Fig. 1. The conference ICHEP 2016 conference audience fully scored, and stand in the doorway waiting for news about the two-photon peak. Image from the site twitter.com
The conference ICHEP 2016 published new results of the LHC on the mysterious two-photon peak at a mass of 750 GeV, hints of which were six months ago. Now, based on four times larger statistics, collaboration ATLAS and CMS have taken out an unambiguous verdict: in the new data is no hint of this peak is not. What excites the theoretical physicists of recent months, it was a statistical fluctuation.
The end of the tale
The two-photon surge at a mass of 750 GeV, which put on the ears of the scientific community in the physics of elementary particles and became the subject for more than 500 theoretical papers – turned out to be a mirage, a fairy tale. The beautiful, believable, thrilling fantasy – but a fairy tale. She left us with strong feelings and warm memories, but we need to move on.
On the morning of 5 August passing these days conference ICHEP 2016, the main event of the year in elementary particle physics, the representatives of the ATLAS and CMS have shown the long-awaited new the results of the Large hadron collider on the creation of two high-energy photons. Last year when the collider was just beginning his second season after a long pause, the statistics has been set up is very modest, it is consistent with the luminosity of a little more than 3 fb-1. This year the collider came to a record rate of the data set, so that the results presented at the conference for the year 2016 were prepared on the luminosity of about 12 fb-1. Last session dataset caught in this sample was on the night of 14 July 15,
In March, the deviation from the Standard Model, it seemed very important -. Look at Fig. 2 and 3 of the March news – and have reason to hope that the five-fold increase in the statistics lead to the guaranteed opening. Provided, of course, that the observed deviation then -. Actually, it’s not a statistical fluctuation
But the new data destroyed these hopes. Fig. 2 shows the same schedule that the last news, but only with the new statistics. From spike at 750 GeV was gone. Moreover, the statistics exactly evil whim at this value there is even a shortage of some events, but not exceeding. In the whole area of the invariant mass of up to 2500 GeV, as there is no significant deviation from the Standard Model background (color curves in Fig. 2). Fluctuations up and down in the data are present, but they look exactly the way from normal statistical fluctuations and expect. Even some semblance of spike in the CMS database (. Figure 2, right) in the 600-650 GeV is not drawn to any significant deviation – the more so because it is not supported in the ATLAS experiment
<. img src = "https://readweb.org/upkeep/uploads/2016/08/550c5aa94a1a5d007784ce85841c4def.jpg" />
Fig. 2. two-photon events invariant mass distribution in the ATLAS data ( left ) and CMS ( right ). Colored curves show the best simulation of the background; below shows the data net of the curve in absolute terms (for ATLAS) and in terms of statistical error (for the CMS). Images of the discussed reports
Comparison of accumulated data with the background is best shown in Fig. 3 ATLAS experiment example for moderately wide scalar resonance; CMS have observed a similar picture. Here are the local p are the values, then it is likely that only one background processes of the Standard Model, with no new physics may be due to statistical attacks to give the masses at a given value, this (or even more) deviation, which saw in physics experiment. When p is the value drops to very small quantities of the order of one-thousandth or less, it gives reason to suspect physicists background that one can not do here, that perhaps there is a new effect.
Fig. 3. Local Schedule p are the values for different invariant mass of two photons in the ATLAS data. red and blue shows the color data 2016 and 2015; black curve – aggregate result. Image source atlas.web.cern.ch
The data in 2015 and it was: a blue dashed line at a value of 750 GeV plummets down. However, the data in 2016 (red dotted line) nothing like this occurs. Even if the two samples to combine, the total local effect falls to 2σ, and then only by the old data. From the perspective of the experiment – it is nothing; fluctuations of this magnitude will inevitably occur in any way large statistics.
In order to really dot the «i», CMS collaboration showed how should look like a signal of a new particle, if it were real and its mass and cross section would be such, which pointed out the March analysis. It is shown in Fig. 4 in the area of the invariant mass of 600-900 GeV. The peak would be standing proudly over a background of nearly 4 standard deviations. Instead – failure
Fig. 4. The same data as in Fig. 2, to the right, in the 600-900 GeV. red shows how should look like a signal, at which point the analysis of the March. Image source indico.cern.ch
Conclusion unambiguous and inexorable: two-photon spike at 750 GeV is closed . What we have seen in last year’s data, was only a statistical fluctuation. The new findings are consistent with the expectations of the Standard Model.
Negative search resonances, as in such cases is allowed physicists to set upper limit on the cross section of a hypothetical particle. Fig. 5 shows the ATLAS data limitations. Almost the entire mass range limits have increased by 2-4 times compared with the old data, and in particular, they have closed the cross section at a mass of 750 GeV, which was estimated in March (5-10 fb).
Fig. 5. The upper limit on the cross section of a hypothetical severe resonance two-photon decay. red and blue shows the color data 2016 and 2015; black curve – aggregate result. Image source atlas.web.cern.ch
Issues and Lessons
brewing was a sensation disappeared as lightning, as it came. But she left behind a question, in a sense, coming out even from a purely scientific plane: What was that? How is the result? ?
Are expectations were justifiedFirst of all, we emphasize once again that neither in the experiment, even more so in the behavior of experimenters no puncture there. Accusations that the participants in the experiments allegedly wishful thinking – are groundless. The approach to the analysis of the data was weighed and impartial analysis of the results reported in the most conservative terms, and in communicating with the media theorists and experimenters stressed that it is too early to draw conclusions, we need new data. The strongest statement that the collaboration officially issued in March, after the new results confirm sounded « Now we can feel cautiously optimistic ». The word “discovery” of the experimenters did not tell even close. And it is very different from the situation with superluminal neutrinos, when all the hype has arisen due to a defective cable, and by hasty posts BICEP2 of registration of primary gravitational waves in the CMB polarization – it turned out that was not correctly assessed the galactic background
<. p> Here, apparently, with the experiment, everything was in order from beginning to end. The difference in the results could be attributed to the fact that in the last analysis of the physics of something not taken into account, and now taken into account. But the method of analysis, in fact, remained the same, only now it is a new dataset has been processed (and again, by the way, revalidated old). It turns out that this data is played with physicists cruel joke – and soon in both experiments. The coincidence of strange, no doubt, but no other explanations from representatives of collaboration is not (Fig. 6). They even have estimated the probability that the same physical situation may give different results so in two successive data samples. The probability of getting at least one percent (2,7σ, ATLAS) and a couple of percent (2,4σ, CMS). The probability of such a joke of nature right in both experiments was not shown.
Fig. 6. The representative of the ATLAS collaboration Dave Charlton (David Charlton), responding during a press conference on the question of the journalist, agrees that the match is so close to statistical fluctuations in the two experiments looks amazing, but another explanation, he does not see. Screenshot of video broadcasting
The enormous – and, as it turned out, groundless – the expectations are, rather, on the conscience of theorists, including many people with big names that are on the first day threw remarkable strength to the full discussion of deviations in terms of new physics and thereby lead by example for many others. Source ATLAS and CMS analysis showed deviation level (unofficially combined) statistical significance of about 3σ. For this, in general, a modest deviation of the response of physicists was unduly optimistic. Now, of course, there is a backlash, including not only a joke, but also direct accusations among other theorists. Therefore, if talk about the “loss of contact” is not between physicists and the press, but rather between experimentalists and theorists. It is possible, even, that the outcome of this story will be reviewed, how much weight should be given to deviations in the 4-5 Sigma, especially when they occur in a previously unexplored area.
In this story there are, however, some positive points . First, the effort of thousands (!) Of theorists will not be lost as a minimum because there were universal phenomenological developments in the case of new unexpected spike in only one of the channels without confirmation in the other. Secondly, well, that the anomaly has disappeared quickly and unconditionally. It would be much worse, if such a spike showed up at the turn of the first session of the collider in 2012: then it would be three years Physics amused ourselves with illusions. Equally unpleasant situation would have been if the new data did not close the deflection, and kept him on the edge of the curious. Physicists have recently passed through the situation with painful uncertainty, when the Higgs boson decay into two photons (two-photon channel again!) For a long time seemed unusual.
In general, the scientific community in elementary particle physics wakes up and returns to the weighted , justified the theoretical interpretation of the data. Fortunately, most of these data will now be a lot. Collider is working properly, statistics even triple by the end of this year – and then grow back in several times to the end of the session Run 2. Finally, the 2020s will be pleased with another approximately ten-fold increase in the statistics compared to the whole sample Run 2. This program has already been approved by the CERN and it will be implemented regardless of the developments in the coming years. So far, the accumulated less than percent of the total amount of data, which is calculated the LHC, and in anticipation of new horizons, this percentage should be to clearly and carefully consider.
Links
ATLAS Collaboration:
Search for a high mass diphoton resonance using the ATLAS detector // report to the conference ICHEP 2016;
Search for scalar diphoton resonances with 15.4 fb-1 of data collected at sqrt (s) = 13 TeV in 2015 and 2016 with the ATLAS detector // Advance publication ATLAS-CONF-2016-059;
Diphoton Searches at ATLAS // seminar at CERN (PDF, 29 Mb);
High-mass di-photon resonances: the first in 2016 ATLAS results // note on the ATLAS website.
CMS Collaboration:
Searches for BSM physics in diphoton final state at CMS // report to the conference ICHEP 2016;
Search for resonant production of high mass photon pairs using 12.9 fb-1 of proton-proton collisions at sqrt (s) = 13 TeV and combined interpretation of searches at 8 and 13 TeV // Advance publication CMS-PAS-EXO-16-027;
Diphoton Searches at CMS // seminar at CERN (PDF, 21 Mb).
Igor Ivanov
No comments:
Post a Comment