New Physics Model Solves Decades-Old Charmonium Decay Puzzle

Physicists have developed a new mathematical model that successfully reconciles decades of contradictory data regarding how heavy subatomic quark pairs bind together into heavy particles known as charmonium. By analyzing 286 separate collision measurements from CERN’s Large Hadron Collider (LHC), researchers managed to simultaneously extract the fundamental parameters governing the creation of three distinct charmonium states.

Charmonium particles, which include the well-known $J/\psi$, $\psi(2S)$, and $\chi_c$ states, consist of a charm quark and its antimatter partner bound together by the strong force. Explaining how these quark pairs transform from initial high-energy collisions into stable bound states has remained one of the most stubborn puzzles in quantum chromodynamics. Traditionally, scientists analyzed each charmonium state independently, ignoring how heavier states decay into lighter ones and muddying the resulting calculations. To solve this limitation, researchers Yubo Liu and Rongrong Ma created a unified statistical approach available on the arXiv preprint server that handles all parent production and daughter decay chains simultaneously.

How Joint Bayesian Inference Untangles Subatomic Decays

The research team employed a statistical method called Bayesian inference to calculate the probability distributions of nine long-distance matrix elements, which describe the non-perturbative transition phases where quark pairs settle into observable mesons. Using PYTHIA 8.317 software, the authors simulated roughly 1.5 billion charmonium production events for each collision energy studied, covering proton-proton collisions with center-of-mass energies ranging from 5.02 to 13 TeV. By tracing the ancestry of every particle decay, the model accounted for feed-down effects, meaning situations where a lighter particle is actually the debris of a heavier parent particle decaying shortly after creation.

When tested against the massive LHC dataset, the joint framework achieved an overall quality metric of $\chi^2/\mathrm{ndf} = 1.18$, indicating an exceptionally accurate description across several orders of magnitude in cross section. In contrast, standard default simulation settings overpredicted $J/\psi$ production by up to sixty percent and missed other states by factors of two or three. The authors also generated a positivity-constrained version of their parameters specifically designed to feed directly into standard particle physics Monte Carlo event generators.

What the Parameter Correlations Reveal About Quark Binding

The statistical analysis uncovered deep mathematical links between different production channels. Most parameter pairs showed weak correlations, but three specific pairs exhibited strong anticorrelations. For example, the color-octet $1S_0^{[8]}$ and $3P_0^{[8]}$ channels for both $J/\psi$ and $\psi(2S)$ produced nearly identical transverse momentum shapes in the fitted kinematic region. Because their individual contributions compensate for one another, the underlying data strictly constrain only their combined sum rather than their individual values.

This overlapping behavior explains why unconstrained fits sometimes yield negative values for certain matrix elements, a mathematical feature that is physically acceptable in quantum field theory but requires careful handling. When the researchers forced all parameters to remain positive, the overall fit quality only degraded slightly, proving that the model remains robust even under strict physical boundaries.

Testing Predictions at Lower Energy Colliders

Armed with these newly extracted parameters, the authors calculated fresh cross-section predictions for prompt $J/\psi$ and $\psi(2S)$ production at a center-of-mass energy of 200 GeV, matching the operating profile of the Relativistic Heavy Ion Collider (RHIC). Because no RHIC data was used to train the model, these calculations serve as a rigorous test of universality, checking whether the same fundamental rules govern quark binding across vastly different collision energies.

While existing high-momentum measurements at RHIC still contain contamination from beauty-quark hadron decays, upcoming high-precision tracking upgrades at detectors like sPHENIX are expected to isolate prompt charmonium production cleanly. The research team emphasizes that their Bayesian framework is fully modular, meaning future theoretical improvements like next-to-leading-order calculations or polarization observables can be easily integrated without rewriting the core statistical engine.

The study was posted to the arXiv preprint server on November 5, 2026.