⚛️ nuclear theory

A Resummed Hydrodynamic Description of Relativistic Heavy-ion Collisions

This paper introduces a resummed hydrodynamic scheme that enforces non-linear causality conditions to constrain viscous stress tensors within tunable limits, demonstrating through event-by-event simulations of Pb+Pb and p+Pb collisions that this approach quantifies theoretical uncertainties in final-state flow observables while recovering standard second-order theories in the small-viscosity limit.

Original authors: Cheng Chiu, Gabriel Denicol, Matthew Luzum, Chun Shen

Published 2026-09-30

📖 3 min read🧠 Deep dive

CC BY 4.0

Original authors: Cheng Chiu, Gabriel Denicol, Matthew Luzum, Chun Shen

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

When scientists smash heavy atomic nuclei together at nearly the speed of light, they create a fleeting, super-hot soup of subatomic particles known as the quark-gluon plasma. This state of matter existed only microseconds after the Big Bang, before the universe cooled enough for protons and neutrons to form. To understand how this primordial soup behaves, physicists treat it not as a gas of individual particles, but as a fluid that flows and expands. This approach, called relativistic viscous hydrodynamics, has been remarkably successful at describing the large, chaotic collisions of heavy ions. However, a problem arises when the collision is smaller or the initial conditions are more extreme. In these scenarios, the fluid can be pushed so far from a calm, balanced state that the standard mathematical rules used to describe its flow begin to break down, potentially predicting impossible speeds or unstable behavior.

A new study by Cheng Chiu, Gabriel Denicol, Matthew Luzum, and Chun Shen addresses this breakdown by introducing a refined way to calculate how this exotic fluid evolves. The researchers developed a "resummed" hydrodynamic scheme, a method that adjusts the fluid's internal resistance to flow based on how far it is from equilibrium. In standard models, the equations governing the fluid's stress and pressure can grow without limit, leading to mathematical errors when gradients become too steep. The new approach acts as a built-in safety valve: as the fluid becomes more turbulent and distant from a calm state, the method automatically dampens the effects of viscosity, effectively turning the fluid into a more ideal, frictionless flow to prevent the equations from violating the fundamental rule that nothing can travel faster than light.

The team tested this new framework by running thousands of computer simulations of collisions between lead nuclei and between protons and lead nuclei at an energy of 5.02 TeV, a standard high-energy setting at the Large Hadron Collider. They compared their new method against two other approaches: a standard calculation that simply cuts off unrealistic numbers when they appear, and a method that strictly enforces the rules of causality by manually reducing viscous effects whenever they threaten to break the laws of physics. The results showed that for large, central collisions where the system has plenty of time to settle down, all three methods produced nearly identical results for the particles detected at the end. This suggests that the new method is consistent with established physics when the system is behaving well.

However, the differences became significant when the simulations started with the fluid in a highly disturbed, far-from-equilibrium state, which is more likely to happen in smaller collisions or the very earliest moments of a large one. In these extreme cases, the new resummed scheme predicted that the fluid would evolve more like an ideal, frictionless liquid. This led to stronger outward flows and different patterns in the final distribution of particles compared to the other methods. The researchers found that for small collision systems, the choice of mathematical framework introduces a substantial uncertainty in the predicted outcomes. This implies that our current understanding of how the quark-gluon plasma behaves in its most violent, early moments is still incomplete. By providing a flexible tool that can impose these necessary physical limits without breaking the equations, the study offers a way to better quantify these uncertainties and move closer to a precise description of the universe's first moments.

Technical Summary: A Resummed Hydrodynamic Description of Relativistic Heavy-ion Collisions

Problem Statement
Relativistic viscous hydrodynamics is the standard effective framework for describing the macroscopic dynamics of the quark-gluon plasma (QGP) created in heavy-ion collisions. However, formulating a causal relativistic viscous fluid dynamics theory remains a significant challenge, particularly when systems evolve far from thermal equilibrium. In conventional second-order theories (such as DNMR), the equations of motion allow viscous stress tensors to grow indefinitely in the presence of large velocity gradients. This unbounded growth can lead to violations of causality conditions, especially in small collision systems (e.g., p+Pb) where strong pressure gradients drive the system far from equilibrium. While non-linear causality conditions have been derived, standard numerical implementations often lack mechanisms to strictly enforce these bounds during evolution, leading to potential theoretical uncertainties in precision comparisons with experimental flow measurements.

Methodology
The authors propose a new phenomenological "resummed" hydrodynamic scheme to evolve viscous stress tensors while enforcing necessary non-linear causality conditions. The core methodology involves:

Renormalization of Transport Coefficients: Instead of treating transport coefficients (shear viscosity $\eta$ , bulk viscosity $\zeta$ , and various second-order coefficients) as constants, the scheme promotes them to be functions of the local inverse Reynolds numbers ( $\tilde{R}_\pi$ for shear and $\tilde{R}_\Pi$ for bulk).
Resummation Mechanism: This functional dependence is interpreted as resumming an infinite set of high-order gradient terms. A specific renormalization factor, $f$ , is introduced to scale the transport coefficients on the right-hand side of the evolution equations. The factor is defined as:
$f \equiv \frac{1}{1 + \text{arctanh}^2(\alpha(\tilde{R}_\Pi + \tilde{R}_\pi))}$
where $\alpha$ is a tunable parameter.
Causality Enforcement: The arctanh function is only defined for arguments within $[-1, 1]$ . Consequently, as the sum of inverse Reynolds numbers approaches $1/\alpha$ , the factor $f$ approaches zero. This effectively suppresses the viscous terms, causing the viscous stress tensors to decay exponentially (reducing the system to ideal hydrodynamics in the limit) and preventing the inverse Reynolds numbers from exceeding the bound $\tilde{R}_\Pi + \tilde{R}_\pi \leq 1/\alpha$ .
Numerical Implementation: The scheme is implemented within the MUSIC hydrodynamic framework. The authors test three scenarios for initializing hydrodynamic simulations at $\sqrt{s_{NN}} = 5.02$ $s_{N N} = 5.02$ TeV for Pb+Pb and p+Pb collisions:
- Standard: Conventional DNMR with numerical stability cuts.
- Impose Necessary Cond.: Ad-hoc reduction of viscous tensors when causality is violated.
- Resummed Scheme: The proposed renormalization method.
  Initial conditions are derived from the IP-Glasma model, tested at three levels of non-equilibrium: complete equilibrium, half-viscous tensor, and full viscous tensor.

Key Contributions

A Causal Numerical Framework: The paper introduces a numerically agnostic scheme that dynamically enforces non-linear causality conditions by bounding inverse Reynolds numbers, rather than relying on post-hoc corrections.
Theoretical Uncertainty Quantification: By comparing the resummed scheme against standard and ad-hoc methods, the authors quantify the theoretical uncertainty introduced by causality enforcement on final-state observables.
High-Order Gradient Resummation: The work provides a practical interpretation of resummation, showing how non-linear corrections from high-order gradients can naturally limit the magnitude of viscous stress tensors.

Results
The event-by-event simulations yield the following findings:

Near Equilibrium: When simulations are initialized near thermal equilibrium (or with small viscous perturbations), the resummed scheme produces results nearly identical to the standard DNMR theory for both Pb+Pb and p+Pb collisions. The corrections are negligible, indicating the scheme behaves correctly in the linear regime.
Far from Equilibrium (Large Systems): For Pb+Pb collisions initialized with the full IP-Glasma viscous tensor (far from equilibrium), the resummed scheme shows robustness in central and semi-peripheral collisions. The long lifetime of these systems washes out the initial impacts of the resummed coefficients.
Far from Equilibrium (Small Systems): Significant differences emerge in peripheral Pb+Pb and small p+Pb collisions initialized far from equilibrium. In these cases, the resummed scheme evolves the system with effectively lower viscosities (due to the suppression of viscous terms), leading to reduced viscous entropy production and enhanced radial and anisotropic flows.
Sensitivity: The results indicate that theoretical uncertainties regarding causality conditions are most pronounced in small collision systems and peripheral events where the system is driven far from equilibrium.

Significance and Claims
The authors claim that their resummed hydrodynamic scheme offers a flexible and robust numerical framework for investigating the impact of resumming large gradient terms on heavy-ion phenomenology. They emphasize that the work is not a derivation of a new microscopic theory but rather an "agnostic" tool to explore theoretical uncertainties.

The paper asserts that:

The scheme successfully imposes non-linear causality conditions without destabilizing the simulation.
The choice of $\alpha = 1.5$ effectively enforces these conditions for the relevant range of the speed of sound squared ( $0.13 < c_s^2 < 1/3$ ).
The observed differences in small systems highlight a significant theoretical uncertainty in modeling collective dynamics in these regimes, suggesting that current precision comparisons with experimental data must account for these causality-related effects.
The framework is designed to be compatible with future Bayesian inference, allowing for data-driven constraints on the form of resummed hydrodynamics once microscopic theoretical inputs become available.

Technical Summary: A Resummed Hydrodynamic Description of Relativistic Heavy-ion Collisions

More like this