ATLAS finds evidence of off-shell Higgs boson production and measures Higgs boson’s total width

15 November 2022 | By

Since discovering the Higgs boson 10 years ago, the ATLAS Collaboration has been carrying out numerous measurements of its parameters and interactions, showing that they are consistent with Standard Model predictions. The Higgs boson’s mass, for instance, has been measured to be 125 GeV with a precision of 0.1%. However, one property that remains inaccessible via direct measurements is the Higgs boson’s “total width”.

Figure 1: The relativistic Breit-Wigner distribution of the Higgs boson resonance with a width ( Γ_H) of 4.1 MeV. For comparison, the width of the Z boson is more than 600 times larger (2.495 GeV), allowing us to measure it directly from the Breit-Wigner line shape. (Image: M. Javurkova/ATLAS Collaboration)

Width is a fundamental parameter of any unstable particle with a finite lifetime – the shorter the lifetime, the broader the decay width (ΓH) – a direct consequence of Heisenberg's uncertainty relation. It is usually determined from the invariant mass of its decay products, which takes the shape of a relativistic Breit–Wigner distribution (see Figure 1). This makes the Higgs boson more likely to be produced around the peak of 125 GeV (“on-shell”), while also allowing the possibility of its production at high masses (“off-shell”) away from the peak.

The Higgs boson's width is predicted to be 4.1 MeV – too small to be directly measured by the ATLAS experiment. However, its value can be determined by comparing the rate of Higgs-boson production in the on-shell and off-shell regions. This relies on the fact that the amount of Higgs bosons produced on-shell depends not only on the Higgs boson’s interactions with other particles, but also on the Higgs-boson width. By contrast, the off-shell signal rate is independent, to a good approximation, of the Higgs-boson total width. The Higgs boson's tiny decay width is an intriguing feature, as it would be modified by the presence of new particles entering the gluon-gluon fusion production of the Higgs boson or new interactions with the Higgs boson. Any deviation from its predicted value would therefore indicate the presence of new physics.

In a new highlight result, the ATLAS Collaboration measured the total width of the Higgs boson as 4.6 ± 2.6 MeV.

In a highlight result presented at the Higgs 2022 conference, the ATLAS Collaboration studied the production of off-shell Higgs bosons with a high invariant mass decaying into two on-shell Z bosons using the data collected during LHC Run 2 (2015-2018). For their new search, ATLAS physicists focused on events where the two Z bosons decay into four charged leptons (ZZ→4l channel) or two charged leptons plus two neutrinos (ZZ→ 2l2v channel), as they offered the highest signal sensitivity.

The probability (or cross section) of off-shell Higgs boson production is modified by sizeable quantum-mechanical interference with background processes, which share the same initial and final states but do not involve the Higgs boson. Such interference effects make this measurement challenging, as they require researchers to precisely model not just the signal and background processes but also how they interfere with each other. ATLAS physicists developed and applied a multi-class, dense neural network to the ZZ→ 4l channel to better identify events originating from a Higgs boson.

Figure 2: Comparisons between data and the Standard Model prediction for the neural network discriminant (a) and 𝑚𝑍𝑍T (b) distributions in the signal regions targeting the gluon-gluon fusion Higgs boson production in the ZZ→ 4l and ZZ→ 2l2v channels, respectively. The hatched area represents the total systematic uncertainty. The last bin in both figures represents the counts expected in the displayed interval, and above. (Image: ATLAS Collaboration/CERN)

Researchers combined results from both channels to measure the ratio of the off-shell Higgs boson production rate to its Standard-Model prediction, 𝜇off-shell. The observed data was found to be consistent with Standard Model predictions, as shown in Figure 2. The data rejects the background-only hypothesis, which assumes no off-shell Higgs boson production, with an observed (expected) significance of 3.2σ (2.4σ). Figure 3 (left) shows the corresponding profile likelihood scans, providing experimental evidence of off-shell Higgs-boson production. The observed (expected) upper limit at 95% confidence level on 𝜇off-shell is found to be 2.3 (2.4). Physicists performed a further statistical combination with on-shell Higgs measurements, and measured the total width of the Higgs boson as 4.6 + 2.6 -2.5 MeV, and an observed (expected) upper limit on the total width is set at 9.7 (10.2) MeV at 95% confidence level, as shown in Figure 3 (right).

Figure 3: The profile likelihood (y axis) as a function of (a) the off-shell Higgs signal strength 𝜇off-shell, for the combination of the ZZ→ 4l and ZZ→ 2l2v off-shell analyses, and (b) Γ𝐻𝐻SM. The horizontal dotted lines correspond to the one and two standard deviation (σ) confidence intervals on the measurement. (Image: ATLAS Collaboration/CERN)

Evidence of off-shell Higgs production is now strongly established, and tight constraints have been placed on relevant parameters. A CERN-LHC Seminar held today, Tuesday 15 November, was dedicated to the discussion of these results, which are compatible with those from the CMS Collaboration. With the increased collision energy and greater accumulated data from expected from LHC Run 3, more precise measurements of off-shell Higgs production and the Higgs-boson width are forthcoming.

Learn more