⚛️ high-energy experiments

Off-shell Higgs boson measurements: Yukawa couplings, self-coupling, compositeness, and width

Using 138 fb $^{-1}$ of 13 TeV proton-proton collision data, the CMS experiment presents the first direct test of composite Higgs models and the first off-shell constraint on the Higgs self-coupling, while providing the tightest model-independent limits on light-quark Yukawa couplings and a precise measurement of the Higgs boson width of 5.1 $^{+2.0}_{-1.8}$ MeV.

Original authors: CMS Collaboration

Published 2026-07-28

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: CMS Collaboration

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine the universe as a giant, cosmic construction site where everything is built from tiny, invisible Lego bricks. For decades, scientists have been trying to figure out exactly what these bricks are made of and how they stick together. The "Standard Model" is their current instruction manual, a brilliant set of rules that explains almost everything we see, from the glow of a lightbulb to the orbit of a planet. But there's a missing piece in this manual: a special, invisible field that gives all the other particles their weight. Without this field, the universe would be a chaotic mess of massless particles zipping around at the speed of light, unable to form atoms, stars, or you.

In 2012, scientists found the physical "excitation" of this field—a particle named the Higgs boson. Think of the Higgs field as a thick, sticky snowfield. When particles move through it, they get "stuck" and slow down, which we perceive as mass. The Higgs boson is like a snowball thrown through that field; it proves the snow is there. But just finding the snowball isn't enough. Scientists want to know if the snowball is a perfect, solid sphere (as the manual predicts) or if it's actually a cluster of smaller, stranger things stuck together, or if it has secret powers we haven't discovered yet. This is where the story gets exciting: by studying how these particles behave when they are moving very fast and have a lot of energy, physicists can peek behind the curtain to see if the universe's instruction manual needs a rewrite.

This paper from the CMS Collaboration at CERN's Large Hadron Collider is like a high-speed detective story. The team took a massive dataset of 138 billion billion proton collisions (a number so big it's hard to imagine) and looked for a very specific, rare event: the Higgs boson decaying into four "leptons" (a type of particle like an electron or muon). While most people look for the Higgs boson when it's sitting still (the "on-shell" region), this team decided to look at the Higgs when it's in a high-energy, "off-shell" state. Imagine a guitar string: when you pluck it gently, it vibrates at a specific, clear note. But if you hit it hard, it vibrates wildly, creating a messy, complex sound. The "off-shell" region is that messy, high-energy vibration.

The researchers used this messy, high-energy data to test four major ideas about the Higgs boson. First, they checked if the Higgs is a "composite" object, meaning it's made of smaller, tighter-bound parts, like a grapefruit made of smaller grapes. If it were, the high-energy collisions would look different than if it were a single, solid point. They found no evidence of this; the Higgs behaves like a single point down to a scale defined by a compositeness limit of 870 GeV, setting a new bound that any potential sub-structure must be smaller than this energy scale.

Second, they looked at the "glue" holding the Higgs together. In the Standard Model, the Higgs interacts with heavy particles like the top quark in a specific way. The team tested if there were any secret, heavy particles (like a fourth generation of quarks) running in the background loops of these interactions. They found no signs of these ghosts; the Higgs seems to be interacting exactly as the standard rules predict, with no hidden heavy partners.

Third, they investigated the Higgs's "self-coupling," or how much the Higgs likes to interact with itself. This is like checking if a magnet repels or attracts another magnet of the same kind. The paper provides the first constraint on this self-interaction in the off-shell region, finding that while the data isn't precise enough to pin down the exact strength yet, it rules out some extreme possibilities and offers a new, independent way to measure this property compared to other methods.

Finally, and perhaps most importantly, they used all this data to measure the "width" of the Higgs boson. In particle physics, "width" is a measure of how quickly a particle decays or falls apart. A narrow width means it lives a long time; a wide width means it dies quickly. By combining the "on-shell" (quiet) and "off-shell" (loud) data, the team calculated the Higgs width to be 5.1 MeV, with a range of uncertainty between 1.7 and 9.3 MeV. This is the tightest constraint ever achieved. They also confirmed with high confidence (more than 5 standard deviations) that the Higgs boson does exist in this off-shell, high-energy state, ruling out the idea that it only appears when it's calm and slow.

In short, this paper doesn't find a new particle or a broken law of physics. Instead, it acts as a rigorous stress test. It squeezes the Higgs boson in high-energy collisions and says, "You are still behaving exactly like the perfect, elementary particle the Standard Model predicted." While this might sound like "nothing new," in the world of particle physics, proving that the universe is exactly as we think it is, down to the tiniest details, is a massive victory. It tells us that if there are new, exotic particles hiding in the shadows, they are even more elusive than we thought, and the Standard Model remains the most accurate description of reality we have.

Technical Summary: Off-shell Higgs boson measurements: Yukawa couplings, self-coupling, compositeness, and width

Problem and Motivation
While the Higgs boson ( $H$ ) was discovered in 2012 with properties consistent with the Standard Model (SM), significant gaps remain in understanding its interactions, particularly regarding its total width ( $\Gamma_H$ ), self-coupling, and potential substructure. The on-shell production region (around $m_H \approx 125$ GeV) is dominated by the resonant peak, where the intrinsic width is negligible compared to detector resolution, making direct width measurements impossible. Furthermore, on-shell analyses often rely on model-dependent assumptions, such as custodial symmetry ( $\kappa_Z = \kappa_W$ ) or bounds on coupling modifiers ( $|\kappa_Z| \leq 1$ ), to resolve degeneracies between the total width and production cross-sections.

The off-shell region (where the invariant mass of the diboson system $m_{4\ell} > 2m_Z$ ) offers a unique probe. In this regime, the Higgs propagator is far from the pole, and the production cross-section is proportional to the square of the couplings ( $g_p^2 g_d^2$ ) rather than inversely proportional to the width ( $g_p^2 g_d^2 / \Gamma_H$ ). This allows for the extraction of $\Gamma_H$ by comparing on-shell and off-shell rates. Additionally, the high momentum transfer ( $q^2$ ) in off-shell production provides sensitivity to the Higgs boson's internal structure (compositeness) and loop-induced effects from heavy or light particles that are inaccessible in on-shell measurements.

Methodology
This analysis utilizes proton-proton collision data collected by the CMS experiment at $\sqrt{s} = 13$ TeV, corresponding to an integrated luminosity of 138 fb $^{-1}$ . The study focuses on the $H \to ZZ \to 4\ell$ decay channel ( $\ell = e, \mu$ ) and includes a combined analysis with $H \to ZZ \to 2\ell2\nu$ and $H \to WW \to 2\ell2\nu$ channels to enhance sensitivity.

Event Selection and Categorization: Events are reconstructed using the particle-flow algorithm. Four prompt, isolated leptons are required. Events are categorized into three topologies based on associated jet activity: VBF-tagged (Vector Boson Fusion), VH-tagged (Associated production with a vector boson), and untagged (dominated by gluon-gluon fusion).
Kinematic Discriminants: To separate signal from the large $qq \to ZZ$ background and to isolate interference effects, the analysis employs kinematic discriminants ( $D_{sig}$ and $D_{int}$ ) derived from the Matrix-Element Likelihood Approach (MELA). These discriminants utilize up to 13 kinematic observables. The MILOMERGE procedure is used to optimally compress the two-dimensional $(D_{sig}, D_{int})$ space into a single dimension while preserving statistical power.
Statistical Framework: An extended maximum likelihood fit is performed across the $m_{4\ell}$ spectrum (binned from 220 GeV to higher masses) and the discriminant distributions. The fit model includes signal, background, signal-background interference, and cross-feed contributions (e.g., on-shell $ZH$ events misreconstructed in the off-shell region).
BSM Scenarios: The analysis tests several Beyond-the-Standard-Model (BSM) hypotheses by introducing free parameters in the likelihood fit:
- Compositeness: A form-factor $F(q^2)$ dependent on a compositeness scale $\Lambda_H$ is applied to the Higgs production and decay vertices.
- Heavy Particles: CP-even and CP-odd couplings of the top quark and a generic heavy quark ( $Q$ ) in the $ggH$ loop are varied ( $\kappa_t, \kappa_Q, \tilde{\kappa}_t, \tilde{\kappa}_Q$ ).
- Light Quark Yukawas: Couplings of light quarks ( $u, d, s, c$ ) to the Higgs are constrained, relaxing the on-shell assumption $|\kappa_Z| \leq 1$ .
- Self-Coupling: The trilinear Higgs coupling modifier $\kappa_\lambda$ is measured via loop corrections in single Higgs production.
- Width: $\Gamma_H$ is extracted under progressively general BSM hypotheses, including the relaxation of custodial symmetry and the profiling of heavy and light sector couplings.

Key Contributions and Results

Compositeness Limits: The first direct test of composite Higgs models in the off-shell region is performed. A lower limit on the compositeness scale is set at $\Lambda_H > 870$ GeV at 95% confidence level (CL) when combining on- and off-shell data (assuming $\Gamma_H \leq \Gamma_H^{SM}$ ). Using off-shell data alone yields $\Lambda_H > 310$ GeV.
Heavy and Light Quark Couplings:
- Constraints are placed on CP-even and CP-odd couplings of heavy states ( $\kappa_t, \kappa_Q, \tilde{\kappa}_t, \tilde{\kappa}_Q$ ) in the gluon-fusion loop.
- Crucially, the off-shell analysis allows for the measurement of light-quark Yukawa couplings ( $\kappa_u, \kappa_d, \kappa_s, \kappa_c$ ) without imposing the restrictive on-shell assumptions $|\kappa_Z| \leq 1$ and $\kappa_Z = \kappa_W$ . While the sensitivity to light quarks is lower in the off-shell region, the combination with on-shell data breaks the degeneracy between $\Gamma_H$ and $\kappa_Z$ , providing model-independent constraints.
Higgs Self-Coupling: The first constraint on the Higgs boson self-coupling ( $\kappa_\lambda$ ) in the off-shell region is obtained. The analysis finds no significant deviation from the SM value, with an observed 95% CL interval of $[-21, 27]$ . This provides a complementary probe to double-Higgs production searches.
Higgs Boson Width:
- The analysis demonstrates the robustness of the off-shell method: even when profiling over a wide range of BSM effects (custodial symmetry breaking, heavy/light particle loops), the constraint on $\Gamma_H$ remains strong, with only modest degradation in precision.
- A combined analysis of $H \to ZZ$ and $H \to WW$ channels yields the tightest constraint to date: $\Gamma_H = 5.1^{+2.0}_{-1.8}$ MeV, with a 95% CL interval of $[1.7, 9.4]$ MeV.
Significance of Off-Shell Production: The hypothesis of no off-shell Higgs boson production is excluded with a significance exceeding 5 standard deviations (observed), compared to an expected significance of 2.9 standard deviations.

Significance
The paper claims that these measurements substantially broaden the Higgs boson physics program. By exploiting the off-shell region, the analysis delivers shape-based sensitivity to Higgs substructure, loop effects, light-quark Yukawa couplings, and self-coupling. The primary significance lies in the ability to provide more model-independent results; specifically, the off-shell method allows for the relaxation of standard on-shell assumptions (such as bounds on vector boson couplings) while still deriving meaningful constraints on fundamental parameters. The robustness of the width measurement against a wide variety of BSM scenarios further validates the off-shell method as a reliable tool for precision Higgs physics.

More like this