The search for new physics: take three

2 May 2022

Improved experimental techniques and new guidance from lower-energy experiments put the LHC in a better position than before to address the question of naturalness, describe Patrick Rieck and Aurelio Juste

An ATLAS mono-jet event

Aside from the discovery of the Higgs boson, the absence of additional elementary-particle discoveries is the LHC’s main result so far. For many physicists, it is also the more surprising one. Such further discoveries are suggested by the properties of the Higgs boson, which are now established experimentally to a large extent. The Higgs boson’s low mass, despite its susceptibility to quantum corrections from heavy particles that should push it orders-of-magnitude higher, and its hierarchy of coupling strengths to fermions present extreme, “unnatural” values that so far lack an explanation. Therefore, searches for new physics at the TeV energy scale remain strongly motivated, irrespective of the no-show so far. 

Naturalness has triggered the development of many new-physics models, but the large extent of their parameter space allows them to evade exclusion again and again. Whereas the discoveries of the past decades, including that of the Higgs boson, were driven by precise quantitative predictions, the search for physics beyond the Standard Model (SM) simply requires more perseverance.

LHC Run 3 will bring long-awaited new insights to the question of naturalness with respect to Higgs physics, as well as to many other SM puzzles such as the nature of dark matter or the cosmological matter–antimatter asymmetry. With considerably more data and a slightly higher centre-of-mass energy than at Run 2, in addition to new triggers and improved event reconstruction and physics-analysis techniques, a significant increase in sensitivity compared to the current results will be achieved. Searches for new phenomena with Run 3 data will also benefit from a much improved definition of the physics targets, thanks to information gathered during Run 2 and the various anomalies observed at lower energies.

The story so far

During the past 12 years, a broad search programme has emerged at the LHC in parallel with precision measurements (see “Pushing the precision frontier”). Initially, the most favoured new-physics scenario was supersymmetry (SUSY), a new fermion–boson symmetry that gives rise to supersymmetric partners of SM particles and naturally leads to a light Higgs boson close to the masses of the W and Z bosons. SUSY is expected to produce events containing jets and missing transverse energy (MET), the study of which at Run 2 placed exclusion limits on gluino masses as high as 2.3 TeV. More challenging searches for stop quarks, with background processes up to a million times more frequent than the predicted signal, were also performed thanks to the excellent performance of the ATLAS and CMS detectors. Yet, no signs of stops have been found up to a mass of 1.3 TeV, excluding a sizeable fraction of the SUSY parameter space suggested by naturalness arguments. Further SUSY searches were performed, including those for only weakly interacting SUSY particles (“electroweakinos”), where the Run 2 data allowed the experiments to surpass the sensitivity achieved by LEP in some scenarios. Half a century since SUSY was first proposed, ATLAS and CMS have demonstrated that the simplest models containing TeV-scale sparticle masses are not realised in nature (see “Stop quarks and electroweakinos” figure).

Stop quarks and electroweakinos

In fact, a large number of new-physics searches during LHC Run 1 and Run 2 targeted models other than SUSY, many of which also address the question of naturalness. Signs of extra spatial dimensions have been searched for in “mono-jet” events containing a single energetic jet and large MET, which could be caused by excited gravitons propagating in a higher dimensional space. Searches for vector-like quarks, as suggested by models with a composite Higgs boson, covered numerous complex final states with decays into all of the heavier known elementary particles. In these and other searches, the Higgs boson has entered the experimental toolkit, for example via the identification of high-momentum Higgs-boson decays reconstructed as large-radius jets.

The Higgs sector itself has been the subject of new-physics searches. These target additional Higgs bosons that would arise from an extended Higgs sector and exotic decays of the known Higgs boson, for instance into weakly interacting massive particles (WIMPs), which are candidates for dark matter. Improvements in both theoretical and data-driven background determinations have also allowed searches for Higgs-boson decays into invisible particles, with the Run 2 dataset setting an upper limit of 10% on their rate.

Searches for dark matter also continued to be performed in traditional channels, for example via the mono-jet signature. To increase the accuracy of this search using the full Run 2 statistics, theorists contributed differential background predictions that go beyond the next-to-leading order in perturbation theory to achieve an unprecedented background uncertainty of only 3% at MET values above 1 TeV. The resulting constraints on WIMP dark matter are complementary to those achieved with ultrasensitive detectors deep underground as well as astroparticle experiments. The absence of dark-matter signals in such established search channels led to the development of new models that predict a number of relevant but previously unexplored signatures.

LHC Run 3 will allow searches to go significantly beyond the sensitivity achieved with the Run 2 data 

In several respects, searches for new physics at the LHC experiments have gone well beyond what was foreseen at the time of their design. “Scouting” data streams were introduced to store small-size event records suitable for di-jet and di-muon resonance searches such that recording rates could be increased by up to two orders of magnitude within the available bandwidth. Consequently, the mass reach of these searches was extended to lower values whereas previously this was impossible due to the high background rates at low masses. Long-lived particle searches also opened a new frontier, motivating proposals for new LHC detectors.

Overall, LHC Run 1 and Run 2 led to an enormous diversification of new-physics searches at the energy frontier by ATLAS and CMS, with complementary searches conducted by LHCb targeting lower invariant masses. The absence of new-physics signals despite the exploration of a multitude of signatures with unforeseen precision is a strong experimental result that feeds back to the phenomenology community to shape this programme further. While the analysis of Run 2 data is still ongoing, the experience gained so far in terms of experimental techniques and investigated signatures puts the experimental collaborations in a better position to search for new physics at Run 3.

Experimental improvements

LHC Run 3 will allow searches to go significantly beyond the sensitivity achieved with the Run 2 data. ATLAS and CMS are expected to collect datasets with an integrated luminosity of up to 300 fb–1, adding to the 140 fb–1 collected in Run 2. Taking into account the additional, smaller benefit provided by the increase in the centre-of-mass energy from 13 to 13.6 TeV, new-physics search sensitivities will generally increase by a factor of two in terms of cross sections. Additional gains in sensitivity will result from the exploration of new territory in several respects.

Already at the level of data acquisition, significant improvements will increase the sensitivity of searches. The CMS higher level trigger system has been reinforced using graphics processing units to increase the recording rate in the data scouting stream from 9 to 30 kHz. ATLAS has extended this technique to encompass more final states, including photons and b-jets. These techniques extend the sensitivity to hadronic resonances with low masses and weak coupling strengths to a domain that has never been probed before.

Mass exclusions for spin-1 leptoquarks

The particularly challenging searches for new long-lived particles will also benefit from experimental advances. ATLAS has improved the reconstruction of displaced tracks, reducing the amount of fake tracks by a factor of 20 at similar efficiencies compared to the current data analysis. New, dedicated triggers have been developed by ATLAS and CMS to identify electrons, muons and tau-leptons displaced from the primary interaction vertex. These trigger developments will allow the collection of signal candidate events at unprecedented rates, for example to test exotic Higgs-boson decays into long-lived particles with branching ratios far below the current experimental limits. 

Likewise, ongoing developments in machine learning will contribute to the Run 3 search programme. While Run 1 physics analyses used generic, simple algorithms to distinguish between hypotheses, in Run 2 more powerful approaches of deep learning were introduced. For Run 3 their development continues, using a multitude of different algorithms tailored to the needs of event reconstruction and physics analysis to increase the reach of new-physics searches further.

New signatures

The Run 3 data will also be scrutinised in view of final states that either have been proposed more recently or that require a particularly large dataset. Examples of the latter are searches for electroweakinos, which have a production cross-section at the LHC at least two orders of magnitude smaller than strongly interacting SUSY particles. First results based on Run 2 data surpassed the sensitivity of the LEP experiments, including tests of unconventional “R-parity violating” scenarios in which electroweakinos can decay into only SM particles. This results in complicated final states containing electrons, muons and many jets but relatively low MET. Here, the challenging background determination could only be achieved thanks to machine-learning techniques, which lay the ground for further searches for particularly rare and challenging SUSY signals at Run 3.

If R-parity is not a symmetry, SUSY does not provide a WIMP dark-matter candidate. Among alternative explanations of the nature of this substance, models with bound-state dark matter are gaining increasing attention. In this new approach, strong interactions similar to quantum chromodynamics determine the particle spectrum in a dark sector that includes stable dark-matter candidate particles such as dark pions. At the LHC, coupling between such dark-sector particles and known ones would result in “semi-visible” jets comprising both types of particle (traditional dark-matter searches at the LHC have avoided such events to reduce background contributions). With the Run 2 data, CMS has already provided the very first collider constraints on these dark sectors, and more results from both ATLAS and CMS will follow in this and other proposed dark-sector scenarios.

Multiple deviations from the SM observed at lower energies are starting to shape the search programme at the energy frontier. The long-standing anomaly in the magnetic moment of the muon has recently reached a significance of 4.2σ, motivating increased efforts in searching for possible causes. One is the pair-production of a supersymmetric partner of the muon, for which models fit the low-energy data if the mass of this “smuon” is below 1 TeV and hence within the reach of the LHC. Another is to look for vector-like leptons, which are suggested by consistent extensions of the SM apart from SUSY, using final states containing a large number of leptons.

Multiple deviations from the SM observed at lower energies are starting to shape the search programme at the energy frontier

Moreover, the anomalies in B-meson decays consistently reported by BaBar, Belle and LHCb (see “A flavour of Run 3 physics”) have a strong and growing impact on the Run 3 search programme. Explanations for these anomalies require new particles with TeV-scale masses to fit the size of the observed effects and a hierarchy of fermion couplings to fit the deviations from lepton-flavour universality. Intriguingly these two requirements happen to coincide with the two peculiarities of the Higgs boson. Particular attention is now given to leptoquark searches investigating several production and decay modes. ATLAS and CMS have already started to probe leptoquark models suggested by the B-meson anomalies using Run 2 data (see “Leptoquarks” figure). While the analysis of key channels is ongoing, Run 3 will allow the experiments to probe a large fraction of the relevant parameter space. Furthermore, consistent models of leptoquarks include more new particles, namely colour-charged and colour-neutral bosons, vector-like quarks and vector-like leptons. These predict a variety of new-physics signatures that will further shape the Run 3 search programme.

In summary, searches for new physics at Run 3 will bring significant gains in sensitivity beyond the benefit provided by the increased amount of data. In particular, potential explanations of the anomalies observed at lower energies will be tested. Assuming that these anomalies point to new physics, the relevant searches with Run 3 data have a good chance of finding the first deviations from the SM at the TeV energy scale. Such an outcome would be of the utmost importance for particle physics, strengthening the case for the proposed Future Circular Collider at CERN.

bright-rec iop pub iop-science physcis connect