Hacker News new | past | comments | ask | show | jobs | submit login
Atlas observes light scattering off light (atlas.cern)
182 points by ars on March 17, 2019 | hide | past | favorite | 36 comments



Light by light scattering was first observed at SLAC in the mid 90s by experiment E-144.

https://physicsworld.com/a/light-is-seen-to-scatter-off-ligh...

Link to paper from 1997: Positron Production in Multiphoton Light-by-Light Scattering

(published in Phys. Rev. Lett., Vol. 79, p. 1626 (1997).) http://www.slac.stanford.edu/exp/e144/ps/positron_slacpub.ps


They do try to mitigate this criticism in the first line of the article:

"Light-by-light scattering is a very rare phenomenon in which two photons – particles of light – interact, producing again a pair of photons."

This is the process that they observe (for the first time) whereas E144 observed the Breit-Wheeler process.

You can argue that this is too narrow a definition of light-by-light scattering but they have at least tried to make it self consistent...


I think this mainly centres on just what should count, here[0] is a paper from 1996 about using astrophysical gamma ray sources to detect the effect of the intergalactic background light, which is also photon photon interactions.

[0]: http://adsabs.harvard.edu/full/1996ApJ...456..124M


Thank you

Would have been nice if that info was featured somewhere in the article. I don't like to say it but this am I correct in saying that this article gives off the impression that this is an experimental first in terms of directly observing this photonic interaction?


Any potentially practical use?


Purely speculating as a physics-interested layperson, but I imagine this probably either changes or confirms/enhances confidence of various cosmological observations where photon-photon interaction might play a role. For example, maybe certain parameters/characteristics of the CMB have uncertainties potentially related to this phenomenon which can now be tightened up, which could have ramifications for inflation theory, Higgs/particle theory, string theory, etc


Sure

making real Jedi saber


Not a physics expert.

Projecting in the air without a screen could be a potential use.


Lol, no. Photon-photon scattering is very rare; you only expect it to happen in very high-energy situations. To scatter enough photons to create a visible image in the air you would need to pour so much energy into the air that it would turn into a plasma. A deadly radioactive plasma that explodes outward from your holographic display and kills everyone nearby. This would not be considered practical.


The lol isn’t needed


I think an important part of our humanity is our ability to laugh. I literally laughed when I read the question, because of the absurd impracticality of it. That's an important part of the answer.

On the other hand, it's true that a purely text-based medium like a comment doesn't have room for a lot of nuance. So for the sake of clarity, I will say that it was a good question. We should always look for ways to use new discoveries to our advantage. Let's just stick to things which are possible, like Dyson spheres, and leave the impossible for science fiction.

See also https://what-if.xkcd.com/1/.


Neither is the rouge but here we are.


If you wanted to do this kind of holography, you’d be using multiple lasers (or masers) to stimulate different excitation states of atmospheric gases - you “just” have to have them sweep fast and accurately enough to intersect at each pixel. The engineering challenge is far from insignificant, but there is some progress being made in this general direction, coming from the 3d printing world (volumetric/tomographic printing).


Like these but less plasma? https://youtu.be/AoWi10YVmfE

My favorite part of this is looking at all the toasted skin on the fingertips in the video.


Is the scattering probability a function of flux density or simply of photon energy?

If of photon energy, then the device is clearly impossible -- you would need to use invisible (and harmful) gamma-rays for scattering.

If of energy density, it still may not be theoretically feasible. For instance, you cannot make monochromatic pulses arbitrarily short -- eventually they will become spread in frequency (more and more energy in higher and potentially harmful frequencies), such that the total dosage necessary to get the desired visible flux is too large. (some calculations would need to be made...)


It's the flux. To scatter any photons takes a lot of flux, and to scatter enough to see would take quite a lot. Of course I haven't done the math either, so I'm just guessing what it would do to the atmosphere that got in the way.


If the electric field from a concentrated laser is strong enough it can spark the air and make a voxel.


Yes, this is much more practical. Short pulses to create a tiny ball of plasma that doesn't expand very much, but which can emit enough light to see. Still not what we want (which is Star Trek), but probably quite fun as long as the total volume of plasma stays small enough to be safe.


If you want Star Trek ( and so do I, of course :) ) then the simplest way I can imagine has to be with headsets as currently is.

Making them imperceptible, if that is even so important, has some challenges but not impossible (e.g. increasing resolution, lowering weight, or even adding devices to compensate its weight if the former is infeasible).


I know a couple rouge nation-states that might be interested


Rouge? Communist?

“Better dead than rouge?” Hm.


That still seems impractical, at least until we miniaturize a particle accelerator. The event is so improbable (two photons banging into each other) that you need a super dense cloud of them to get a measurable effect.


Light by light scattering is also why we can't see gamma rays above 80 TeV or so from far away (the gamma rays would pair produce e+/e- pairs with CMB photons).


How do we detect high-energy cosmic rays, which to my knowledge can exceed this "limit" by several orders of magnitude?


They're not gamma rays :)

But a similar limit happens to protons (the GZK limit at 50 EeV, where protons and CMB photons can make a Delta+). For heavier nuclei, photodisintegration will happen at even higher energies.


If you are talking about [1], I think nobody knows how they are possible.

https://en.wikipedia.org/wiki/Oh-My-God_particle


That article links to a much better written one that appears to discuss this exact issue: https://en.wikipedia.org/wiki/Greisen–Zatsepin–Kuzmin_limit. Thanks!


Those rays are unlikely to be photons, so EM-self interaction is not likely to stop them. The articles linked in this sub-thread all speculate/assume that the cosmic rays were protons.


Dumb question but is this unexpected because photons are bosons?


This is not unexpected, QED (quantum electrodynamics) predicts this. Classically it does not happen because classical electromagnetic fields are in linear superposition. QED has loop diagrams involving virtual fermions that mediate scattering, this is also the reason why these events are rare.


I agree. More details:

The main problem is that photons only can interact with charged particles, but photons have no charge, so the photons don't interact (directly) with photons.

The gluons that are the bosons of the strong force can interact between them, so it's not a problem of bosons vs fermions. In particular, gluons have color that is the equivalent of charge for the strong force.

The idea is that the photons can sometimes create a virtual electron-positron pair that is very short lived [1], and the other photon can colide with the electron or positron before they annihilate. This is very rare, and you can ignore it unless you have a really huge amount of photons flying around.

[1] Insert here a technical remark about "virtual" and "very short lived". It's more complicated. Take this as a metaphor to hide a lot of math.


Quantum physics of this type has always intrigued me, but I've never been able to find a good resource to properly start learning it.

Could anyone point me to some good starting material? Thanks :)


If you have two photons that have been emitted at different times, and no time passes at the frame of reference of the photon when they travel, what time is it when they interact?


> what time is it when they interact?

In which frame of reference?


I think that's the detail underlying my question. Does the interaction collapse the frames of reference into one? I guess it doesn't have to.


The time of collision is not a problem, it happens at a different time depending on which photon's frame of reference we're talking about.

Which frame of reference the resulting particles might "inherit", well, that's a very interesting question that I've never thought about!




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

Search: