3D Initial-State Dynamics across scales: A Comparative Study of saturation and string-based descriptions

The Gist

Imagine two giant, ultra-dense trains (atomic nuclei) smashing into each other at nearly the speed of light. When they collide, they don't just bounce off; they create a tiny, super-hot fireball of matter called the Quark-Gluon Plasma (QGP). This is the state of matter that existed just after the Big Bang.

To understand what happens in this fireball, scientists need to know exactly how the "ingredients" (energy, protons, and electric charge) are distributed right at the moment of impact. This is called the Initial State.

This paper compares two different "recipes" or computer models that scientists use to predict this initial state. The authors want to see which recipe works better, especially in the tricky middle ground of collision energies where neither recipe is perfect.

Here is a breakdown of the two models and what the study found, using simple analogies:

The Two Competing Recipes

1. The "String" Model (SMASH)

• The Analogy: Imagine the colliding nuclei are like two bundles of tangled rubber bands. When they crash, these rubber bands stretch, snap, and turn into new particles (hadrons).
• How it works: This model is based on hadronic transport. It treats the collision as a series of individual particle interactions and "string" excitations (like stretching rubber bands). It works very well for lower-energy crashes where the particles behave more like solid objects bumping into each other.
• The Flaw: At very high speeds, this model struggles. It tends to keep too many "heavy" particles (baryons) stuck in the middle of the crash, whereas experiments show they should fly further apart.

2. The "Saturation" Model (McDipper)

• The Analogy: Imagine the nuclei are like dense clouds of fog made of invisible glue (gluons). When they collide, the fog gets so thick and "saturated" that it behaves like a single, fluid sheet rather than individual drops.
• How it works: This model is based on Color Glass Condensate (CGC) theory. It assumes that at high speeds, the particles inside the nuclei are so packed together that they act like a unified wave of energy. It excels at high-energy collisions (like those at the Large Hadron Collider).
• The Flaw: It might be too simplified for lower energies where individual particle interactions matter more.

The Experiment: A Race Across Speeds

The authors ran simulations of these two models across a wide range of collision speeds, from "moderate" (62.4 GeV) to "ultra-fast" (5.02 TeV). They looked at three main things being deposited into the collision zone:

1. Transverse Energy: How much heat/energy is created sideways.
2. Baryon Number: How many protons/neutrons get stopped in the middle.
3. Electric Charge: How the electric charge is distributed.

The Findings

1. At Low Speeds (The Middle Ground):

• The Result: Both models agreed reasonably well. They produced similar amounts of energy in the center of the collision.
• The Takeaway: There is an "overlap zone" where both the "rubber band" (string) and "fog" (saturation) recipes give similar answers. This is a good sign for scientists studying intermediate energies.

2. At High Speeds (The Breakup):

• The Result: The models started to disagree significantly.
  • Energy: The "Fog" model (McDipper) predicted much more energy than the "Rubber Band" model (SMASH). This makes sense because at high speeds, the "glue" (gluons) becomes the dominant force, which the Fog model captures better.
  • Stopping Power (Baryons): This was the biggest difference. The "Rubber Band" model (SMASH) kept too many protons stuck in the middle of the crash. It acted like a traffic jam that wouldn't clear. The "Fog" model (McDipper) correctly predicted that at high speeds, these protons should fly further out, leaving the center emptier.

3. The Shape of the Fireball:

• Surprisingly, despite these huge differences in how the energy and particles were distributed, both models predicted a very similar shape for the fireball's initial geometry (specifically, how elliptical or triangular it was).
• The Analogy: Think of two different chefs making a cake. One uses a sponge recipe, the other a flour recipe. They might use very different ingredients and mixing techniques, but if they both aim for a round cake, the final shape looks the same. The authors found that the overall shape of the collision is mostly determined by the size and angle of the crash, not the tiny details of the recipe.

The "Why" Behind the Failure

The paper digs into why the "Rubber Band" model (SMASH) fails at high speeds.

• The Issue: In the SMASH model, when a "leading" particle (a piece of the original train that flies forward) is created, the model gives it a special "pass" to interact immediately, even before it fully forms.
• The Consequence: This causes these leading particles to crash into other incoming particles too early, effectively acting like a wall that stops them from flying away. This creates a "traffic jam" of protons in the middle that doesn't match reality.

The Conclusion

• For Low/Medium Energies: Both models are useful and give similar results.
• For High Energies: The "Saturation" model (McDipper) is superior. It correctly handles the physics of high-speed gluon clouds and predicts that protons should fly further out, rather than getting stuck in the middle.
• The Shape Factor: Regardless of the recipe, the overall geometric shape of the collision remains surprisingly consistent between the two models.

In short: If you are studying a slow crash, you can use either model. If you are studying a high-speed crash, you should use the "Saturation" model because the "String" model keeps the particles stuck in the middle when they should be flying apart. The authors also suggest that future experiments need to look more closely at the "edges" of the crash (forward and backward regions) to understand exactly how these particles stop or fly away.

Technical Summary

Technical Summary: 3D Initial-State Dynamics across scales: A Comparative Study of saturation and string-based descriptions

Problem Statement
The initial state of ultra-relativistic heavy-ion collisions remains a primary source of theoretical uncertainty in heavy-ion phenomenology. While standard modeling often assumes longitudinal boost invariance (Bjorken picture), deviations are expected even at mid-rapidity, necessitating 3+1D hydrodynamical simulations. A significant gap exists in the intermediate collision energy regime ($\sqrt{s_{NN}} \sim 10\text{--}200$ GeV), where neither purely partonic descriptions (valid at LHC energies) nor purely hadronic transport approaches (valid at HADES energies) provide a complete picture. Furthermore, the longitudinal deposition of conserved quantities (transverse energy, net-baryon number, and electric charge) and the resulting initial-state eccentricities are insufficiently constrained across a broad energy range.

Methodology
The authors perform a comparative study of two initial-state models with fundamentally different microscopic assumptions and dynamical mechanisms:

1. McDipper: A saturation-based model rooted in the Color Glass Condensate (CGC) effective description of QCD. It utilizes the $k_T$-factorization limit and leading-order (LO) single parton production formulas. Energy and charge densities are computed as moments of single-inclusive gluon and quark distributions, derived from the convolution of dipole correlators (using the IP-Sat saturation model) and collinear parton distribution functions (PDFs). This approach naturally links light-cone momentum fractions to rapidity.
2. SMASH (SMASH-IC): A string-based approach utilizing the hadronic transport code SMASH. It propagates hadrons based on the relativistic Boltzmann equation, employing Pythia to model inelastic interactions via string fragmentation. Initial conditions are constructed by defining a hypersurface of constant proper time ($\tau_{sw} = 0.5$ fm) and distributing conserved quantities from the particle list onto a grid using a Lorentz-contracted Gaussian smearing kernel.

The study analyzes nucleus-nucleus collisions (Au-Au and Pb-Pb) across a wide range of center-of-mass energies ($\sqrt{s_{NN}} = 62.4, 200, 5020$ GeV) and impact parameters ($b = 2, 5, 9$ fm). The comparison focuses on:

• The longitudinal deposition of transverse energy ($dE_T/d\eta_s$).
• The deposition of net-baryon and electric charge.
• The calculation of initial-state eccentricity coefficients ($\epsilon_n$) in the transverse plane.

Key Contributions and Results

• Energy Deposition: At lower energies ($\sqrt{s_{NN}} = 62.4$ GeV), both models produce similar mid-rapidity energy densities, suggesting an overlapping region of applicability. However, the longitudinal profiles differ: McDipper exhibits a broader profile extending further to forward/backward rapidities, while SMASH shows a sharper depletion. As collision energy increases, McDipper deposits substantially more transverse energy than SMASH, reflecting the importance of gluon and small-$x$ dynamics inherent in the saturation framework but absent in the string-based model.
• Baryon Stopping and Charge Deposition: The models diverge significantly in the deposition of conserved charges. SMASH retains more baryons at mid-rapidity than McDipper. In the high-energy regime ($\sqrt{s_{NN}} = 5.02$ TeV), SMASH exhibits a plateau of nearly constant baryon density rather than the expected quasi-vanishing deposition. This behavior is attributed to the treatment of "leading hadrons" in the Pythia string fragmentation within SMASH; these hadrons possess reduced but non-zero cross-sections during their formation time, allowing them to interact immediately with incoming nucleons and stop baryons regardless of collision energy. In contrast, McDipper follows the expected exponential decrease in baryon stopping with increasing rapidity shift, consistent with experimental data and saturation physics.
• Eccentricities: Despite pronounced differences in the microscopic energy-density structure (McDipper showing granular hot spots vs. SMASH's smoother distributions), both models yield similar eccentricity coefficients ($\epsilon_2, \epsilon_3$). This suggests that initial-state eccentricities are governed primarily by global geometrical features of the collision rather than the detailed local structure of energy deposition. However, SMASH eccentricities show an artificial increase at high rapidities due to particle depletion, whereas McDipper maintains a concave profile consistent with partonic physics.

Significance and Claims
The paper claims to identify the respective domains of applicability for string-based and saturation-based initial-state models. It demonstrates that while SMASH works reasonably well at lower energies, it breaks down in the ultra-relativistic regime regarding baryon stopping and total energy deposition. Conversely, the McDipper model, based on gluon saturation, provides a consistent description of baryon stopping and energy deposition across a large energy range, including the intermediate regime where effective degrees of freedom are less clearly separated.

The authors emphasize that the modeling of string excitations and the specific treatment of leading hadrons are crucial for determining baryon stopping in hadronic transport approaches. They conclude that experimental data in the forward/backward fragmentation regions is necessary to further constrain these dynamics. The study does not claim to resolve the final-state observables but highlights that the differences in initial-state dynamics identified here will impact the subsequent hydrodynamic and hadronic evolution, warranting further investigation within a complete hybrid framework.