Physicists demonstrate existence of new subatomic structure | Hacker News Reader

Physicists demonstrate existence of new subatomic structure

las.iastate.edu

135 pointsupen36 comments

Mmm, scientific click bait titling. More accurate would be "Physicists demonstrate a theoretical subatomic structure in simulation". That's important, to be sure, but it has nothing to do with demonstrating the existence of a particle - we've determined something "should" exist based on simulation and data-interpretation plenty of times, and many of those have ended up being wrong. And that's good: good science comes from discovering what _isn't_ true, but that makes it all the more silly to have this title on the article. We've got the simulation worked out... now we need to see if it holds up to reality. It might not. If so, that's valuable information.

Apparently you didn't read the article in full before criticizing it. The article says that the simulation was used to verify real world observations from a particle accelerator experiment in Japan.

"Verify" is a strong word. The RIKEN experiment didn't come close to the usual standard of evidence for declaring a new particle exists, it just showed a reaction for which creation of a tetraneutron is a possible explanation. There need to be a lot more experiments (different ones, not just repeats of RIKEN's) before anybody would be willing to declare tetraneutrons "exist".

Simulations don't "verify" physical findings, they let us tweak the parameters under which we might encounter findings, and now we're not done:

  1. practical observation (optional)
  2. a hypothesis (either from first principles or to explain a practical observation)
  3. testing that hypothesis
  4. establish as theory if it holds up, or mark as "known to not be true" if it doesn't.

You've not scienced until you're done with step 3, because step 2 on its own cannot "confirm" or "verify" anything found in step 1.

The most important part is to confirm that a theory is not accidental, giving simulations a disadvantage: it is far easier to accidentally find parameters in a simulation than it is when you're on paper from first principles - even if that paper's Mathematica or R. You can always find some set of values, parameters, and functions that result in the thing you're looking for, so once you find such a set, you then still need to go back to the lab, verify that it holds up, and have others verify it holds up, too.

(the result of a single study is merely a very interesting anecdote. It's not data until your findings can be reproduced, and it's not a scientific theory until it's proven to hold up. Until then it's just a hypothesis)

"On their own, neutrons are very unstable and will convert into protons — positively charged subatomic particles — after ten minutes. " ... "For the tetraneutron, this lifetime is only 5×10^(-22) seconds (a tiny fraction of a billionth of a nanosecond). "

10 minutes is an eternity and 5×10^(-22) seconds is closer to what I'd consider 'very unstable'

Compared to the almost-forever you can get inside a nucleus, or from protons, it's quite unstable.

Inside a nucleus---yes; from protons, not really. Helium-2 is as unstable or possibly more unstable than the tetraneutron; lithium-3 and beryllium-4 don't exist at all.

They will break apart, but the stability of the protons themselves is enormous.

Yes, but the article is about multi-particle systems.

The discovery is. The comment about neutrons decaying in ten minutes isn't.

> The research in Japan used a beam of Helium-8, Helium with 4 extra neutrons, colliding with a regular Helium-4 atom. The collision breaks up the Helium-8 into another Helium-4 and a tetraneutron in its brief resonance state, before it, too, breaks apart, forming four lone neutrons.

I don't know how fast these particles were going in the experiment, but if they were going at the speed of light, then the tetraneutron would make it 150 femtometres from the site of the collision before decaying. By way of comparison, neutrons and protons are 1 - 2 fm across, and a uranium nucleus is 15 fm across [i]. So if the tetraneutron was actually going at a tenth of the speed of light, it would barely make it a nucleus's diameter away before exploding.

[i] https://en.wikipedia.org/wiki/Atomic_nucleus

That's not how relativistic speeds work. If this was going near enough to the speed of light, it could travel almost any distance before decaying. Time dilation would be observed of the particle, where it would appear that time is ticking slower for the particle compared to our own frame of reference.

Actually what the parent comment said would be correct. Time dilation and length contraction do not arbitrarily change the proper distances that particles travel at near light speeds.

As observers, the detectors would see precisely what we expect if we calculate the distance using the detector's proper time when compared with the distance we experimentally measure.

Please explain why you think this is so.

This is super basic stuff https://en.wikipedia.org/wiki/Time_dilation_of_moving_partic...

The linked wiki article says that rapidly moving particles take longer to decay.

Mmm... I have the strange feeling that you both agree, let's write this with more details to avoid confusion.

The half life of a static tetraneutron is (theoreticaly) 5E-22s.

If you are at a laboratory and the tetraneutron is moving at relativistic speed, the apparent half life in the laboratory frame will increase substantially.

To calculate the distance that the tetraneutron will travel you must multiply the velocity x the apparent half life. I.E

d ~= v * 1/(1-v^2/c^2) * t_hl

where t_hl = 5E-22s = the half life of the tetraneuton

not the classic version that is

d_clasic_wrong = v * t_hl

For a more detailed explanation, with numbers and graphic you all can see the analogue experiment with muons: http://hyperphysics.phy-astr.gsu.edu/hbase/relativ/muon.html

Anyway, IIRC a half life of 5E-22s is still very small. IIRC you can use a bubble chamber or something similar to see the trajectory of the particles. You can only see a collision and examine the debris and you will note that when the collision has some specific energy you will get an unexpected excess of collisions. You only can examine the direction and energy of the debris to make a good guess of the intermediate steps of the collision and discover that for a very short time you had a tetraneutron.

I do agree that time dilation occurs, what I disagreed with/corrected was with this: it could travel almost any distance before decaying, which I took to imply that the lengthening of particle lifetimes was not measurable in the way that you just pointed out.

In retrospect perhaps the parent comment was saying theoretically it could travel any distance which is true, but doesn't really help in estimating whether it could clear nuclear radii or not.

That's a good point, i didn't think of that! Do we know how fast the tetraneutrons were moving in this experiment? If so, we can work out the degree of time dilation, and how far it would have got.

You forgot the sentence just before the last one. The weird thing is that they call the 5×10^(-22) seconds "stable for a period of time", but the 10 minutes "very unstable".

>four neutrons together can form a resonance, a structure stable for a period of time before decaying.

You're adding context to "a period of time" that's not there. 1 ns, 1ms, 1 day, and 1 millennia are all "periods of time." As for stable vs unstable you have to contrast lifetimes across other particles.

Oh, I understand that; I just think that in the interest of clarity they should have added a bit more explicit context. Now it looks odd.

Physics is one science where the seemingly impossible winds up being possible all the time. Strange to think people spend a good portion of their lives studying things thought to not exist.

>Physics is one science where the seemingly impossible winds up being possible all the time.

Quite on the contrary.

Actually, physicists are desperate for an upset that will disprove the standard model. The standard model is, in many ways, too good, in the sense that it predicts most (but certainly not all) measurable phenomena to very high accuracy, but is not a complete theory. So we're looking for examples that break the standard model so we can figure out how to expand it to be complete, but we're having trouble finding useful mispredictions by the standard model. A few areas where the standard model is deficient: dark matter, matter/antimatter asymmetry, hierarchy problem.

> where the seemingly impossible winds up being possible all the time.

No, it's where the possible is often portrayed as impossible by popular reporting all the time.

We've updated the link from http://sciencebulletin.org/archives/7339.html to this, which looks like the original source.

I read the headline as "demand" and thought this was a rather petulant reaction to the "no new physics" development this year.

https://home.cern/about/updates/2016/05/theory-theoretical-p...

The original paper [1].

[1] https://arxiv.org/abs/1607.05631

Is it really "subatomic" though? Seems more like it could be considered isotope 4 of element zero.

Technically subatomic refers to the size of an atom with electrons, not a nucleus (which is much smaller). It's especially appropriate here because neutrons aren't electrically charged, so their interactions with other particles are very short range.

I don't know, maybe this discovery has scientific value. But could it be the sometimes scientists report small "breakthroughs" just so that they keep the funding coming ?

Still no turtles.

it's subatomic particles all the way down i tell you!

I would have sworn I've seen an animation implying that at sufficient magnification the whole thing loops around to galactic superclusters. Non-intuitive, to be sure...

Maybe we just need a bigger particle accelerator to discern the turtle sub-structure of current known particles.

Reply on news.ycombinator.com