One of the great unexplained mysteries is the nature of dark matter. So far, its existence has only been established through gravitational effects observed in space; no dark-matter particles with the needed properties have (yet) been detected. Could the Higgs boson be the key to their discovery?

Looking for dark matter in particles of light

The high-energy proton–proton collisions of the LHC could be the perfect environment to study dark-matter particles. In particular, if these elusive particles are produced in association with the Higgs boson, the ATLAS experiment should be sensitive to their presence. Although the dark-matter particles would escape the direct detection – as they would not interact with the sensitive material of the detectors – they could result in collision events with a substantial amount of missing transverse momentum. The experiment would detect a large momentum imbalance among the reconstructed particles in the plane perpendicular to the LHC beam.

At this week’s Higgs 2020 conference, the ATLAS Collaboration presented a new search for dark-matter particles produced in association with a Higgs boson decaying to two photons. ATLAS researchers studied the full Run 2 dataset collected between 2015 and 2018, selecting for events with two high-quality photons and large missing transverse momentum. Their search focused on three types of dark-matter models – namely the Z_’B, Z’-2HDM, and 2HDM+a models – all of which include new vector or pseudoscalar mediator particles.

To identify a potential dark-matter signal, the ATLAS physicists used a machine-learning algorithm (boosted decision tree) to spit events into categories with increasing sensitivity by analysing the kinematic information of the selected photon pair and the event’s missing transverse momentum. In each event category, they looked at the invariant-mass spectrum of the two photons, where a Higgs boson produced in association with dark-matter particles would show an enhanced peak at around 125 GeV over a smooth, non-Higgs-boson background.

Figure 1 shows the diphoton invariant-mass spectrum in the most sensitive event category, comparing the data with the fitted number of expected signal and background events. The data has a small mass peak at around 125 GeV that is fully consistent with a fluctuation of the Standard Model background.

The ATLAS Collaboration observed no significant excess of signal events in any of the categories, and thus set limits on three dark-matter models. In the context of the Z’B model, for dark-matter particles of 1 GeV, the new result extends the limit on the Z’_B mediator mass to 1150 GeV and significantly improves upon previous ATLAS results.

Seeking out “invisible” particles

Another ATLAS search for dark matter looked for Higgs-boson decays into particles invisible to the detector, leading to missing transverse momentum. Such decays are actually possible in the Standard Model; the Higgs boson can decay via two Z bosons into four neutrinos, which escape the detector undetected and thus resemble dark matter particles. However, this only happens in ~0.1% of Higgs-boson decays and is thus negligible; if ATLAS were to see such events, it would be a clear sign of new physics.

Since all known massive elementary particles interact with the Higgs boson, it is reasonable to assume that as-yet undiscovered dark-matter particles would behave the same way. New-physics models where dark matter interacts with known particles through the Higgs boson are called “Higgs portal” models. If the dark matter particles have a mass of less than half the Higgs-boson mass, the Higgs boson could decay into a pair of dark-matter particles.

ATLAS physicists considered different production mechanisms of the Higgs boson to search for such events in different event topologies; to increase the overall sensitivity, these searches can be statistically combined. Such a combination was conducted by ATLAS in 2019 with partial Run 2 data, resulting in an upper limit of 26% (17% expected) on the Higgs-to-invisible decay probability (branching fraction).

The ATLAS Collaboration has now released a new combination using the full LHC Run 2 dataset, further constraining the invisible decay channel. The result combines several measurements, including the recent search for invisible particles via Higgs bosons produced through vector-boson fusion – a result which, by itself, already sets an upper limit of 13% on the Higgs-to-invisible branching fraction. The new combination also incorporates studies of the Higgs boson produced in association with a top–antitop pair, focusing on final states with zero or two leptons. Physicists were able to reinterpret these results – originally conducted outside the Higgs-to-invisible context – to look for dark-matter particles.

ATLAS researchers then went one step further, combining their analysis of the full Run 2 dataset with a similar Higgs-to-invisible analysis using the full Run 1 dataset, which examined the Higgs-boson production via vector-boson fusion and its associated production with a W or Z boson. Their final result found no evidence for invisible particles, and thus physicists set a new upper limit of 11% (11% expected) on the Higgs-to-invisible branching fraction, at a 95% confidence level. This is the strongest direct limit on this process up to date.

Making important progress

The new results from the ATLAS Collaboration set strong constraints on many dark-matter models, making important progress in the ongoing search for new physics. In the future, ATLAS physicists will further improve the discovery potential of using the Higgs boson as a tool to search for dark matter and other new physics phenomena.

Links