levels. The frequency of that jump is incredibly stable, so you use it as your "tick." Simple enough, right?

The ion most labs use is ytterbium-171 (¹⁷¹Yb⁺). Its clock transition is a quantum jump so unlikely — so "forbidden" by the rules of physics — that once the electron makes the leap, it stays put for about 1.6 years before coming back down. That extreme reluctance to decay is exactly what makes it such a great clock. But here's the catch.

Now scale that up to multiple ions and the problem gets messy fast. A laser beam is brighter in the middle than at the edges, so ions sitting at different spots in your crystal each feel a different intensity — which means each one gets a slightly different frequency error. It's like trying to keep three metronomes in sync when each one is being poked with a different strength. For years, this has been the main reason the world's best clocks still run on a single lonely ion rather than a whole team of them.

So What's Special About This ¹⁷³Yb⁺ Ion?

Here's where it gets genuinely interesting. Back in 2016, theorists Dzuba and Flambaum had a hunch about the odd ytterbium isotope ¹⁷³Yb⁺. Unlike its more common cousin ¹⁷¹Yb⁺, the ¹⁷³ version has a nuclear spin of 5/2 — and crucially, its nucleus is physically deformed. Not "broken" deformed, just shaped more like a slightly squashed rugby ball than a perfect sphere. That shape gives it an unusually large electric quadrupole moment, which is a fancy way of saying the nucleus and the surrounding electrons interact with each other much more strongly than usual.

That stronger interaction does something remarkable: it mixes the normally very long-lived clock state with other nearby energy states that decay much faster. Certain hyperfine sub-levels of the clock state — the fine-grained energy rungs created by the nuclear spin — end up with a completely new, much faster decay pathway available to them. Scientists call this effect nuclear spin quenching. The clock state doesn't disappear, but it becomes a lot more "accessible" to the laser. And if you need less laser power to drive the transition, the pesky Stark shift shrinks right along with it.

What Did the Experiment Actually Show?

The PTB team didn't just theorise — they got their hands dirty and measured the whole thing. They trapped a single ¹⁷³Yb⁺ ion in a radio-frequency Paul trap, cooled it to below 1 millikelvin (colder than anywhere in deep space), and prodded it with an ultrastable laser tied to a cryogenic silicon cavity. Then they measured the Rabi frequency — basically, how fast the electron wobbles back and forth between the two clock states — for several different hyperfine sub-levels.

For the Fₑ = 4 sub-level, the Rabi frequency came out 4.22 times higher than the pure physics model predicted. That's not a measurement error — that's a clear fingerprint of the nuclear spin quenching kicking in and opening up a much more efficient decay channel. The inferred lifetime of that state works out to roughly 49 days, compared to 1.6 years for the unquenched reference state in ¹⁷¹Yb⁺. About 12 times shorter. And along the way, they mapped out 15 previously unknown transition frequencies — a complete atlas of the clock state's hyperfine structure that nobody had ever measured before.

Why Should Anyone Outside a Physics Lab Care?

Fair question! Here's the honest answer: better atomic clocks touch more of your life than you might expect. The precision timekeeping from atomic clocks already underpins GPS navigation, financial transaction timestamps, and internet synchronisation. The next generation of clocks — the kind this research is helping to build — could detect the tiny gravitational difference between two cities at different altitudes, help redefine the official SI second, and test whether the fundamental constants of physics are actually drifting over cosmological time.

There's also a quantum computing angle that's genuinely exciting. The rich hyperfine structure of ¹⁷³Yb⁺ makes it a great platform for storing quantum information in multiple ways at once — what some researchers call "polyqubit" processing. And the quenched Fₑ = 4 transition, with its shorter ~49-day lifetime, turns out to be an ideal tool for reading out qubit states quickly and accurately, which is one of the trickiest engineering problems in building practical quantum computers. It's a two-for-one.

The nuclear physics community also gets something valuable here. The ¹⁷³Yb nucleus has been notoriously hard to characterise — there's even a long-running disagreement between two different experimental measurements of its magnetic octupole moment. The 15 new transition frequencies measured in this study give nuclear theorists the best experimental data they've ever had to work with on this particular nucleus.

Okay, What's the Catch? (There's Always a Catch)

The team is refreshingly honest about the limitations, and there are a couple worth knowing about. The quenched Fₑ = 4 state turns out to be quite sensitive to magnetic field wobbles — technically, it has a large quadratic Zeeman sensitivity. In practice, that means magnetic field noise in the lab could limit how long you can interrogate the ions before the measurement gets smeared out. There are ways to manage this (special magnetic field settings, averaging techniques), but it adds engineering complexity to any future clock design.

There's also a genuinely puzzling gap in the theory. The measured transition rates came out about 28 times different from the theoretical predictions made by Dzuba and Flambaum back in 2016. Nobody has a clean explanation for that yet, and it's going to take some fresh theoretical work to unpick. As physics mysteries go, "our experiment was way more dramatic than the theory predicted" is actually a pretty exciting one to have. And the team's own back-of-the-envelope numbers suggest that pairing the quenched ¹⁷³Yb⁺ transition with a segmented Paul trap and a flat-top laser beam could let you run up to 100 ions simultaneously — a tenfold jump in clock stability over today's single-ion systems, and a whole new tier of physics experiments that simply aren't possible today.

• Clocks are about to get a lot quieter. With the Stark shift down 22×, the biggest technical barrier to running many ions together has finally been cleared — and that means stability improvements that would have taken decades to squeeze from single-ion systems.
• Quantum computers get a handy new trick. The quenched transition's ~49-day lifetime makes it perfect for fast, high-fidelity qubit readout — one of the trickiest engineering challenges in building practical trapped-ion quantum processors.
• Nuclear physics also wins. The 15 newly measured transition frequencies give nuclear theorists their best-ever experimental look at ¹⁷³Yb's famously awkward nucleus, potentially resolving a long-running disagreement about its magnetic octupole moment.

"Sometimes the best breakthroughs come from looking carefully at something everybody else wrote off as too complicated — a misshapen nucleus, an odd isotope, a quirk of quantum spin — and asking: what if this isn't a problem, but actually the answer?" — Yu, Prakash, Zyskind et al., Physical Review Letters 136, 023002, 2026.