The shape of transverse momentum spectra in hybrid hydrodynamic models

This study reveals that despite a large parameter space in hybrid hydrodynamic models, the shape of scaled transverse momentum spectra exhibits surprisingly limited flexibility and significant tension with momentum-integrated observables, particularly regarding the nucleon width parameter $w$ and bulk viscosity.

The Gist

The Big Picture: Smashing Atoms to Find the "Perfect Recipe"

Imagine you are a master chef trying to recreate a famous, complex dish (like a perfect soufflé) that you've only seen in a photo. You know the ingredients (quarks and gluons) and the basic cooking method (heat and pressure), but you don't know the exact amounts, the oven temperature, or how long to let it rise.

In the world of physics, scientists do this with heavy-ion collisions. They smash heavy atoms (like lead) together at nearly the speed of light to create a tiny, super-hot drop of liquid called the Quark-Gluon Plasma (QGP). This is the state of matter that existed just after the Big Bang.

The scientists in this paper are trying to figure out the "perfect recipe" for this cosmic soup. They use a massive computer simulation (a "hybrid model") that has 17 different knobs they can turn. These knobs control things like:

• How big the initial "dough" is.
• How sticky the liquid is (viscosity).
• How long the dough rises before it hits the oven.

The New Tool: The "Scaled Spectrum"

Usually, to check if their recipe is right, scientists look at two main things:

1. How many particles came out? (The total yield).
2. How fast were they moving on average? (The average momentum).

However, this paper introduces a new way to look at the data called the Scaled Transverse Momentum Spectrum.

The Analogy:
Imagine you have a bag of marbles of different sizes.

• The old way: You count how many marbles you have and measure the average size.
• The new way (Scaled Spectrum): You take the average size of the marbles in your bag and use that as a "ruler." Then, you measure every single marble relative to that ruler.

By doing this, the scientists found something amazing: No matter how they changed the collision (big or small, central or edge-on), the shape of the marble distribution looked almost exactly the same. It was "universal." It's like if every time you baked a cake, regardless of the flour brand or oven temp, the crumb structure looked identical.

The Investigation: Turning the Knobs

The team asked: "If this shape is so universal, which of our 17 recipe knobs actually control it?"

They used a statistical method (like a super-smart detective) to test every knob. They found that only four of the 17 knobs really mattered for this specific shape:

1. The "Free-Streaming" Time: How long the particles zoom around freely before they start bumping into each other like a crowded dance floor.
2. Bulk Viscosity: How much the liquid resists being squeezed or expanded (like the thickness of honey vs. water).
3. The "Peak" Temperature: The specific temperature where the liquid gets thickest.
4. Nucleon Width: How "fuzzy" or "sharp" the edges of the initial atoms are.

The Surprise: Even though they had 17 knobs to play with, the model was surprisingly rigid. It couldn't easily change the shape of the spectrum. The "universal shape" is a very strict rule that the physics of the early universe seems to follow.

The Conflict: The "Tension"

Here is where the plot thickens. The scientists tried to tune their recipe to match the Scaled Spectrum (the shape).

• Result: To get the shape right, they had to turn the "Nucleon Width" knob to make the initial atoms very small and "grainy" (like coarse sand).

Then, they tried to tune the recipe to match the Average Speed (the old way).

• Result: To get the speed right, they had to turn the "Nucleon Width" knob to make the atoms very smooth and large (like fine flour).

The Problem: They cannot have both. The model is like a car that can drive fast or steer well, but not both at the same time. If they tune the car to steer perfectly (match the shape), it drives too slow (wrong average speed). If they tune it to drive fast, it steers poorly.

The Conclusion: Missing Ingredients

The paper concludes that our current "recipe" for the universe is missing something.

The fact that the model fails to describe both the shape and the speed simultaneously suggests that we are missing a piece of physics.

The Analogy:
Imagine you are trying to bake a cake. You have the flour, sugar, and eggs. But no matter how you mix them, the cake never rises right. You realize, "Oh! I forgot the baking powder!"

In this case, the scientists suspect the missing "baking powder" might be non-equilibrium dynamics (specifically related to Goldstone modes). This is a fancy way of saying that when the particles are first created, they aren't perfectly balanced yet, and this imbalance creates extra particles at low speeds that our current models ignore.

Summary

1. The Discovery: The shape of particle speeds in these collisions is surprisingly universal and rigid.
2. The Test: Scientists tested 17 variables and found only 4 control this shape.
3. The Conflict: The model can match the shape OR the average speed, but not both at the same time.
4. The Takeaway: Our current understanding of the "cosmic soup" is incomplete. We likely need to add new physics to the recipe to make the whole picture fit together.

Technical Summary

1. Problem Statement

In relativistic heavy-ion collisions, the Quark-Gluon Plasma (QGP) is characterized using hybrid hydrodynamic models that simulate the evolution from initial conditions through a viscous fluid phase to a hadronic afterburner. While these models successfully describe $p_T$-integrated observables (such as particle multiplicity $N_{ch}$ and mean transverse momentum $\langle p_T \rangle$) and anisotropic flow coefficients ($v_n$), the full shape of the transverse momentum ($p_T$) spectra contains complementary information sensitive to late-stage hadronic interactions and the QGP-hadron transition.

A specific observable, the scaled transverse momentum spectrum $U(x_T)$, defined as:
$$U(x_T) \equiv \frac{\langle p_T \rangle}{N} \frac{dN}{dp_T} = \frac{1}{N} \frac{dN}{dx_T}, \quad \text{where } x_T = \frac{p_T}{\langle p_T \rangle}$$
has been observed to exhibit a surprising universality across different collision systems (Pb-Pb, Xe-Xe, p-Pb) and centralities. However, it remains unclear:

1. Which specific model parameters govern this universal shape?
2. Whether current state-of-the-art hybrid models can simultaneously reproduce this scaled shape and traditional $p_T$-integrated observables.
3. How sensitive this observable is to the specific prescription used for particlization (converting fluid to particles), particularly regarding non-equilibrium corrections ($\delta f$).

2. Methodology

The authors employed a comprehensive Bayesian analysis framework using the JETSCAPE hybrid simulation model. The study involved the following steps:

• Simulation Framework: The model chain includes:
  • Initial Conditions: TRENTo parametric model (controlled by nucleon width $w$, generalized mean $p$, etc.).
  • Pre-equilibrium: Free-streaming evolution.
  • Hydrodynamics: MUSIC (second-order viscous hydrodynamics) with temperature-dependent shear ($\eta/s$) and bulk ($\zeta/s$) viscosities.
  • Particlization: Cooper-Frye prescription on an isothermal switching surface ($T_{sw}$).
  • Afterburner: SMASH kinetic transport code.
• Particlization Variations: The study systematically compared four different models for the non-equilibrium correction $\delta f$ to the particle distribution:
  1. Grad (14-moments): Linear expansion in momentum.
  2. Chapman-Enskog (CE): Linearized expansion in relaxation time approximation.
  3. PTM (Pratt-Torrieri-McNelis): Exponentiated ansatz.
  4. PTB (Pratt-Torrieri-Bernhard): Exponentiated ansatz with different scaling.
• Statistical Tools:
  • Gaussian Process (GP) Emulation: To handle the high computational cost of the full simulator, the authors trained GP emulators on 318–321 design points (Latin Hypercube Sampling) for each $\delta f$ model.
  • Dimensionality Reduction: Principal Component Analysis (PCA) was applied to the scaled spectra ($U(x_T)$) across 7 centrality bins and 41 momentum bins. The first 6 principal components captured >98% of the variance, allowing efficient emulation.
  • Bayesian Inference: Used to calibrate the model against experimental data (ALICE Pb-Pb at 2.76 TeV) and derive posterior distributions for 17 model parameters.
  • Global Sensitivity Analysis (GSA): Sobol indices were calculated to quantify the influence of each parameter on the shape of $U(x_T)$.

3. Key Contributions

• First Systematic Calibration of Scaled Spectra: This work is the first to include the dimensionless scaled spectrum $U(x_T)$ in a global Bayesian calibration of hybrid hydrodynamic models, moving beyond $p_T$-integrated quantities.
• Quantification of Model Tension: The study rigorously demonstrates a tension between calibrating to scaled spectra versus calibrating to $p_T$-integrated observables. The two strategies favor mutually exclusive regions of the parameter space.
• Sensitivity Mapping: It identifies the specific physical mechanisms (initial state granularity vs. bulk viscosity) that control the spectral shape, distinguishing them from those controlling integrated yields.
• Robustness Check: It evaluates the impact of different particlization prescriptions ($\delta f$ models) on the universal scaling, finding that the shape is largely robust to these choices, with the exception of the PTB model.

4. Key Results

A. Sensitivity to Model Parameters

The Global Sensitivity Analysis revealed that the shape of $U(x_T)$ is primarily controlled by four parameters:

1. Free-streaming time scale ($\tau_R$): Governs pre-equilibrium dynamics.
2. Maximum bulk viscosity ($(\zeta/s)_{max}$): Regulates entropy production near the crossover.
3. Bulk viscosity peak temperature ($T_\zeta$): Determines where bulk effects are strongest.
4. Nucleon width ($w$): Controls the initial state granularity (effective size of nucleons in TRENTo).

B. The "Tension" in Simultaneous Description

A critical finding is the inability of the current model to simultaneously describe $U(x_T)$ and $\langle p_T \rangle$:

• Scaled Spectra ($U(x_T)$): Prefer a small nucleon width ($w \approx 0.6$ fm), implying a more granular, fluctuating initial state.
• Mean Transverse Momentum ($\langle p_T \rangle$): Prefers a large nucleon width ($w \approx 1.0$ fm), implying smoother initial profiles.
• Consequence: When the model is calibrated to $U(x_T)$, it fails to reproduce the experimental $\langle p_T \rangle$ (overestimating it), and vice versa. This suggests that the current "TRENTo + free-streaming + hydro + afterburner" framework is missing physics.

C. Impact of Particlization ($\delta f$) Models

• Universality: The overall shape and universality of $U(x_T)$ are remarkably robust against the choice of $\delta f$ (Grad, CE, PTM). This suggests the shape is an emergent property of collective expansion rather than microscopic freeze-out details.
• PTB Anomaly: The PTB (exponentiated) model showed distinct behavior, with larger posterior uncertainties and stronger sensitivity to bulk viscosity parameters, likely due to its non-linear functional form. However, it did not resolve the tension with $\langle p_T \rangle$.

D. Fit Quality

Even after Bayesian calibration, the model predictions for $U(x_T)$ show systematic deviations from ALICE data:

• Underestimation at low $x_T$.
• Overestimation at intermediate $x_T$.
• Underestimation at high $x_T$.
  This indicates that the "universal" shape observed in data is not perfectly captured by current hydrodynamic evolution, even with optimized parameters.

5. Significance and Conclusion

This paper establishes the scaled transverse momentum spectrum $U(x_T)$ as a powerful, complementary observable for constraining heavy-ion collision models. Its key significance lies in:

1. Revealing Missing Physics: The persistent tension between $U(x_T)$ and $\langle p_T \rangle$ suggests that current models lack specific physical mechanisms. The authors speculate that non-equilibrium Goldstone mode dynamics (enhancing low-$p_T$ pion yields) might be the missing ingredient required to resolve this discrepancy.
2. Refining Initial State Constraints: The study proves that $U(x_T)$ is uniquely sensitive to initial state granularity ($w$), offering a new constraint that is orthogonal to traditional flow and multiplicity measurements.
3. Model Validation: It demonstrates that while hydrodynamic universality is a robust feature, the specific details of the freeze-out ($\delta f$) have a limited impact on the global spectral shape, reinforcing the dominance of collective dynamics.

In summary, the work highlights that while hybrid models are successful, they are not yet complete. The scaled spectrum serves as a sensitive probe that exposes limitations in the current theoretical description of the QGP-hadron transition and initial state geometry.