⚛️ phenomenology

Using Higgs Boson Production in Association with Three Jets via Gluon Fusion at the LHC to Investigate the CP Properties of the Higgs Boson

This paper presents a comprehensive study of Higgs boson production in association with three jets via gluon fusion at the LHC, incorporating full top- and bottom-quark mass effects and all partonic subprocesses to demonstrate how specific kinematic observables, particularly azimuthal correlations, can effectively probe the CP properties of the Higgs boson and reveal the significant impact of bottom-quark loops that are often neglected in effective field theory approximations.

Original authors: Arunima Bhattacharya, Francisco Campanario, Michael Kubocz, Terrance Figy, Tunde Kushimo, Ivan Rosario

Published 2026-10-28

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Arunima Bhattacharya, Francisco Campanario, Michael Kubocz, Terrance Figy, Tunde Kushimo, Ivan Rosario

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

In the vast, high-energy collisions occurring deep inside the Large Hadron Collider, scientists are hunting for the subtle fingerprints of a particle that gives mass to the universe: the Higgs boson. Since its discovery in 2012, physicists have been meticulously checking whether this particle behaves exactly as the Standard Model of physics predicts, or if it hides a secret. One of the most intriguing possibilities is that the Higgs boson might not be a simple, symmetrical object, but rather a mixture of two different types of behavior: one that looks the same in a mirror (scalar) and one that flips like a left hand becoming a right hand (pseudoscalar). This property, known as CP symmetry, is crucial because the known laws of physics do not contain enough of this "mirror-flipping" violation to explain why the universe is made of matter rather than antimatter. To find out if the Higgs boson holds this secret, researchers must look at how it is created alongside other particles, specifically jets of debris, and analyze the angles at which these particles fly apart.

A team of theorists has now performed a detailed simulation of a specific, complex scenario: the creation of a Higgs boson alongside three jets of particles, all produced through a process called gluon fusion. In this process, two particles of light called gluons smash together, and their energy briefly transforms into a heavy loop of quarks before materializing as a Higgs boson and three jets. While previous studies looked at this process using simplified assumptions about the mass of the heavy quarks involved, this new work calculates the entire event with exact precision, accounting for the real masses of both the top quark and the bottom quark. Crucially, the researchers included the complete set of partonic subprocesses, not just the purely gluonic ones. They found that channels initiated by quarks (such as gluon-quark and quark-quark interactions) are numerically vital, supplying close to one half of the total cross section for pseudoscalar Higgs production with three jets. The researchers also found that these details matter immensely. When they included the bottom quark, which had been largely ignored in earlier approximations for certain conditions, the predicted rate of these events changed dramatically. In scenarios where the interaction strength of the bottom quark is enhanced, it can dominate the production process, altering the total number of events by hundreds of times compared to predictions that only considered the top quark.

The study also tested how well simplified models could describe these high-energy collisions. Scientists often use an "effective theory" that treats the heavy quarks as if they are infinitely heavy to make calculations easier, sometimes adding a correction factor to fix the results. The researchers discovered that while this shortcut works well for the overall number of events at lower energies, it fails completely when looking at the most energetic parts of the collision. When the jets or the Higgs boson fly apart with high momentum, the simplified model predicts a much harder, more energetic spray of particles than what actually occurs when the full, complex physics is calculated. The correction factor, which fixes the behavior of the simplest loop diagram, cannot account for the complex geometry of the larger loops involved when three jets are present. This means that for precise measurements at the highest energies, the simplified models are unreliable, and the full, heavy-quark calculations are necessary.

Beyond the total number of events, the researchers examined the shapes of the distributions, looking at how the particles are arranged in space. They found that some measurements are surprisingly robust; the relative positions of the jets in the forward and backward directions, and the centrality of the third jet, remained largely the same regardless of whether the full calculation or the simplified model was used. However, other measurements were highly sensitive. The energy of the jets and the mass of the pair of leading jets showed clear distortions when the bottom quark contribution became significant. Most importantly, the team analyzed a specific angle formed by the jets on either side of the Higgs boson. For a particle that is purely one type or the other, this angle follows a predictable pattern. But if the Higgs boson is a mixture of both types, a new, distinct pattern emerges in the data: an asymmetry where the particles prefer to fly in one direction over the opposite one. This specific signature, which cannot be produced by a pure particle, acts as a clear signal of a mixed state.

The work confirms that to truly understand the nature of the Higgs boson and its potential role in the matter-antimatter imbalance of the universe, physicists cannot rely on simplified shortcuts or ignore the contributions of lighter heavy quarks like the bottom. The full complexity of the quantum loops must be included, especially when analyzing the most energetic collisions where new physics might hide. By providing these precise predictions for the Large Hadron Collider operating at its current energy levels, the study offers a reliable map for experimentalists. They can now compare their real-world data against these exact calculations, looking for the tell-tale signs of a mixed Higgs boson or deviations that point to physics beyond our current understanding. The results show that the bottom quark, often overshadowed by its heavier cousin, plays a starring role in these high-energy events, reshaping the landscape of what we expect to see and how we interpret the fundamental symmetries of nature.

Technical Summary: Higgs Boson Production in Association with Three Jets via Gluon Fusion

Problem and Motivation
Following the 2012 discovery of the Higgs boson, a primary objective of the LHC physics program is to determine the CP structure of the Higgs interactions. While the Standard Model (SM) predicts a CP-even scalar, the observed baryon asymmetry of the Universe suggests the potential existence of additional CP-violating sources. Higgs production in association with jets via gluon fusion is a sensitive probe of the Higgs–gluon interaction tensor structure.

Previous studies of Higgs-plus-two-jet production have established that azimuthal correlations between tagging jets are sensitive to CP properties. However, the inclusion of a third hard jet, necessary for higher-order QCD radiation studies and background discrimination, complicates the analysis. The standard two-jet azimuthal variable becomes unnatural in the presence of a third jet. Furthermore, existing calculations for Higgs-plus-three-jet production ( $pp \to \Phi jjj$ ) were largely restricted to:

The purely gluonic subprocess ( $gg \to ggg\Phi$ ), neglecting quark-containing channels.
The heavy-top effective field theory (EFT) approximation, which integrates out the top quark.
Calculations that often omitted the bottom-quark loop contribution or its interference with the top loop.

The heavy-top EFT approximation is known to break down in the high-transverse-momentum tails of distributions, and its validity is further compromised in scenarios where bottom-quark loops are enhanced, such as in Type-II Two-Higgs-Doublet Models (2HDM) at large $\tan\beta$ . There is a need for a complete calculation that includes all partonic subprocesses, retains exact finite quark-mass dependence for both top and bottom quarks, and investigates CP-sensitive observables in the three-jet regime.

Methodology
The authors present a leading-order (LO) calculation of the process $pp \to \Phi jjj$ at order $\mathcal{O}(\alpha_s^5)$ , where $\Phi$ is a generic neutral Higgs state with mixed scalar ( $H$ ) and pseudoscalar ( $A$ ) couplings. The calculation is implemented within the VBFNLO framework (specifically the GGFLO component).

Key methodological features include:

Complete Partonic Subprocesses: The calculation includes the full set of partonic channels: $gg \to ggg\Phi$ , $gg \to q\bar{q}g\Phi$ , $gq \to ggq\Phi$ , $gq \to q\bar{q}q\Phi$ , $q\bar{q} \to ggg\Phi$ , and $qq \to qqg\Phi$ , along with all crossing-related processes.
Exact Mass Dependence: The one-loop amplitudes retain the exact mass dependence for both top ( $m_t$ ) and bottom ( $m_b$ ) quarks. The top and bottom contributions are combined coherently, preserving their interference.
Master Topologies: The amplitudes are constructed using a helicity formalism and effective currents. The calculation involves master loop topologies up to hexagons (for purely gluonic processes), pentagons (for two external quarks), and boxes (for four external quarks). Numerical stability for high-rank tensor integrals is ensured via Ward identity tests, with a "rescue system" using quadruple precision for unstable phase-space points.
CP Mixing: The framework allows for arbitrary scalar and pseudoscalar Yukawa couplings, covering pure CP-even, pure CP-odd, and genuinely CP-mixed configurations. A specific phenomenological benchmark is defined for a CP-mixed state $\Phi$ with Type-II 2HDM-like coupling structures.
Observables: The study analyzes integrated cross sections, differential distributions (transverse momenta, dijet mass, rapidity separation), and a specific multi-jet azimuthal observable $\Phi_2$ . The observable $\Phi_2$ is constructed from the vector sums of jet momenta on either side of the Higgs boson in rapidity, designed to retain CP sensitivity in the presence of additional radiation.

Key Contributions and Results

Importance of Quark-Containing Channels:
Contrary to previous approximations restricted to the purely gluonic channel, the authors find that quark-initiated subprocesses are numerically dominant. For a pseudoscalar state ( $A$ ) with $m_A = 125.2$ GeV and $\tan\beta = 1$ , the purely gluonic channel ( $gg \to gggA$ ) contributes only $\sim 51.4\%$ of the total cross section. The $gq$-initiated channels contribute $\sim 37.9\%$ , and the remaining quark-quark channels add $\sim 4.8\%$ . Thus, restricting calculations to the gluonic channel misses nearly half of the cross section.
Bottom-Quark Dominance and Non-Monotonicity:
In a Type-II 2HDM, the bottom-quark contribution scales with $\tan\beta$ . While negligible at $\tan\beta=1$ (changing the rate by $\sim 1\%$ ), the bottom loop becomes the dominant contribution at large $\tan\beta$ . The total cross section exhibits non-monotonic behavior: it decreases as $\tan\beta$ increases (due to the suppression of the top loop by $\cot^2\beta$ ) until a minimum is reached, after which the enhancement of the bottom Yukawa coupling ( $\propto \tan\beta$ ) drives the cross section back up. At $\tan\beta=30$ , the full top-plus-bottom result is $\sim 578$ times larger than the top-only prediction.
Breakdown of Effective Theory in Hard Tails:
Comparisons with the heavy-top EFT and its form-factor-improved (EFT+FF) version reveal significant discrepancies in differential distributions. While EFT+FF reproduces the inclusive rate reasonably well at low $\tan\beta$ , it fails to describe the "hard tails" of the transverse momentum distributions ( $p_{T,\Phi}$ , $p_{T,j1}$ , $p_{T,j3}$ ) and the high invariant mass of the leading dijet ( $m_{j1j2}$ ). The form factors, derived from triangle amplitudes, do not capture the full momentum dependence of the box, pentagon, and hexagon amplitudes entering the three-jet process. Consequently, EFT+FF predicts significantly harder spectra than the exact calculation when $p_T$ approaches or exceeds the top-quark mass.
Stability of Centrality Observables:
In contrast to transverse momentum observables, the normalized distribution of the third-jet centrality ( $z^\star$ ) and the rapidity separation of the leading jets ( $|\Delta\eta_{j1j2}|$ ) are found to be largely insensitive to the heavy-quark treatment (EFT vs. exact mass) and to the value of $\tan\beta$ . This suggests these observables are robust against finite-mass effects.
CP Sensitivity via Azimuthal Correlations:
The study investigates the azimuthal correlation $\Phi_2$ .
- CP Eigenstates: For pure CP-even ( $h$ ) and CP-odd ( $A$ ) states, the distribution is symmetric under $\Phi_2 \to -\Phi_2$ , modulated by a $\cos(2\Phi_2)$ term. The amplitude of this modulation differs between the two states ( $A_2 \approx 0.40$ for $h$ and $0.30$ for $A$ at $\tan\beta=1$ ).
- CP-Mixed States: For a CP-mixed state $\Phi$ , the interference between scalar and pseudoscalar couplings introduces a $\sin(2\Phi_2)$ term. This breaks the symmetry of the distribution. The authors propose a counting asymmetry, $A_{\sin}$ , which isolates this term. This asymmetry vanishes for any CP eigenstate but is non-zero for the mixed state (e.g., $-21\%$ at $\tan\beta=1$ ).
- Bottom-Loop Effects: The shape of the $\Phi_2$ distribution is modified by the bottom-quark contribution even for a state of definite CP, indicating that heavy-quark mass effects influence angular observables used for CP studies.

Significance and Claims
The paper claims that a reliable description of Higgs-plus-three-jet production requires both the complete set of partonic subprocesses and the exact heavy-quark mass dependence. The authors demonstrate that:

Neglecting quark-containing channels leads to a significant underestimation of the cross section.
The heavy-top EFT approximation, even when improved with form factors, is insufficient for precision studies in the high- $p_T$ regime and for scenarios with enhanced bottom-quark contributions.
The multi-jet azimuthal observable $\Phi_2$ , combined with a counting asymmetry, provides a robust, fit-free method to isolate CP-violating interference terms in the effective Higgs–gluon coupling, distinguishing CP-mixed states from CP eigenstates even in the presence of significant bottom-quark loop contributions.

These results are presented for LHC conditions at $\sqrt{s} = 13.6$ TeV and $m_\Phi = 125.2$ GeV, providing a theoretical baseline for precision measurements of Higgs properties and searches for CP violation in gluon-fusion production.

More like this