Physicist Proposes LHC Dijet Excesses Trace to Fourth-Generation Tetraquarks

A theoretical physicist has proposed that puzzling excesses of di-dijet events recorded at the Large Hadron Collider could be the fingerprints of an exotic particle that has never been observed: a tetraquark built from four fourth-generation quarks, each weighing roughly 2 teraelectronvolts. The proposal, published in The European Physical Journal C, argues that two separate anomalies in CMS and ATLAS data can be explained within a single framework, something earlier model-building attempts failed to achieve.

The anomalies in question come from searches in which proton-proton collisions produce pairs of dijet resonances, each resonance itself decaying into two jets. CMS observed two events with a four-jet invariant mass near 8 TeV and an average dijet resonance mass near 2 TeV, standing out with a local significance above 3.6 standard deviations because the quantum chromodynamic background in this region is extremely small. ATLAS reported one event at 6.6 TeV with a dijet mass of 2.2 TeV, and CMS one at 5.8 TeV with 2.0 TeV. Separately, a non-resonant search revealed an excess for four-jet masses between 3 and 4 TeV with an average dijet resonance mass of 0.95 TeV, which a reinterpretation attributed to a resonant structure near 3.6 TeV carrying a local significance of 3.9 and a global significance of 2.2 standard deviations.

The new work, by Hsiang-nan Li, postulates that these excesses arise from the production of four fourth-generation bottom-prime quarks, denoted b′, forming a b′b′b̄′b̄′ tetraquark. In this picture, the signal around a four-jet mass of 8 TeV comes from resonant tetraquark production, while the dijet resonances near 2 TeV correspond to first excited states of b′b̄′ pairs, which behave as color-octet scalars. The excess near 3.6 TeV, by contrast, is attributed to non-resonant production of the same four-quark system, with dijet resonances of about 0.95 TeV corresponding to the ground states of the pairs, which act as color-octet vectors.

The binding mechanism is unusual. The b′ quarks are assumed to be bound not by the strong force but by a Yukawa potential created by Higgs boson exchanges. Previous work has shown that fermions with masses above the teraelectronvolt scale form bound states in such a potential, which is how superheavy fermions with large Yukawa couplings can evade the perturbative unitarity bound. The Yukawa potential supports only a finite number of bound states, unlike, for example, the infinite tower of Kaluza-Klein excitations in theories with extra space-time dimensions, a distinction the author draws explicitly.

To make the numbers work, the paper adopts a b′ quark mass of about 2.0 TeV at a scale of order 1 TeV. This is anchored in earlier analyses of the sequential fourth-generation model, in which renormalization-group evolution of the Yukawa couplings drives the b′ mass from 2.7 TeV at the electroweak scale down to 1.6 TeV at a symmetry restoration scale of order 10 TeV, making an intermediate value of roughly 2 TeV reasonable. Solving a relativistic Dirac equation with the Yukawa potential, the author obtains a ground-state mass of 0.94 TeV and a first excited state mass of 2.1 TeV, values consistent with the observed dijet resonance masses. The calculation requires tuning a spin-symmetry parameter to −1.6 times the b′ mass, and the binding energies turn out to be of the order of the constituent mass itself, a pattern that differs from ordinary QCD bound states because the Yukawa interaction strengthens with mass.

The framework also predicts a third bound state: the n=3 level of the b′b̄′ system comes out at 2.8 TeV. The author notes that a future detection of four-jet events with dijet masses near 2.8 TeV would provide direct support for the proposal. The four-jet events at 6.6 and 5.8 TeV are deliberately left aside, since the tetraquark mass spectrum itself requires further investigation, though they might be associated with lower-lying tetraquark states, and the 5.8 TeV event may be compatible with a broad resonance centered near 8 TeV.

To connect the full four-quark calculation to measurable cross sections, the paper translates the fourth-generation model into effective theories already available in the literature, containing color-octet scalars and vectors. The 8 TeV resonance is matched to a coloron of mass 8 TeV decaying into two color-octet scalars, and the matching fixes an effective coupling at approximately 2.8 times the strong coupling. Scaling an earlier calculation accordingly boosts the expected event yield to about 0.7 events in 78 inverse femtobarns of integrated luminosity, or 1.3 events in 138 inverse femtobarns, which the author argues accommodates the CMS excess at 8 TeV. For the 3.6 TeV excess, the matching to a model with two distinct colorons yields an effective coupling of 0.31, larger than the 0.07 extracted from data when the mediator is treated as a resonance. The resolution proposed is that the mediator is an off-shell coloron of mass 8 TeV, whose propagator suppresses the rate by a factor of roughly 0.2, compensating the coupling enhancement and leaving a cross section close to the one inferred from data, hence the interpretation of the 3.6 TeV excess as non-resonant production.

The whole scenario is presented as a teraelectronvolt-scale analogue of the fully charmed tetraquark X(6900), which LHCb discovered in the four-muon channel and which ATLAS and CMS later confirmed. Just as X(6900) consists of four approximately on-shell charm quarks decaying into pairs of charmonium states, the proposed b′b′b̄′b̄′ resonance at 8 TeV would decay into two excited b′b̄′ states, and the non-resonant production at 3.6 TeV would yield two ground states, mirroring the J/psi and psi(2S) decay patterns of the charmed sector.

The paper also situates itself within the broader fourth-generation program, in which the masses and mixing patterns of quarks and leptons are dictated by analyticity requirements of the underlying dynamics, echoing the S-matrix bootstrap ideas of Geoffrey Chew. The author argues that superheavy fourth-generation quarks bypass experimental constraints from Higgs boson production and decay, since their contributions to gluon fusion and the diphoton decay are estimated at only the 10⁻³ and 10⁻² level of the top quark’s, and that the model survives constraints from the oblique parameters and delta rho. The competing diquark-scalar explanation is disfavored in this framework because the required transition would be suppressed by a tiny CKM matrix element of order 10⁻⁴.

Important caveats remain. The tetraquark mass spectrum has not been computed, the matching procedure relies on effective couplings treated as free parameters fixed by comparison with the full theory, and the interpretation of the 6.6 and 5.8 TeV events is explicitly deferred to future work. High-mass dijet events observed by ATLAS near 8 TeV, and a weaker-than-expected limit below 1 TeV, are flagged for future analysis within the same framework. Whether nature actually hosts a fourth generation of quarks bound by Higgs-mediated forces will depend on whether the predicted spectrum, including the 2.8 TeV state, shows up in future LHC data.

Subject of Research: A theoretical proposal attributing LHC di-dijet excesses to tetraquarks of fourth-generation quarks bound in a Yukawa potential

Article Title: LHC di-dijet excesses as signals of fourth-generation tetraquarks

Article References: Li, H.-N. (2026). LHC di-dijet excesses as signals of fourth-generation tetraquarks. The European Physical Journal C, 86(10), Article 1250. https://doi.org/10.1140/epjc/s10052-026-16517-y

Image Credits: AI Generated

DOI: 10.1140/epjc/s10052-026-16517-y

Keywords: LHC, dijet excess, tetraquark, fourth-generation quarks, Yukawa potential, CMS, ATLAS, color-octet scalars, colorons, particle physics, b′ quark, X(6900)