The CMS collaboration has completed a more than a year search for new Higgs bosons that decay into two photons. Her article confirms preliminary data from last year. In particular, at an energy of about 95 GeV, a deviation from the Standard Model is still noticeable. Curiously, it was in this area that the LEP collider also saw a faint hint of deviation 15 years ago. This coincidence does not leave theorists indifferent; they are already discussing which versions of New Physics are capable of accommodating both deviations.
There is no doubt that there must be phenomena in the world of elementary particles that do not fit into the Standard Model. The search for such phenomena, which are collectively called “New Physics”, is the main task of the Large Hadron Collider. Many physicists hope that the “Higgs” sector of the microworld will become the first swallows of the New Physics. In 2012, the CMS and ATLAS collaborations discovered the long-awaited Higgs boson and thus launched a long-term program to study all of its properties. So far, all the measured properties of this particle are in good agreement with the expectations of the Standard Model. However, it may turn out that the Higgs sector is much richer and that there are additional Higgs bosons in our world, only for some reason they have so far escaped detection.
There are no miracles or artificial assumptions here. In many theories with an extended Higgs sector, it naturally turns out that one Higgs boson is standard and therefore visible well, but the rest, on the contrary, very reluctantly interact with ordinary particles and therefore it is difficult to identify them in experiment. Nevertheless, experimenters are intensively searching for new Higgs bosons, both heavier and lighter than the familiar 125 GeV Higgs boson.
At the end of November, in the preprint arXiv: 1811.08459, the CMS collaboration published the results of the next search for additional Higgs bosons in the Run 1 data and in the first part of Run 2 (data for 2016, integrated luminosity 36 fb−1). Physicists were looking for new bosons with masses from 70 to 110 GeV, decaying into two photons. Although the probability of a new particle decaying into two photons may be small, the two-photon channel is convenient for studying, and therefore there is a chance to see a new particle in it. We recall that the Higgs boson with a mass of 125 GeV was just discovered in the two-photon and 4-lepton decay channels.
In fig. 1 shows the number of selected events from the 2016 statistics depending on the invariant mass of two photons. There were many such events – hundreds of thousands. The Standard Model predicts that they should occur quite often due to the usual process of independent emission of photons in a hard proton collision. Their distribution over the invariant mass should have the form of a fairly smooth, but rapidly decreasing function, which is confirmed by the data obtained. New particles, if present in this mass range and capable of decaying into two photons, will show up as bursts on this smooth graph. In order for such deviations to be considered, in the same Fig. 1, below, the difference between the data and the general smooth curve is shown. Among the statistical fluctuations on it, a suspicious upward deviation of several experimental points at once in the vicinity of 95 GeV is clearly visible. This could, of course, have happened by chance, but optimistic theorists can see here the first manifestation of a new particle, maybe even a new Higgs boson.
It should be emphasized that the CMS collaboration did not make any loud statements. Yet this deviation, no matter how beautiful it may look to the theoretician, has too little statistical significance to classify it even as an indication of the existence of a new particle. In addition to this, the smooth background curve, which was discussed above, is not at all fixed theoretically, but is built on the basis of the same data. Depending on which function to choose for this smooth curve, the deviation is different (the mark Class 1 in Fig. 1 just refers to the shape of the curve). On the other hand, approximately the same deviation near 95 GeV can be seen in the Run 1 data, albeit weaker.
General analysis of the CMS showed that the local statistical significance of the deviation was 2.8σ. This is especially clearly seen in Fig. 2, which shows an upper bound on the cross section for the production and two-photon decay of a new hypothetical particle.
Let us remind you how to “read” this graph. When experimenters are looking for new particles, but do not see their manifestations, they set an upper bound on the probability of their appearance in the process under study (in this case, in the process of two photons production). This limit is shown at the top with a black line. In addition, experimenters carry out numerical simulations of the process and estimate how this limitation should have turned out in the framework of the pure Standard Model, without any new particles. It is shown with a dotted line. Of course, statistical fluctuations are superimposed on this: after all, in a real experiment, the number of events may slightly differ from the expected one. Therefore, a small deviation is not yet a reason for increased attention. The green and yellow stripes show the area on the graph through which the curve can pass in a real experiment just like that, due to fluctuations. This happens almost everywhere on the graph – with the exception of the region around 95 GeV. There the black curve crawls out far beyond the color band. This means that the data in this area do not allow the experimenter to set the upper limit that they hoped to establish: the data oppose it. This happens when the data contains “extra” events that the Standard Model does not take into account. The discovery of new particles begins with such a “creep of the graph”.
Experience has taught physicists to be very critical of such deviations, especially in the case when it is not known in advance where the new particle may be. Therefore, for a more balanced assessment, one should look not at the local, but at the global statistical significance. In this case, it is very modest 1.3σ, but the experimenters considered it necessary to mention this in their work.
And this is where … LEP, CERN’s previous flagship collider, comes into play. At the turn of the century, he made the finishing spurt and tried to find the Higgs boson at the maximum energy then available. The dash, alas, was unsuccessful, and in 2003 all four LEP experiments published a joint legacy article Search for the Standard Model Higgs Boson at LEP with the results of this “boson hunt.” Among other things, there is also an interesting graph shown in Fig. 3. Here is shown the same upper bound on the Higgs boson production and its decay into b-anti-b pair normalized to the prediction of the Standard Model. The entire mass region where this curve falls below unity (up to 115 GeV) is a “closed” region, there is no Higgs boson. As we now know, this is not surprising, because our “native” boson is at a mass of 125 GeV.
However, the same graph can be read in another way – as a restriction on additional Higgs bosons, which do not need to correspond to the boundary of the Standard Model. And here the region near 95 GeV also stands out from the general background. Here, too, there is a difference – albeit weak and in itself statistically insignificant, which is why special attention was not paid to it until recently. However, when the CMS collaboration began to see something similar in the Run 1 data, albeit in a different decay channel, and then, back in 2017, confirmed the hints in the first Run 2 data, the coincidence could not go unnoticed.
The first articles of theorists appeared at the end of 2017 (see, for example, articles arXiv: 1710.01743, arXiv: 1710.04663, arXiv: 1710.07649), and they tried to understand which versions of the Higgs mechanism the discovered pair of deviations fit into. If we assume that a new particle with a mass of about 95 GeV really exists, then from the LEP data it turns out that the signal intensity in b-anti-b channel is about 10% of the standard Higgs boson. For the two-photon deflection seen by the CMS, the signal strength is higher, about half what a standard Higgs boson would give. No deviations were observed in other decay channels. Therefore, when theorists build models, they must take into account all of these constraints. In particular, this deviation cannot be accommodated in the minimal supersymmetric model (MSSM), however, more complex versions of supersymmetric theories can cope with this quite well (see articles arXiv: 1712.07475, arXiv: 1807.06322).
Now the situation is evolving, but rather slowly. The current article of the CMS collaboration has not taken anyone by surprise. She just confirmed the preliminary data, which the collaboration had already reported earlier. Interestingly, another collaboration, ATLAS, also performed such a search in the mass range from 65 to 110 GeV, but it did not find any deviations worth mentioning (see preliminary publication ATLAS-CONF-2018-025 for July 2018). However, the upper limit obtained there turned out to be slightly worse than that of the CMS. So as a result, the two datasets do not yet contradict each other, which was clearly demonstrated in the recent report by S. Heinemeyer, T. Stefaniak, A Higgs Boson at 96 GeV ?!, which appeared in the archive of preprints last week.
I must say that, having burned themselves on the failed discoveries of past years, theorists now treat such deviations rather cautiously. Therefore, theoretical publications with their discussions are going on, but there is no excitement. However, the situation will change dramatically if this deviation is confirmed in the complete statistics of the Run 2. However, this data will not be processed soon. So far, this situation can be considered another argument in favor of the fact that we need a new, much more perspicacious electron-positron collider.
A source: CMS Collaboration. Search for a standard model-like Higgs boson in the mass range between 70 and 110 GeV in the diphoton final state in proton-proton collisions at √s = 8 and 13 TeV // arXiv preprint: 1811.08459 [hep-ex]…