Ekkehard Peik聽is a clock-maker. But instead of spending his days looking at tiny cogs and springs through a magnifying glass, the tools of his trade are powerful lasers, wires and, occasionally, radioactive atoms. Peik, director of the German metrology institute (PTB), is one of a handful of physicists who have spent the best part of three decades trying to make the most accurate timepiece in the universe.
Since the 1950s, researchers have been constructing atomic clocks, the very best of which are now so accurate they only lose a second in around 31 billion years. But these are about to be replaced by a new model: the nuclear clock.
Advertisement
This promises to outperform its atomic counterparts both in terms of precision and accuracy. A nuclear clock would, in principle, only drop a second every 300 billion years. Why, you might ask, would we ever need something with such mind-blowing precision? Because it will be used for something much more exciting than simply telling the time. Nuclear clocks could help probe some of the deepest mysteries of the universe, including the nature of dark matter and some of the elusive fundamental forces that shape our cosmos.
The tick of today鈥檚 atomic clocks is the result of electrons that oscillate between a pair of shells around the nucleus of an atom. The transitions between these shells are driven by shining lasers at the atoms involved at just the right frequency to match that of the oscillations, a state that is known as resonance. This resonant frequency, the number of oscillations of light per second, sets the tempo of the clock.
In the case of atomic clocks, the number of oscillations each second is extremely stable, so we can use them as very precise timekeepers. The best atomic clocks use ytterbium or strontium atoms and have ultra stable resonances. This exceptional accuracy enables GPS, keeps financial systems in check and allows us to test the limits of fundamental laws of physics, like the time dilation predicted by Albert Einstein鈥檚 general theory of relativity.
But atomic clocks have limitations. Electrons form the outer part of an atom, and as such they are vulnerable to interference from electric and magnetic fields that can cause the resonances to shift. 鈥淲hat limits these clocks is that in the lab you have to control all the environmental parameters 鈥� the smallest magnetic fields, the smallest electric fields, everything,鈥� says , a physicist at KU Leuven, a university in Belgium. 鈥淪o, in practice, we鈥檙e always fighting that.鈥�
What is a nuclear clock?
In 1996, , a physicist at Moscow State University, realised there was a better solution. Instead of using electrons on the edge of an atom, a neutron moving between two energy levels inside an atom . There were clear reasons such a device would outperform an atomic clock. While the electron shells are affected by stray electric and magnetic fields, neutrons are impervious to these forces. Instead, they are only affected by the strong nuclear force, which interacts over extremely short distances 鈥� less than the width of an atom. A nuclear clock would measure time with orders of magnitude more precision than an atomic clock.
For most atomic nuclei, driving such a resonance would require a laser more powerful than we have yet invented. There is an exception though 鈥� the radioactive element thorium-229 (Th-229). This needs an unusually low amount of energy to create its oscillation. 鈥淚t鈥檚 still high compared to what people typically use nowadays in atomic clocks, but it鈥檚 so low that one could imagine driving this with a laser,鈥� says , a physicist at the University of California, Los Angeles.
Physicists set to work to be the first to achieve this neutron oscillation. The initial, and biggest, hurdle was that no one knew exactly what energy was required. Only by knowing this very precisely could a laser be tuned to produce the resonance. This value couldn鈥檛 be calculated, only determined through careful experimentation.
Probing the strong force
The reason for this lies in the strong nuclear force. As the name suggests, it is the strongest of the fundamental forces, but only over the shortest distances. It confines fundamental particles called quarks together in a bubble to make protons and neutrons. But move a tad outside the bubble, and the strong force peters out to nothing.
To this day, the physics behind the strong nuclear force makes it extremely difficult to make practical calculations of the entire nucleus. 鈥淭here is a pretty good model of strong interaction in the standard model [of particle physics],鈥� says Kraemer. 鈥淗owever, to describe a whole nucleus, there are so many potential interactions.鈥� As a rule, in such cases, physicists rely on approximations. Yet, even with these simplifications, our largest supercomputers don鈥檛 have enough capacity to calculate the physics of heavy atoms like Th-229. This means values related to the inner workings of the nucleus, like the Th-229 resonance, must be found by trial and error.
The number of possibilities was so large that finding the right one was like looking for a needle in a haystack. Over several years, starting in 2009, physicists like at the Lawrence Livermore National Laboratory in California narrowed down the Th-229 transition energy to . Then, Hudson hit on a novel approach. Instead of measuring the transition from a single Th-229 atom, as others had been doing, he proposed by embedding billions of atoms into crystals. This would finally allow a more precise measurement of the energy required, and make a nuclear clock feasible.
Fast-forward to 2023, and physicists working at CERN, the particle physics laboratory near Geneva, Switzerland, in the energy gap even further. On the back of this, both Hudson鈥檚 and groups refocused their search efforts to this much narrower region. 笔别颈办鈥檚 group started scanning from the upper end of the new range, while Hudson鈥檚 started from the lower end.
Then, one day in November 2023, Peik wrapped up a meeting and headed to his lab, where he was surprised to walk into a room full of excited colleagues. Finally, after decades of searching, they had found . 鈥淚t鈥檚 a rare event in the life of a researcher where you really see something for the first time,鈥� he says. 鈥淭hat鈥檚 a great experience.鈥� Hudson鈥檚 group soon , too. 鈥淭hey beat us by a little bit, but that鈥檚 OK,鈥� says Hudson. 鈥淚t鈥檚 an exciting time for everyone.鈥�
With this huge hurdle cleared, the only step left to making a usable clock was to boost the resonance signal from the nucleus of Th-229, which was still too weak. To get a usable signal for a working clock would require a laser much more sharply tuned at exactly the right energy. In late June, at the JILA research institute in Boulder, Colorado, and his group revealed that they . 鈥淪eeing the direct frequency measurement sent a chill to my body,鈥� says Ye.
In their paper, Ye and his team describe their set-up as amounting to a nuclear clock, but not everyone agrees. 鈥淛un has made spectacular progress,鈥� says Peik, but it doesn鈥檛 quite meet the definition of a clock. 鈥淎 clock is a device that shows time,鈥� he says. 鈥淚t should be able to provide a ticking at a well-controlled frequency and to do so for a reasonably continuous period.鈥�
While Ye鈥檚 set-up isn鈥檛 stable enough to meet this definition, since it worked only for a limited number of cycles, it has proven useful as a tool for probing questions in fundamental physics. 鈥淚t鈥檚 the dawn of precision metrology for nuclear physics,鈥� says Ye. For his first experiment with the device, Ye and his team used it along with a strontium atomic clock to measure something called the nuclear electric quadrupole moment inside a Th-229 atom, a measure of how symmetrical the charge is within a nucleus, which naturally has a shape like a stretched sphere.
The team hoped to find out if the quadrupole moment changes slightly when the nucleus moves to a slightly higher energy state. 笔别颈办鈥檚 lab ran a similar experiment in 2018, and no such changes were detected. But Ye鈥檚 lab found a tiny change of 1.8 per cent. 鈥淣o one has ever been able to observe this experimentally,鈥� said Kraemer. 鈥淭hat鈥檚 a really very big step forward.鈥�
Ye and his colleagues used the shifting quadrupole measurements to show that a property called the nuclear volume of Th-229, which is related to the shape of the nucleus, must be altering too. They discovered that the minuscule shape shifts can be used to determine the sensitivity of the nuclear transition frequency to a universal constant known as the fine structure constant.
This research relates to an overarching question in physics to do with these universal constants, which are observed values 鈥� like the strength of the various forces 鈥� that define how the universe works. The tiniest change in any of these constants would have an impact on everything from subatomic physics to cosmic structures. The question is: have these values really always been the same? 鈥淵ou think about nature, how many things are actually constant?鈥� says Hudson. 鈥淓verything is dynamic at some level.鈥�
Testing nature鈥檚 fundamental constants
Nuclear clocks are the best bet to answer this question. Researchers could construct a nuclear and an atomic clock 鈥� one ruled primarily by the strong nuclear force and the other by the electromagnetic force 鈥� and let them tick, looking for subtle changes in how they keep time. Any differences could be traced back to teensy shifts in nature鈥檚 constants, such as the fine structure constant. Others hope to use the clocks to answer other questions too (see 鈥淭he clock that hunts dark matter鈥�, below).
There are plenty of applications in tech, too. Unlike the electrons in an atomic clock, which repel each other, the functioning of neutrons in a nuclear clock is unaffected when the atoms are packed densely together. This means a nuclear clock based on Th-229 atoms could be far more stable than an atomic clock, because the large number of atoms would average out the noise inherent to such systems. This means the clock would be a good candidate for next-generation GPS satellites, as such traits are required to make very precise measurements of location. 鈥淭he atomic clocks in GPS satellites are nowhere near as good as what we have in the lab,鈥� says Hudson, who also points out that nuclear clocks are portable and can run at any temperature.
Eventually, nuclear clocks could be used to redefine time. Currently, a unit of time is based on the rate of oscillation of an electron between two electron shells inside a caesium atom, and it hasn鈥檛 changed since 1967. But perhaps in future, the second could be defined according to neutron oscillations. 鈥淚t takes a long time for standards to change,鈥� says Hudson, 鈥渂ut I imagine a [thorium-based] clock will eventually be the standard.鈥�
There鈥檚 just one fly in the ointment: Th-229 is incredibly scarce. Because of its radioactive decay, it can鈥檛 reliably be found in nature. Today鈥檚 stockpile is a byproduct of nuclear weapons programmes that ended in the 1990s, leaving only 40 grams of high-quality Th-229 on Earth. Kraemer says CERN could produce more, but the facility isn鈥檛 ideal for manufacturing large quantities. 鈥淥n the other hand, the amount needed is also rather low,鈥� he says. A recent experiment only needed 0.17 micrograms, so today鈥檚 supply could potentially enable around 200 million nuclear clocks. It is certainly enough to get started.
The clock that hunts dark matter
Nuclear clocks could be a surprisingly useful tool in the hunt for dark matter, the mysterious stuff thought to make up more than 80 per cent of all the matter in the universe. In particular, these clocks, which are currently being developed (see main story), might be sensitive to a specific candidate known as ultralight dark matter.
Dark matter is only known to interact with other matter gravitationally, making it notoriously hard to detect. But some models predict that ultralight dark matter through the strong nuclear force. If physicists could , it would provide evidence that ultralight dark matter exists.
