Updates from ICHEP: 750 GeV bump (and other things we didn’t find)

Article: Search for resonant production of high mass photon pairs using 12.9/fb of proton-proton collisions at √s = 13 TeV and combined interpretation of searches at 8 and 13 TeV
Authors: CMS Collaboration
Reference: CERN Document Server (CMS-PAS-EXO-16-027, presented at ICHEP)

In the early morning hours of Friday, August 5th, the particle physics community let our a collective, exasperated sigh. What some knew, others feared, and everyone gossiped about was announced publicly at the 38th International Conference on High Energy Physics: the 750 GeV bump had vanished.

Had it endured, the now-defunct “diphoton bump” would have been the highlight of ICHEP, a biennial meeting of the high energy physics community currently being held in Chicago, Illinois. In light of this, the scheduling of the announcements for a parallel session in the middle of the conference – rather than a specially arranged plenary session – said anything that the rumors had not already: there would be no need for champagne or press releases.

While the exact statistical significance depends upon the width and spin of the resonance in question, meaning that the paper presents multiple p-value plots corresponding to different signal hypotheses, the following plot is a good representative.

Combined background-only p-values for a new scalar particle in the CMS diphoton data from both 2015 and 2016. The dip in blue near 750 GeV is the excess observed during 2015, and the flat red line indicates the result from the 2016 data: no excess. Had a new particle existed at 750 GeV, we would have expected to see a similar dip in the red curve.
Combined background-only p-values for a new scalar particle in the CMS diphoton data from both 2015 and 2016. The dip in blue near 750 GeV is the excess observed during 2015, and the flat red line indicates the result from the 2016 data: no excess. Had a new particle existed at 750 GeV, we would have expected to see a similar dip in the red curve.

We hoped that the 2016 LHC dataset would bring confirmation that the excess seen during 2015 was evidence of a new particle, but instead, the 2016 data has assured us that 2015 was merely a statistical fluctuation. When combining the data currently available from 8 TeV and 13 TeV, the excess at 750 GeV is reduced to <2σ local significance. The channel with the largest, which was 3.4σ local significance excess before the addition of 2016 data, has now been reduced to 1.9 sigma local significance, and other channels have seen analogous drops. As a result, CMS reports that “no significant excess” is observed over the Standard Model predictions.

The excess disappearing was clearly the less-desirable of the two possible outcomes, but is there a silver lining here? This CMS result puts the most stringent limits to date on the production of Randall-Sundrum (RS) gravitons, and the excitement generated by the diphoton bump sparked a flurry of activity within the theory community. A discovery would have been preferred to exclusion limits, and the papers published concerned a signal that has subsequently disappeared, but I would argue that both of these help our field move forward.

However, as we continue to push exclusion limits further across all manner of search for new physics, particle physicists become understandably antsy. Across all manner of searches for supersymmetry and exotica, the jump in energy from 8 to 13 TeV has allowed us to place more stringent exclusion limits. This is great news, but it is not the flurry of discoveries that some hoped the increase in energy during Run II would bring. It seems that there was no new physics ready to jump out and surprise us at the outset of Run II, so if we are to discover new physics at the LHC in the coming years, we will need to pick it out of the mountains of background. New physics may be lurking nearby, but if we want it, we will have to work harder to get it.

 

References and Further Reading

  • CERN Press, “Chicago sees floods of LHC data and new results at the ICHEP 2016 Conference” (link)
  • Alessandro Strumia, “Interpreting the 750 GeV digamma excess: a review” (arXiv:1605.09401)

750 GeV Bump Update

Article: Search for resonant production of high-mass photon pairs in proton-proton collisions at sqrt(s) = 8 and 13 TeV
Authors: CMS Collaboration
Reference: arXiv:1606.04093 (Submitted to Phys. Rev. Lett)

Following the discovery of the Higgs boson at the LHC in 2012, high-energy physicists asked the same question that they have asked for years: “what’s next?” This time, however, the answer to that question was not nearly as obvious as it had been in the past. When the top quark was discovered at Fermilab in 1995, the answer was clear “the Higgs is next.” And when the W and Z bosons were discovered at CERN in 1983, physicists were saying “the top quark is right around the corner.” However, because the Higgs is the last piece of the puzzle that is the Standard Model, there is no clear answer to the question “what’s next?” At the moment, the honest answer to this question is “we aren’t quite sure.”

The Higgs completes the Standard Model, which would be fantastic news were it not for the fact that there remain unambiguous indications of physics beyond the Standard Model. Among these is dark matter, which makes up roughly one-quarter of the energy content of the universe. Neutrino mass, the Hierarchy Problem, and the matter-antimatter asymmetry in the universe are among other favorite arguments in favor of new physics. The salient point is clear: the Standard Model, though newly-completed, is not a complete description of nature, so we must press on.

Background-only p-values for a new scalar particle in the CMS diphoton data. The dip at 750 GeV may be early evidence for a new particle.
Background-only p-values for a new scalar particle in the CMS diphoton data. The dip at 750 GeV may be early evidence for a new particle.

Near the end of Run I of the LHC (2013) and the beginning of Run II (2015), the focus was on searches for new physics. While searches for supersymmetry and the direct production of dark matter drew a considerable deal of focus, towards the end of 2015, a small excess – or, as physicists commonly refer to them, a bump – began to materialize in decays to two photons seen by the CMS Collaboration. This observation was made all the more exciting by the fact that ATLAS observed an analogous bump in the same channel with roughly the same significance. The paper in question here, published June 2016, presents a combination of the 2012 (8 TeV) and 2015 (13 TeV) CMS data; it represents the most recent public CMS result on the so-called “di-photon resonance”. (See also Roberto’s recent ParticleBite.)

This analysis searches for events with two photons, a relatively clean signal. If there is a heavy particle which decays into two photons, then we expect to see an excess of events near the mass of this particle. In this case, CMS and ATLAS have observed an excess of events near 750 GeV in the di-photon channel. While some searches for new physics rely upon hard kinematic requirements or tailor their search to a certain signal model, the signal here is simple: look for an event with two photons and nothing else. However, because this is a model-independent search with loose selection requirements, great care must be taken to understand the background (events that mimic the signal) in order to observe an excess, should one exist. In this case, the background processes are direct production of two photons and events where one or more photon is actually a misidentified jet. For example, a neutral pion may be mistaken for a photon.

Part of the excitement from this excess is due to the fact that ATLAS and CMS both observed corresponding bump sin their datasets, a useful cross-check that the bump has a chance of being real. A bigger part of the excitement, however, are the physics implications of a new, heavy particle that decays into two photons. A particle decaying to two photons would likely be either spin-0 or spin-2 (in principle, it could be of spin-N where N is an integer and N ≥ 2). Models exist in which the aforementioned Higgs boson, h(125), is one of a family of Higgs particles, and these so-called “expanded Higgs sectors” predict heavy, spin-0 particles which would decay to two photons. Moreover, in models which there are extra spatial dimensions, we would expect to find a spin-2 resonance – a graviton – decaying to two photons. Both of these scenarios would be extremely exciting, if realized by experiment, which contributed to the buzz surrounding this signal.

So, where do we stand today? After considering the data from 2015 (at 13 TeV center-of-mass energy) and 2012 (at 8 TeV center-of-mass energy) together, CMS reports an excess with a local significance of 3.4-sigma. However, the global significance – which takes into account the “look-elsewhere effect” and is the figure of merit here – is a mere 1.6-sigma. While the outlook is not extremely encouraging, more data is needed to definitively rule on the status of the di-photon resonance. CMS and ATLAS should have just that, more data, in time for the International Conference on High Energy Physics (ICHEP) 2016 in early August. At that point, we should have sufficient data to determine the fate of the di-photon excess. For now, the di-photon bump serves as a reminder of the unpredictability of new physics signatures, and it might suggest the need for more model-independent searches for new physics, especially as the LHC continues to chip away at the available supersymmetry phase space without any discoveries.

References and Further Reading