The energy dependence of exclusive heavy vector meson… — Plain-Language Explanation

Imagine the universe is filled with a thick, invisible fog made of tiny particles called gluons. These gluons are the "glue" that holds the building blocks of matter (protons and neutrons) together. Usually, this fog is thin and easy to see through. But when you smash particles together at incredibly high speeds, like in the Large Hadron Collider (LHC), you create a situation where the fog becomes so incredibly dense that it starts to behave strangely. It's like trying to pack a stadium full of people into a single room; eventually, they can't move freely anymore. This state is called gluon saturation.

The paper you provided is a scientific investigation trying to figure out: "Is this fog actually getting dense enough to saturate, or is it just a very thick, but still normal, fog?"

Here is how the authors tackled this mystery, explained simply:

The Experiment: Taking a Snapshot

The scientists looked at a specific process called exclusive photoproduction. Imagine a photon (a particle of light) zooming in and hitting a proton (a tiny particle inside an atom) or a lead nucleus (a heavy atom). The photon hits, and for a split second, it turns into a heavy "meson" (a particle made of a heavy quark and its anti-particle, like a J/ψ or an Υ).

The J/ψ is like a medium-heavy particle.
The Υ (Upsilon) is like a very heavy particle.

By measuring how often these particles are created at different energy levels, the scientists can learn how the "gluon fog" behaves.

The Two Theories: The "Empty Room" vs. The "Crowded Room"

To understand the data, the scientists used two different mental models (mathematical frameworks):

The "Empty Room" Model (BFKL Evolution): This model assumes the gluon fog is still thin enough that the particles don't really bump into each other. They just pass through. This is the "low density" theory.
The "Crowded Room" Model (Non-linear QCD): This model assumes the fog is so dense that the particles are jamming each other, slowing down the growth of the fog. This is the "saturation" theory.

The goal was to see if the "Empty Room" model could explain the data. If it failed, it would be strong proof that the "Crowded Room" (saturation) is real.

The Method: Starting with a Map

The scientists couldn't just guess where the fog started. They needed a "map" of the fog at a specific point in time (called initial conditions). They used two different maps to start their journey:

Map A (IP-Sat): A complex map that assumes the lead nucleus acts like a collection of individual people (nucleons) crowded together.
Map B (BGK with A¹/³ scaling): A simpler map that treats the lead nucleus as one giant, scaled-up version of a single proton.

They then ran their "Empty Room" simulation (NLO BFKL evolution) forward in time to see if it matched what the LHC actually observed.

The Results: What Worked and What Didn't

1. The Proton Test (The Small Target)
When they tested their simulation on a single proton, the "Empty Room" model (BFKL) did a decent job. It predicted the energy dependence reasonably well, though it was a bit shaky at the very highest energies. This was expected because the proton is small, and the fog isn't as dense there.

2. The Lead Test (The Big Target)
This is where things got interesting.

Using Map A (IP-Sat): When they assumed the lead nucleus was a crowd of individual nucleons, the "Empty Room" model completely failed. It predicted way too many particles were being created. It was like predicting that a crowded stadium would behave exactly like an empty one—it just didn't make sense.
Using Map B (BGK A¹/³): When they treated the lead nucleus as a single, scaled-up object, the "Empty Room" model worked surprisingly well. It matched the data almost perfectly, even for the nuclear modification factor (a ratio that cancels out many errors).

The Big Conclusion

The paper concludes with a few key takeaways:

The "Crowded Room" isn't strictly necessary yet: Surprisingly, the "Empty Room" model (which assumes no saturation) could actually describe the data if you started with the right map (the A¹/³ scaled model). This suggests that we might not need to invoke complex "saturation" physics to explain the current data; the standard "low density" math works if we treat the heavy nucleus as a single, scaled-up unit.
The Shape of the Nucleus Matters: The fact that the "individual nucleon" map failed while the "scaled-up proton" map worked suggests that inside a heavy nucleus, the gluons aren't just sitting in individual cells; they are behaving more like a unified, scaled-up cloud.
The Υ Particle is the Key: The heavier particle (Υ) gave much clearer results than the lighter one (J/ψ). Because it is heavier, it acts like a sharper probe, cutting through the noise and giving a clearer picture of the underlying physics.

In a Nutshell

The authors tried to prove that gluon saturation (a "traffic jam" of particles) is happening. They used a math tool that assumes no traffic jam.

When they treated the heavy nucleus as a crowd of individuals, the math broke.
When they treated the heavy nucleus as one giant, scaled-up blob, the math worked perfectly.

This implies that while we are seeing signs of how heavy nuclei behave, we might not need to invent new "traffic jam" physics just yet to explain the current data. The standard rules work, provided you view the heavy nucleus as a single, scaled-up entity rather than a pile of separate parts.

Technical Summary: Energy Dependence of Exclusive Heavy Vector Meson Photoproduction and NLO BFKL Evolution

Problem Statement
Exclusive photoproduction of heavy vector mesons ( $J/\psi$ and $\Upsilon$ ) in ultraperipheral collisions at the LHC serves as a probe for Quantum Chromodynamics (QCD) at ultra-small Bjorken- $x$ . While perturbative QCD predicts a power-like rise in gluon distributions via the Balitsky-Fadin-Kuraev-Lipatov (BFKL) equation, unitarity bounds suggest this rise must eventually slow down due to gluon saturation. The central challenge is distinguishing whether observed data can be described by linear evolution equations assuming low gluon densities or if non-linear QCD dynamics (such as those described by JIMWLK-BK equations) are required. Previous attempts to test this using collinear factorization (DGLAP) face limitations at low hard scales, while non-linear evolution schemes can describe data in both low and high-density regimes, making it difficult to isolate saturation effects. This study aims to use Next-to-Leading Order (NLO) BFKL evolution as a benchmark for a low-density framework to test its validity against experimental data for both protons and lead nuclei.

Methodology
The authors employ a two-step approach to model the energy dependence of the exclusive photoproduction cross-sections:

Initial Conditions: The study generates unintegrated gluon distributions at an initial scale $x_0 = 0.01$ using dipole models rather than fitting directly to collider data.
- For the proton, the authors utilize the recently refitted Bartels-Golec-Biernat-Kowalski (BGK) dipole model.
- For the lead nucleus ($Pb$), two distinct initial condition scenarios are tested:
  - The Impact Parameter dependent Saturation (IP-Sat) model, which averages over nucleon positions using a Wood-Saxon distribution.
  - A scaled BGK model where the effective saturation scale is rescaled by a factor of $A^{1/3}$ , treating the nucleus as a scaled-up proton.
Evolution: The initial distributions are evolved to lower $x$ values using a solution to the NLO BFKL equation. This solution includes complete NLO corrections and a resummation of collinear-enhanced logarithms, following the formalism established in previous works [46, 47]. The evolution is restricted to zero momentum transfer ( $t=0$ ).
Cross-Section Calculation: The scattering amplitude is calculated via a convolution of the photon-vector meson impact factor (using boosted Gaussian wave functions) and the evolved unintegrated gluon distribution. The total cross-section is derived by integrating the differential cross-section at $t=0$ using a diffractive slope parameter $B_D$ determined from data.
Stability Criteria: To ensure the perturbative validity of the NLO BFKL solution, the authors impose two criteria:
- The magnitude of the NLO correction to the scattering amplitude must remain below 50% of the leading order term at the lowest observed $x$ .
- The scattering amplitude must maintain a positive energy slope ( $\lambda > 0$ ).

Key Contributions and Results

Proton Case: The NLO BFKL evolution, initialized with the BGK model, provides a good description of the energy dependence for both $J/\psi$ and $\Upsilon$ photoproduction on protons. While the central values slightly overshoot data at high energies ( $W > 500$ GeV), the results remain within theoretical uncertainties. This confirms that NLO BFKL is a viable framework for predicting energy dependence in the perturbative regime without initial fits to collider data.
Nuclear Case ( $J/\psi$ on Lead):
- IP-Sat Initial Conditions: When initial conditions are generated via the IP-Sat model, the NLO BFKL prediction fails to describe the data. The predicted energy dependence is too strong, overshooting experimental measurements significantly.
- $A^{1/3}$ Scaled BGK Initial Conditions: Conversely, when initial conditions are generated using the $A^{1/3}$ scaled BGK model, the NLO BFKL evolution provides a very good description of the nuclear modification factor ( $S_{Pb}$ ) and the energy dependence of the cross-section across the measured range.
Nuclear Case ( $\Upsilon$ on Lead): Similar trends are observed for $\Upsilon$ production, though uncertainties are reduced due to the higher hard scale. The IP-Sat based prediction again shows a stronger energy growth than observed, while the $A^{1/3}$ scaled model aligns better with data, particularly for the nuclear modification factor.

Significance and Claims
The paper claims that the success of the NLO BFKL evolution with $A^{1/3}$ scaled initial conditions, contrasted with the failure of the IP-Sat based approach, suggests that for the energy dependence of these processes, the nuclear gluon density per transverse area scales approximately as $A^{1/3}$ .

Crucially, the authors argue that their results indicate that the current energy dependence of the data does not strictly require the inclusion of non-linear terms (gluon saturation) in the evolution equations to be described. The linear NLO BFKL evolution, provided the correct initial conditions are chosen, is sufficient to match the data. However, the authors remain modest in their conclusions, noting:

The successful description relies on a negative NLO perturbative correction; if the renormalization scale were chosen such that this term were small or positive, the description would fail.
This does not completely rule out non-linear dynamics, but implies that if such terms exist, their magnitude must be similar for both proton and lead targets to preserve the observed scaling.
The study highlights the sensitivity of the nuclear modification factor to the choice of initial conditions, making it a robust observable for distinguishing between different saturation models.

The paper concludes that while NLO BFKL evolution can describe the data without explicit non-linear terms, future measurements of $\Upsilon$ production on lead and vector meson production at the future Electron-Ion Collider (EIC) are necessary to further constrain initial conditions and definitively test the onset of gluon saturation.

The energy dependence of exclusive heavy vector meson photoproduction cross-sections and NLO BFKL evolution