杏吧原创

IT鈥橲 LIKE something out of a bad dream. You鈥檙e stuck in a dance hall performing an interminable waltz. The hours go by and the dance continues. The hours melt into days, years, centuries, millennia. Eventually, billions of years have passed in which the universe has transformed into a featureless void populated only by you and your fellow indefatigable waltzers, dancing throughout eternity.

The vision is surreal, nightmarish 鈥� and entirely against the laws of physics. Anything that repeats on loop without an external energy source to power it seems to bend the cast-iron laws of thermodynamics, which govern how energy flows and can be exploited. So when five years ago, Nobel laureate Frank Wilczek speculated about a type of material that he called time crystals whose components could, in fact, do just that, he faced a wave of scepticism. 鈥淚 took a lot of grief,鈥� he says.

In the time since, Wilczek鈥檚 brainchildren have been championed, vilified, proved to be impossible, and now, apparently, made in the lab. If so, it鈥檚 the birth of an entirely new phase of matter, one that is fundamentally bizarre, perhaps confounding 鈥� and possibly even useful.

Time crystals might still be waiting to be invented if Wilczek were not the sort of person who gets bored easily. He won his Nobel prize in 2004 for theoretical insights into the nature of the strong force, which determines how fundamental particles interact within the atomic nucleus. He once described the experience of waiting for experimental verification of his theory as akin to watching grass grow. So when at some point his employer, the Massachusetts Institute of Technology, appointed him to teach a graduate course involving the structure of solid crystals, he was soon looking for ways to spice up the curriculum. 鈥淚t鈥檚 a beautiful subject, but kind of cut and dried by now,鈥� says Wilczek. 鈥淚 wanted to hint at something different.鈥�

Wilczek ended up throwing around ideas about symmetry in crystals with a former student of his, Alfred Shapere, now at the University of Kentucky in Lexington. All crystals, from salt flakes to diamonds, are made up of some basic unit of atoms that repeats over and over again in space. The temptation is to think of such a pattern as symmetrical, but to a mathematician or physicist, this structure actually breaks a kind of symmetry.

To see why, imagine asking for random samples of a carpet. If the carpet is plain, any pair of samples will be impossible to distinguish. But if the carpet is patterned, it鈥檚 likely no two are the same: if you overlay them, the patterns will not line up, just as they won鈥檛 if you take two random chunks out of a repeated solid crystal. In physicists鈥� speak, both the carpet and the crystals break translational symmetry in space.

So far, so ordinary. But Shapere challenged Wilczek to think of an object that would break translational symmetry in time: in other words, it would repeat naturally not in space, but in time.

Such things do not ordinarily exist on their own. A heart may constantly beat, and the hands on a clock may continuously move round the dial, but they need energy to power them 鈥� and hearts eventually run down and clocks cease to work. Claim to have built something that can run on a never-ending loop, with no power source, for as long as you please and you will get laughed out of the patent office. What you have designed is a perpetual motion machine: a long-discredited, thermodynamically impossible device for getting something out of nothing.

Wilczek鈥檚 new form of matter, however, exploited a loophole. Time crystals would do no useful work: they wouldn鈥檛 travel anywhere or power anything, and so technically would need no external energy source to run. For good measure, Wilczek imagined time crystals arising in a superconductor, a material that already exhibits unusual behaviour by allowing electrons to flow without resistance.

When Wilczek and Shapere published a paper outlining their thought experiment, the pick-up was immediate. 鈥淚t generated a lot of excitement,鈥� recalls Vedika Khemani, a researcher at Harvard University.

At the time, Haruki Watanabe was a PhD student at the University of California, Berkeley. One of the questions during the oral examination for his degree was about what would happen if symmetry broke in the way that Wilczek proposed. Watanabe didn鈥檛 know, and the question festered.

As is clear from the title of the paper he published two years later, 鈥溾��, the answer he came up with was unambiguous. Time crystals as proposed by Wilczek were still thermodynamically impossible. The problem, Watanabe showed, lay in Wilczek鈥檚 proposal that the dancers be in constant, seamless motion, entirely independent of the outside world. These dancers would be in thermal equilibrium, said Watanabe, a state he was able to painstakingly prove could never display periodic behaviour.

鈥淭hat was that,鈥� says Khemani. 鈥淓veryone thought that time crystals were impossible.鈥� But the vagueness of Wilczek鈥檚 thought experiment offered a lifeline: a time crystal might still exist, just not in thermal equilibrium.

Khemani ended up exploiting that get-out clause almost by chance. When Watanabe鈥檚 paper came out, she was at Princeton, working on a problem called many-body localisation, which can be summed up as 鈥渁 bunch of tiny particles get stuck鈥�. An example might be if all the air particles in a room, instead of filling the space evenly with gas, were hanging out in one corner. That鈥檚 not a normal happening: thermodynamics says that, left to their own devices, things tend to evolve towards messy, featureless, randomly distributed states.

But tickle a system in precisely the right way, Khemani found 鈥� gently, not imparting any energy to it 鈥� and keep on tickling it in a regular rhythm, and you could prevent that happening. In fact, you could get a bunch of particles to move in space at a rhythm different to the one you鈥檙e poking it at. It鈥檚 a bit like a sponge that you squeeze and release, says Khemani, except that the material doesn鈥檛 follow the rhythm of your squeezing, but expands and contracts at only half the rate, say, as though you were squeezing it half as quickly.

Loophole to hula hoop

It wasn鈥檛 until Khemani鈥檚 paper reached peer review that someone pointed out the connection with time crystals 鈥� although it was of a new sort, as her system was not in thermal equilibrium. With the loophole now expanded into a hula hoop, all time crystals needed was the right experimental home. And Soonwon Choi, a graduate student at Harvard, thought he could provide it. While listening to Khemani lecture about her idea in 2016, it occurred to him that the experiment his colleagues were using to observe strange quantum effects could be an ideal test bed for a time crystal.

Choi and his group were working with a black diamond, a close cousin of those clear rocks that adorn engagement rings, but with a few million carbon atoms swapped out for nitrogen atoms paired with little empty spaces. These impurities, meticulously inserted over months, give them their distinctive colour and extravagant price. More than that, however, they offer just the kind of disorder necessary to allow for the weird effects Khemani had envisioned.

Each nitrogen atom/empty space pair in the diamond has a property called spin, which can be made to flip between two values using va microwave signal. Choi and his colleagues found that they could send the microwave signal twice for every spin flip, just like the sponge moving half as fast as the rhythm of the hand squeezing it.

The Harvard experiment wasn鈥檛 unique. Nearly 7 hours鈥� drive south on I-95, in experimentalist Chris Monroe鈥檚 lab at the University of Maryland, another group had built their own version of a time crystal. It was another possible rebel child of Wilczek鈥檚 theory, based on a slightly different theoretical blueprint. It consisted of just a handful of atoms 鈥� ytterbium ions to be exact 鈥� flipping spins in unison inside a chamber the size of a softball. The atoms were responding to lasers pulsing at twice the frequency of the flips.

鈥淚t鈥檚 as weird as a roomful of dancers leaving the party, going home, and continuing to dance in perfect lockstep鈥�

鈥淢y god,鈥� Wilczek recalls thinking when he read the papers this January. To him, this was a major justification of his idea. What鈥檚 more, he鈥檇 had no idea that the experiments were under way. 鈥淚t was a very delightful surprise.鈥�

Watanabe was unconvinced. To him, an object that needs an external signal to loop in time is no more a time crystal than a beating heart or a swinging pendulum. 鈥淚n my opinion, the phenomenon should occur by itself,鈥� he says. The fact that time crystals were popping up in Massachusetts and Maryland represented little more than overenthusiastic branding.

Wilczek himself doesn鈥檛 entirely disagree, and nor, for that matter, does Khemani. But she does see an important distinction between her work and the everyday phenomena Watanabe describes. 鈥淚t is unfair to say that they are just pendulum-like,鈥� says Khemani. That鈥檚 because the atoms in time crystals are separated in space, and yet are all still flipping together, cooperating at a distance. Pendulums and hearts don鈥檛 do that. It would be as weird as a roomful of dancers leaving the party, going home, and then continuing to dance in perfect lockstep.

That feature makes the creations interesting in their own right, Watanabe admits. And now that they exist, says Wilczek, it鈥檚 easier to see the implications of time crystals. Monroe鈥檚 set-up, for example, is one he typically uses to create elements for quantum computers. Time crystals could, one day, fit into that category. The fact that the spins align and appear to cooperate means that they could serve as a clock for a future machine or, because they have the capacity to maintain their state as time plods on, a way to store information.

They could also help perform MRI scans on tiny things, says Mikhail Lukin, the principal investigator at the Harvard lab. A time crystal could respond to a relatively weak signal coming from, say, a cell or even an individual protein, and flip millions of spins in unison as a result, amplifying the signal.

Happily ever after?

But for now, Monroe points out, the time crystal doesn鈥檛 last long enough to do anything more than let us observe that it exists. Thanks to imperfect equipment, the processes used in the lab are 鈥渓ike deliberately throwing sodium and chlorine crystals together one at a time,鈥� he says, technically capable of creating salt crystals, but ones far too small to use as seasoning.

鈥淥f course, naturally, salt crystals appear all the time,鈥� says Monroe. That means there could be larger time crystals out there in the wild.

Wilczek still believes that time crystals could hypothetically last forever. 鈥淭here鈥檚 no clear upper limit,鈥� he says. Experimentalists are more sceptical 鈥� and there are some practical aspects to consider, too. Phil Richerme, who has been watching the emerging time crystal research from his post at Indiana University at Bloomington, drily notes that 鈥渋n practical terms, the sun is going to explode and take out the lab.鈥�

In the immediate future, the hope is that the properties of this new phase of matter can be explored in more detail. 鈥淚t鈥檚 like we鈥檝e stumbled on to some new territory,鈥� says Wilczek. 鈥淲e don鈥檛 know if it鈥檚 an exotic island, or a continent.鈥�

For Khemani, the biggest consequences of the experiment are more subtle: it means that interesting things happen out of equilibrium. 鈥淧hases have been studied for so long,鈥� she says 鈥� the handful we knew about seemed to be all that were out there. Now, she says, 鈥減eople are thinking about phases in this out-of-equilibrium setting.鈥� It opens up a whole new realm of matter, one sure to keep physicists dreaming for some time to come.

This article appeared in print under the headline 鈥淣ever-ending story鈥�