Hacker News new | past | comments | ask | show | jobs | submit login
Scientists unveil new form of matter: Time crystals (eurekalert.org)
200 points by mpweiher on Jan 28, 2017 | hide | past | favorite | 69 comments



It is funny how advanced physics sounds to my layperson ears like something a stoned teenager might come up with.

This reminds me somehow of the "one-electron universe": https://en.wikipedia.org/wiki/One-electron_universe

(Just to be clear: I am not mocking physicists, it just seems the universe is really, really weird.)


Or, as I would put it, the universe at large is simply vastly richer than our limited intuitions, a natural consequence of the limited range of physical properties that supports human life. It would be really strange if the universe didn't seem weirder and weirder the further we stray from our immediate perceptions.


Our of curiosity, what is the argument against a one-electron universe (or positrons being electrons moving backward in time)?


> The idea was proposed by Frank Wilczek in 2012. His speculation was that a construct would have a group of particles that move and periodically return to their original state, perhaps moving in a circle, and form a time crystal. In order for this perpetual motion to work, the system must not radiate its rotational energy.

I thought perpetual motion violates the laws of thermodynamics. Does quantum physics cause the laws of thermodynamics to break down?


The kind of perpetual motion that violates thermodynamics is the kind that can do useful work, i.e. that you can get energy out of, without putting any energy into it. So it violates conservation of energy.

A planet in orbit, or a spaceship like Voyager travelling through the galaxy, is in perpetual motion, but if you try to harness that kinetic energy, you'll slow it down.


Electrons and photons are in perpetual motion. Perpetual motion is only a thermodynamic problem when the system is shedding energy in some way (friction, radiation, photon smacking into a detector).


Or an example that's easier to visualize would be a spinning rock in space.


Thermodynamics are a statistical model to describe molecules behavior at a macroscopic scale. It is usually not very relevant in the quantum world.


Niels Bohr proposed that electrons don't have enough energy to radiate away, and thus are forced to continue to spin around the nucleus.[0] This was the seed that fruited the quantization of energy and eventually our modern understanding of Quantum Mechanics.

In a way, the quantization of reality at very small scales does appear to prevent the laws of thermodynamics from occurring. This is just one form of Zero-Point Energy[1]. Maybe these time crystals are another form - maybe some large objects radiate some kind of time-entropy, and these are too small (or configured in some other way), such that they are also mechanically prevented from doing so?

IANA theoretical physicist - I merely play one on the internet. Expert opinion and advice is welcome.

[0]https://physics.stackexchange.com/questions/18473/where-do-e...

[1]https://en.wikipedia.org/wiki/Zero-point_energy


Disclaimer: Theoretical physics postdoc, working in an unrelated subject area. Not necessarily an expert, but I do have some comments.

> Niels Bohr proposed that electrons don't have enough energy to radiate away, and thus are forced to continue to spin around the nucleus.[0] This was the seed that fruited the quantization of energy and eventually our modern understanding of Quantum Mechanics.

The first sentence is correct, though the history directly begins in 1900 with Planck's quantization hypothesis in the context of black body radiation. Bohr's paper, which I assume you're referring to, came in 1913.

> In a way, the quantization of reality at very small scales does appear to prevent the laws of thermodynamics from occurring. This is just one form of Zero-Point Energy[1]. Maybe these time crystals are another form - maybe some large objects radiate some kind of time-entropy, and these are too small (or configured in some other way), such that they are also mechanically prevented from doing so?

I do to take an issue with this paragraph, in that it doesn't really map to any well defined concepts in modern physics. The "quantization of reality at very small scales" (by which I assume you're referring to the speculative idea of fundamentally quantized spacetime) doesn't really have anything directly to do with thermodynamics - the laws of thermodynamics are necessarily statistical laws, produced by the behavior of ensembles defined by particular distributions. They have no particular validity in (most) very small scale systems and I don't see how the quantization is relevant here. Nor do any of these remarks help you violate the conservation of energy with respect to any larger systems.

I have no idea what "time-entropy" means, but generally speaking this notion of time crystals doesn't have anything intrinsically to do with violating energy conservation or producing useful work (in the physical sense) in any new fashion. Rather, it represents a fairly ingenious way of applying the established mathematical tools of analyzing symmetries to systems that exhibit complex periodic behavior rather than just spatial symmetries. In that regard, I think the example of satellite motion presented by Latham Boyle [1] was very illustrative without being misleading or overly suggestive.

I hope this response isn't overly snarky, I just wanted to clarify a few misconceptions I've seen a lot in the last few days.

[1] https://arxiv.org/abs/1407.5876


An electron at rest has 511 keV of energy, which is ~10^6 times the energy of visible light (~1 eV), so from energy considerations alone it has more than enough to "create" a photon. Accelerating electrons are a great source of light via Bremsstrahlung and synchrotron radiation.

This is the fundamental reason why electrons can't orbit the nucleus in a classical sense (and why we know the Bohr model is incorrect). The atom is a quantum object and as such it does not behave in a classical manner. You can describe it heuristically as an electron 'orbiting' the nucleus but that is not what's actually happening. The very concept of a path or trajectory does not have any meaning for quantum objects, which exist in probabilistic quantum states.


The first law of thermodynamics is basically conservation of energy. The internal energy in a closed system cannot change. In this case, if 'the system does not radiate its rotational energy', the particles will keep on moving indefinitely.

Since you ask, QM can sort-of 'break' this law though, very temporarily, via the uncertainty principle. Many particle decays are mediated by 'virtual particles', which are heavier than the input particles. For example, a neutron can decay into an electron+neutrino+proton (Beta decay), this is mediated by a W boson which is ~80 times heavier than the neutron. So in a way energy is 'borrowed' to create the W, which then promptly decays into the lighter electron+neutrino, so in the end energy is conserved.


Not really, It depends on the temperature of the particular system, but if the crystals are above something like 15 Kelvin then the particles should have a decent amount of energy each.


It seems remarkable that time crystals were theorized in 2012 and confirmed in 2017. It took several decades for experimental methods to catch up to the Higgs Boson for example.


Well, the Higgs has strict technical requirements -- a collider capable of getting comfortably over the energies needed to create the thing without it fading into background noise. That took a lot of engineering progress, software, money, and political will to get to. Whereas this is a much less.. involved creation. The experimental requirements aren't nearly as vast.

(I found some details on the requirements here: https://physics.stackexchange.com/questions/246841/why-did-i... )


What's the reasoning behind calling it non-equilibrium matter? Sure if you need to keep sending in laser pulses you can't exactly call it an equilibrium, but the original definition had periodic behaviour in the ground-state, which you'd think would be considered an an equilibrium.


It's not a thermodynamic equilibrium. The system does not have maximum entropy, and so, eventually the spin crystal will decay. The problem is eventually might be larger than the age of the universe by a factor of some really really large number


>"Wouldn't it be super weird if you jiggled the Jell-O and found that somehow it responded at a different period?" Yao said. "But that is the essence of the time crystal. You have some periodic driver that has a period 'T', but the system somehow synchronizes so that you observe the system oscillating with a period that is larger than 'T'."

I don't have anywhere close the knowledge and background that these physicists have but no that wouldn't be very odd if you observed oscillations that were something other than T. In music, this is called a harmonic. If you push and pull on a garage door with a period of T or a frequency of 1/T the door will product oscillations at frequencies other than 1/T


Harmonics only increase the frequency though, right? Not decrease? You can't get an octave lower note from plucking a guitar string in a certain way.


Harmonics (overtones) are already there. By definition the fundamental tone is just the lowest frequency. You can isolate harmonics either by amplifying the desired frequency or by muting a longer wave length e.g. by placing a finger on a string in a place that is a node of the higher frequency harmonics but not a node in the lower frequency. Thus by definition you cannot create subharmonics, well unless you cheat :-) http://www.marikimura.com/subharmonics.html


Harmonics are caused by the different modes that any entity vibrates at (physical, electrical, or magnetic). If you strike and object with an impulse a rich array of vibrations will occur (think of a gong, a snare, a garage door) all of those modes will be present if you repeatedly strike or oscillate the object.


This sounds like science fiction, which really tickles me in a positive way (as in, I am not criticizing it but rather it's highly exciting yet very fantastical compared to where physics was at when I studied it). I'd almost see this in an episode of Steven Universe[0], after the creators got stoned with a physicist one afternoon

[0] https://en.wikipedia.org/wiki/Steven_Universe


Great to see links to the papers!


Eurekalert is basically links to papers with some text submitted by the author and/or publisher PR person.


So.. what new kinds of electronics do we get..?


Flux capacitors?



innaccurate quartz crystals


I know the answer is probably going to be "What the hell can't they be used for?!!?" but what the hell can they be used for? Is that maybe a question for later?

All that's coming to mind is permanent storage with the emphasis on permanent.


One thing that comes to mind is reversible computing. (https://en.wikipedia.org/wiki/Reversible_computing). When computation is irreversible (as is the case for all our current information elaboration systems and substrates), it is bounded below in the energy that dissipates. The bound is quantified by the Landauer principle, and has to do with the entropy involved in "forgetting" the previous states. If a computation process is reversible, it doesn't incur in such limit. One can maybe imagine time crystals concocted to carry out useful computation, spending no energy in the process.


Tommaso Toffoli [1] invented the reversible Toffoli Gate [2].

Norman Margolus [3] invented the Margolus Neighborhood [4], which is useful for rotationally symmetrical cellular automata rules [5] like billiard ball cellular automata [6] [7].

Toffoli and Margolus also explored other energy conserving cellular automata like spin glasses [8], which are disordered magnets that store energy in the bonds between atoms.

Edward Fredkin [9] invented reversible second order cellular automata [10], which look back two steps in time, and are useful for simulating the Ising model of ferromagnetism.

Then again, maybe Otis Eugene "Gene" Ray, the "wisest man on earth", caused this article to quantum tunnel through time from April 1 1997 [11] [12].

To play with the following code that implements a spin glass, go here: http://donhopkins.com/home/CAM6/ then click the square in the upper left, click "Rules", pick the rule "von Neumann Spins Only", and draw in the cells by dragging around with the left button. JavaScript sure cooks these days!

    // ruleFunction_VonNeumann_spinsOnly computes the Spins Only rule
    // for VonNeumann neighborhood lookup table.
    //
    // Cellular Automata Machines, p. 190, section 17.3, Spins Only.
    //
    // This models a spin glass, which is a matrix of atoms with
    // magnetic spins (up or down).
    //
    // https://en.wikipedia.org/wiki/Spin_glass
    //
    // A spin glass is a disordered magnet with frustrated
    // interactions, augmented by stochastic positions of the spins,
    // where conflicting interactions, namely both ferromagnetic and
    // also antiferromagnetic bonds, are randomly distributed with
    // comparable frequency. The term "glass" comes from an analogy
    // between the magnetic disorder in a spin glass and the
    // positional disorder of a conventional, chemical glass, e.g.,
    // a window glass.
    //
    // Spin glasses display many metastable structures, leading to a
    // plenitude of time scales which are difficult to explore
    // experimentally or in simulations.
    //
    function ruleFunction_VonNeumann_spinsOnly(ruleDict, state) {

        // This makes a checkerboard pattern that alternates every
        // step, so we can apply the rule to every other cell every
        // other step. That way we know our four neighbors will not be
        // changing at the same time we are changing.
        var activeSite =
            (state.horiz ^ state.phaseTime) == state.vert;

        // Count how many of our four neighbors are set.
        var sum4 =
                 state.n0 +
            state.w0 + state.e0 +
                 state.s0;

        // When it is our turn to run in this cell (at every other
        // step), then we flip our value if exactly two of our
        // neighbors are up, and two are down. Since energy is stored
        // in two adjacent cells with different spins, we can flip our
        // value without changing the energy of the system, because
        // the perimeter between up and down cells remains the same.
        var result =
            (activeSite
                ? [
                    state.c0,
                    state.c0,
                    state.c0 ^ 1,
                    state.c0,
                    state.c0
                  ][sum4]
               : state.c0);

        return result;
    }
[1] https://en.wikipedia.org/wiki/Tommaso_Toffoli

[2] https://en.wikipedia.org/wiki/Toffoli_gate

[3] https://en.wikipedia.org/wiki/Norman_Margolus

[4] https://github.com/SimHacker/CAM6/blob/master/javascript/CAM...

[5] https://github.com/SimHacker/CAM6/blob/master/javascript/CAM...

[6] https://en.wikipedia.org/wiki/Reversible_cellular_automaton#...

[7] https://en.wikipedia.org/wiki/Billiard-ball_computer

[8] https://en.wikipedia.org/wiki/Spin_glass

[9] https://en.wikipedia.org/wiki/Edward_Fredkin

[10] https://en.wikipedia.org/wiki/Second-order_cellular_automato...

[11] https://en.wikipedia.org/wiki/Time_Cube

[12] https://web.archive.org/web/19980629180418/http://www.timecu...


Twelve citations in a HN comment has to be some kind of record.


I was only getting started! There are so many much more interesting cellular automata than Life, which is so overrated. As Marvin the Paranoid Android says, "Life? Don't talk to me about life! Loathe it or ignore it. You can't like it."

The classic book on the subject is "Cellular Automata Machines: A New Environment for Modeling" [1]

This article has a lot of great references: "When–and how–can a cellular automaton be rewritten as a lattice gas?" [2].

And Fredkin's "Digital Mechanics: An Informational Process Based on Reversible Universal Cellular Automata" [3] is also a classic paper about the subject.

Fredkin, Toffoli, Margolus and others have done a lot of interesting research into reversible computation, which has many practical and theological [4] applications, and will some day be very useful at the Restaurant at the End of the Universe [5].

[1] https://mitpress.mit.edu/books/cellular-automata-machines

[2] http://www.sciencedirect.com/science/article/pii/S0304397508...

[3] https://books.google.com/books?id=6o6zx5MRRcQC&pg=PA254&lpg=...

[4] http://www.tandfonline.com/doi/pdf/10.1080/15665399.2010.108...

[5] https://en.wikipedia.org/wiki/Omega_Point


If the 4th dimension structure is stable it seems like they'd make perfect clocks.


Could open up a new area of art, depending on how complex these crystals can get.


I wonder if (in a long long time) 4D memory storage


so, something like a hard drive that gives you correct data once every 2 ms ? :)) That sound like a lot of fun :))


Are these crystals cubic? Like some sort of time cube?


Since when is an arxiv.org paper and a physics letter mean that significant new piece of physics is confirmed?


FYI, 'Physical Review Letters' is probably the most prestigious physics-focused journal one can publish in. Here's a link the the actual PRL paper. http://journals.aps.org/prl/abstract/10.1103/PhysRevLett.118...

Back when I was in physics grad school, getting a paper in Nature or Science was the dream. But a paper in PRL was also considered a big deal.


From the article.

> While we're waiting for the papers to be published, we need to be skeptical about the two claims. But the fact that two separate teams have used the same blueprint to make time crystals out of vastly different systems is promising.


"time crystals"

Am I the only one who thinks science is starting to sound more and more like magic?


The name is whimsical. A "crystal" is a repeated pattern in space. So a "time crystal" is a pattern repeated in time.


Oh. That makes a lot more sense. Thanks


Next we find out Earth really has 4 corner simultaneous 4-day time cube within single rotation.


Gene?


The Wisest Man on Earth!


It made me think of Iron Council, in which "golems" are constructed of more and more esoteric phenomena (wood and stone are "easy", but then they're made from fire and moonlight [during the daytime, apparently it required a deep shaft and the new moon passing over at just the right time]), until at the climax a "time-golem" is created.


Sounds more like acid, IMO.


Oh, god. What is time exactly, I forgot?


Crystals?


Sure.

But according to some crummy philosophy, time as phenomena does not exist, because it requires an observer.

The simplest example would be how life (at the level of molecular biology, proteins) uses only pattern-matching but no counting, binding-sites and concentration-based regulations, but no counters. Never. Because a counting requires an observer. But molecular biology precedes any consciousness, so, no counting or numbers. Only pattern matching of physical shapes, compound structures. (Precedes here means that in a whole process of evolution of life it some patterns emerged prior to others).

Same holds for time. The notion of time, like counting, requires a consciousness. Like any other abstraction.

But, of course, who studies philosophy in the age of hipsters? So lets ask a much simpler question: what exactly replicable experiment proves the existence of time as an independent phenomena (independent of any physical process, the phenomena-in-itself), so any goddamn crystals could exist?


Are you sure? If I remember correctly, glycosylation/mannosylation of folding proteins acts as a timer giving the protein some amount of chances to fold properly before being discarded.

https://www.ncbi.nlm.nih.gov/pubmed/15950873


Nonexistence of time is a fundamental principle, which helps to prune out lots of nonsensical dogmatic hipster's pseudo-science.

It is like an ancient hack - to see things as they are one have to remove an observer. One could hack physics this way.

It has lots of philosophical implications, when one leans to recognize the fundamental difference between what is and chimeras (or memes) created by mind. Time is one such meme. Numbers is another.


Finally, the last piece for my TARDIS.


no humor please, this is HN

edit: I wonder: did I get downvoted because someone does not share my humor, as a joke, or because someone didn't get the irony?


Not sure it was recognizable as a joke. But once it was, it was damn funny!


I think you got downvoted because "no humor please, this is HN"


No DALEKs allowed though.


Last piece? I want to know where you got all the other ones.


ELI5?


From what I gather, and someone can tell me if I'm wrong, time crystals vibrate in a repeatable pattern that can be accelerated by hitting it with a laser. Scientists haven't imagined a use for this yet.


I don't understand.. I read it but it is meaningless jibberish to me


Perhaps Wikipedia can lend a hand, the description there is quite easy to follow: https://en.wikipedia.org/wiki/Space-time_crystal


Is it really meaningful though? I could call a human a space-time worm, and it makes no difference.


It is. Most matter we think about is what you would call a space-time worm. A pattern of 4d continuous worms interweaved together.

If at a given time the pattern stops, we consider that the structure broke. Matters that spontaneously break are unstable.

Materials we call impossible are only impossible under these conditions : stable a d continuous in time. The idea of matter that may be discontinuous in time but still display some characteristics of stable solid matter is pretty new.


Am I wrong but couldn't this be used to create an extremely powerful quantum supercomputer?


How do you think this would help with a quantum computer?

Not that Wilczek's work is unrelated to quantum computers, just not sure that this bit is.

Ed: For anyone wondering about the connection, Wilczek worked on anyons, which are the basis of Microsoft's research into quantum computing.


How so?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: