Atlas Sees No Light From Quantum Black Holes
    By Tommaso Dorigo | February 2nd 2014 10:46 AM | 13 comments | Print | E-mail | Track Comments
    About Tommaso

    I am an experimental particle physicist working with the CMS experiment at CERN. In my spare time I play chess, abuse the piano, and aim my dobson...

    View Tommaso's Profile
    Black holes are all the rage these days, with theorists arguing about firewalls and Hawking's paper being handled by the press in rather improper ways. Meanwhile at the Large Hadron Collider ATLAS and CMS are furthering their searches for microscopic versions of the same objects, which could exist if the energy scale at which quantum gravity effects make themselves felt is orders of magnitude smaller than the place where they ought to be -i.e., at the Planck energy.

    There is a wide variety of phenomena that could signal the production of black holes in the proton-proton collisions of the LHC. The one considered by the latest article by ATLAS is just one of them -the production of photon-jet pairs of high combined mass, which could result from the decay of a black hole that would not have time to thermalize. But in truth ATLAS does the opposite: they search for gamma-jet events of high mass, without specifically focusing on black holes. They thus produce aspecific limits on the rate of processes that would yield that signature, and that is, I think, a very good thing.

    Indeed, exclusive searches for very specific final states predicted by particular new physics models are useful but a bit narrow-minded. With the large amount of data now available in ATLAS and CMS, and with the failure of fashionable models of new physics to provide predictions that stand the experimental test, it is quite reasonable to study the data in a more inclusive fashion, concentrating on rather generic signatures and determining if the corresponding data agree with the Standard Model background or if they depart from it.

    ATLAS considered their while 8-TeV data sample and produced a spectrum of the invariant mass of energetic photons recoiling against a jet. The photon signature at high energy is quite clean, so backgrounds are mainly due to the compton-like scattering process of a quark-gluon collision yielding a quark and a photon; a second background is due to hadronic jets fragmenting in such a way that the leading particle is a photon. These are not "direct" photons, in the sense that they are not the direct result of the hard scattering process, but are secondary by-products of the final state partons.

    Despite the above distinction, the analysis is performed in such a way that the background processes are estimated from the data themselves - a smooth distribution of invariant mass is their collective result, so one may search that smooth spectrum for a Gaussian bump signalling the presence of a resonance or other turn-on phenomena.

    The mass spectrum obtained by ATLAS is shown on the right. As you can see, this is indeed quite smooth and well interpreted by the employed functional form. The presence of a resonance - such as could be an excited quark, q*, deexcitating to a quark by emitting electromagnetic radiation, or a quantum black hole or other fancy new things - would produce enhancements such as the ones shown by dashed histograms in the graph. Also note the lower portion of the graph, where ATLAS incorrectly labels the y axis as a "significance" (it is not, technically, but the discussion of why that is so is for elsewhere). The red bars still give a feeling of the compatibility of the data with the chosen parametrization, which would be otherwise hard to make out from the semi-logarithmic distribution on top, especially at the lowest mass end.

    Using  by now well-tested technology of limit setting with the CLs criterion, ATLAS proceeds to interpret the data in terms of an exclusion of the cross section for quantum black holes and excited quarks as a function of the mass scale of those processes (see e.g. the figure below). However, they also produce a limit for a generic narrow resonance as a function of its mass, so that a future model can compare to the ATLAS result without the need for a new analysis.

    The graph on the left shows the usual "Brazil band" plot, where for any hypothetical quantum black hole treshold energy (for parameters n=6 and MD=Mth, but these are details which can only interest those who will read the paper anyway) the upper limit at 95% confidence level is drawn on the product of cross section times branching fraction to photon-jet pairs times acceptance. The black points show the result, while the band as usual describes the pre-data expectation for the limit, given data size, detector efficiency and analysis technique. The dashed blue line shows that models with a energy scale below 4.55 TeV are excluded in this particular scenario.

    All in all a nice analysis, and I must say I wish I saw more of these signature-based searches. Specifically, I would like to see searches that consider the data even more inclusively, and multivariate techniques that take care of evidencing region of multi-dimensional parameter space of the observable quantities where the data disagree with background expectations. But such searches are still viewed with suspicion in the HEP community, partly because the mindset is still aligned with that of our now aging mentors, who learned how to hunt for new physics when computers still munched hole-punched data cards. Those new ideas will take time to win over the scientific consensus.


    As to "... searches that consider the data even more inclusively, and multivariate techniques that take care of evidencing region of multi-dimensional parameter space of the observable quantities where the data disagree with background expectations ..."

    how is that related to what Narsky and Porter describe as "... In transductive learning, we optimize the predictive power for the OBSERVED unlabeled data. We thus do not care about classifying data the COULD BE observed ... If transductive learning were to be applied to physics data today, it would likely raise strong objections from some physicists, especially those who religiously practice blind analysis ...".

    The microscopic quantum black holes (stabilized with extradimensions) are actually the common atom nuclei, which are produced at LHC routinely.

    John Duffield
    Zephir: no they aren't.

    Tommaso: I don't like to see talk of black holes at CERN. I think it's irresponsible. The sort of thing that could get the LHC shut down. Hawking radiation remains a hypothesis. Try explaining it in a way that makes sense. You can't. And nor can you rely on Hawking radiation to make a micro black hole decay.

    As for "Hawking's paper being handled by the press in rather improper ways", oh come on. Hawking works the press. When he said "The absence of event horizons mean that there are no black holes", that was a media hook. A soundbite. Deliberately planted to get him in the headlines. That's what these celebrity "physicists" do.
    You keep saying this but it isn't true. Hawking is well-known enough he doesn't need to stick that in an arXiv transcription of a year old talk, he could just call up NBC and they would have run it if he had intended to have it used that way. He said matter collapse doesn't lead to an eternal event horizon but rather an apparent one - it isn't new, it's been said for two decades. Sabine at Backreaction has a whole history of it. Ethan at Starts With A Bang clears up some of your confusion about Hawking radiation.

    What did happen was that a writer at Nature saw what Hawking wrote and misunderstood it just like you do. But he put "Stephen Hawking: 'there are no black holes' in his title and it snowballed into mainstream media from there. So an editor and a journalist are to blame, not Hawking, or Hawking would have caused this a year ago when he gave his talk.
    John Duffield
    I'm not confused about Hawking radiation. Ethan doesn't clear anything up. He recounts the "given" explanation, which confuses curved spacetime with curved space, includes the popscience myth that particles pop in and out of existence, and repeats the old adage that one of the pair falls in while one escapes. Read Matt Strassler's blog on virtual particles. They aren't particles. And nor are they vacuum fluctuations. And do you know of any negative energy particles? No. Now read Sabine's comment: "There is a reason I never use the analogy with the pairs created in the vicinity of the horizon. That's because it is extremely misleading". The "given" explanation for Hawking radiation is trash. Don't kid yourself that you understand it. And don't kid yourself that the soundbite isn't deliberate either. When you've been talking black holes for decades, you just don't write black holes don't exist. And as for calling up NBC, Hawking has caught flak before for bypassing the system and going straight to the media, see this.
    Layman's explanations are useful. Visualizations and toy models are helpful, which is why textbooks and even scientific articles have a lot of prose in them. But yes, layman's explanations can be misleading. That is why it is important to actually learn the theory and be able to work with it at the precise mathematical level. Then the simplified explanations become more useful as one understands the limitations of that view point.

    You however refuse to learn anything beyond layman's explanations, and even worse, are incredibly biased towards those explanations that you believe bolster your strange beliefs. Without any depth of knowledge, arguing by layman's descriptions is just silly and wasteful. So learn the theory before criticising and providing your "take". For instance, we understand quantum field theory in flat spacetime very well, and even in flat spacetime, Rindler horizons radiate very analogous to Hawking radiation. This isn't some extra postulated effect, this is a derived result of quantum field theory.

    "...confuses curved spacetime with curved space..."

    Please STOP it with this stupid talking point of yours. If your "interpretation" of GR is mathematically equivalent with regard to predictions for experiments, then empirically there can be no difference, so stop complaining. If however your "interpretation" gives different predictions, then IT IS A DIFFERENT THEORY. You have disagreed with the predictions of GR many times so far, and complain when people use math to show you that you are wrong, so it is clear you don't actually understand GR and are pushing a different theory. Your continued complaints basically amount to complaining that scientists don't use your crackpot version of gravity instead of GR. And no, your random quotes from scientists don't back your claims that Einstein agrees with you, because he specified his theory precisely with mathematics, and anyone can check the predictions and see that you are wrong.

    Stop pushing your crackpot gravity theory all over science20. Keep it to your blog.

    F*** that John. Do you think we should not produce papers about black hole searches or explain what we do because somebody could be scared ? We have to educate the people, not follow their ignorance and be conditioned by it.

    John Duffield
    I think you should explain what you do. But I also think you should take great care about what you say about black holes at CERN. Because Hawking radiation remains hypothetical. We've got Hawking saying black holes don't exist, and doubtless luxuriating in the media storm that gave him. What if there's a media storm saying Hawking radiation doesn't exist? If I was Rolf-Dieter Heuer I'd be very concerned about this issue. Don't get me wrong, I'm utterly confident that black holes cannot be created in a collider. For my own reasons. That's not the issue. The issue is the public perception of physicists who think they can. And who think Hawking radiation would save everybody's bacon.
    /* Zephir: no they aren't. Tommaso: I don't like to see talk of black holes at CERN. I think it's irresponsible. */

    LOL, why not? It's irresponsible - but not because it's actually dangerous, but because the scientists don't know, that it cannot be dangerous (they do believe in stable black holes instead). If you're trying to explode the munition in your kitchen, it's irresponsible despite this munition is already deactivated, until you don't know about it.

    One can also learn from trying to understand the continuum as you mentioned.

    I was under the understanding that evidence of black holes production at LHC (or lack thereof) would give support (or not) to higher-dimensions theories like SuSy or String theory. Is that correct? Do the observations you report about tell us anything about the likeliness of spatial extra-dimensions?

    Hi Filippo,

    no, the idea that the scale of quantum gravity is at the TeV or so is a pretextuous one, and disproving it has no impact on SUSY or String Theory. There is a rather wide range of values
    from the TeV to the Planck scale to investigate...

    Thanks a lot. I was probably misinterpreting articles such as

    Keep up the good work!