Shear and bulk viscosities of the gluon plasma across the transition temperature from lattice QCD

Using high-precision lattice QCD calculations with controlled continuum extrapolations and gradient flow, this study determines the temperature dependence of shear and bulk viscosities in gluon plasma across the transition region, revealing that the shear viscosity ratio $\eta/s$ reaches a minimum near $T_c$ while the bulk viscosity ratio $\zeta/s$ decreases monotonically with increasing temperature.

The Gist

Imagine a pot of water on the stove. As it heats up, it goes from a calm liquid to a bubbling, chaotic boil. Now, imagine a substance even more extreme: Quark-Gluon Plasma (QGP). This is the "primordial soup" that existed microseconds after the Big Bang, where the building blocks of matter (quarks) and the glue holding them together (gluons) are free-floating in a super-hot, super-dense fluid.

Physicists want to know: How "thick" or "sticky" is this cosmic soup?

In everyday terms, this is about viscosity.

• Shear Viscosity: Think of honey vs. water. Honey resists being stirred (high viscosity); water flows easily (low viscosity).
• Bulk Viscosity: Think of a sponge. If you squeeze it, does it resist compression? That's bulk viscosity.

This paper is a high-precision measurement of how "sticky" this cosmic soup is at different temperatures, specifically around the point where it changes from a solid-like state (confinement) to a fluid state (deconfinement).

Here is the story of how they did it, explained simply:

1. The Problem: The "Frozen" Snapshot

To measure viscosity, you usually watch a fluid flow in real-time. But in the world of quantum physics (specifically Lattice QCD), we can't watch time flow forward easily. Instead, we take a series of "snapshots" of the system in a frozen, mathematical state called Euclidean time.

The challenge is like trying to figure out how a car drives by looking at a single, blurry photograph of its shadow. You have to mathematically "reverse engineer" the motion from the static image. This is a notoriously difficult math puzzle called an "ill-posed problem" because there are infinite ways to draw a line through a few dots.

2. The Tools: Smoothing the Noise

The researchers used two powerful tricks to get a clear picture:

• Gradient Flow (The "Blur" Filter): Imagine taking a noisy, grainy photo and applying a blur filter. In physics, this "smears" out the tiny, chaotic quantum jitters (noise) that hide the real signal. It's like turning down the static on a radio so you can hear the music.
• Blocking (The "Puzzle" Solver): Even with the blur filter, the signal is still faint. The researchers used a "blocking" technique. Imagine you are trying to hear a whisper in a crowded room. Instead of listening to one person, you group people together and average their voices. This boosts the signal without needing more microphones (computing power).

By combining these, they achieved percent-level precision. They didn't just get a guess; they got a very sharp, clear image of the fluid's behavior.

3. The Experiment: Heating the Pot

They simulated this plasma at seven different temperatures, ranging from just below the "boiling point" (0.76 times the critical temperature, $T_c$) to well above it (2.25 times $T_c$).

They used three different "magnifying glasses" (lattice sizes) for each temperature. This is crucial because it allows them to zoom out and see the "true" picture, removing the distortion caused by the pixelation of their computer grid (the continuum extrapolation).

4. The Results: The "Perfect Fluid" and the "Sponge"

Here is what they found about the two types of stickiness:

A. Shear Viscosity (The "Stirring" Resistance)

• The Finding: As the temperature rises toward the transition point ($T_c$), the plasma becomes less sticky (more fluid). It hits a "sweet spot" right at the transition where it is the most perfect fluid imaginable (approaching the theoretical limit of $1/4\pi$).
• The Analogy: Imagine a crowd of people. When they are cold, they are stiff and resist moving. As they warm up, they start dancing and flowing past each other effortlessly. But if they get too hot, they start bumping into each other again, becoming slightly "thicker" again.
• The Trend: It gets slippery near the transition, then gets a bit thicker as it gets super hot.

B. Bulk Viscosity (The "Squeezing" Resistance)

• The Finding: This one behaves differently. As the temperature goes up, the resistance to compression monotonically decreases.
• The Analogy: Think of a sponge. When it's cold and dry, it's hard to squeeze. As it heats up and becomes the plasma, it becomes incredibly easy to compress. The hotter it gets, the less it resists being squished.
• The Trend: It starts high near the transition and just keeps dropping as the temperature rises.

5. Why This Matters

• The "Perfect" Fluid: The fact that the shear viscosity drops so low near the transition temperature confirms that the Quark-Gluon Plasma created in particle colliders (like the LHC) behaves like the "perfect fluid" predicted by string theory (AdS/CFT). It flows with almost zero friction.
• The "Sponge" Effect: The behavior of the bulk viscosity helps explain how the universe expanded and cooled right after the Big Bang.
• Better Math: Previous studies had to guess a lot about the "shape" of the data because their measurements were noisy. This paper's high precision allows them to constrain those guesses much tighter, giving us a more reliable map of the early universe.

Summary

The authors took a difficult, blurry math problem and used clever "noise-canceling" techniques to get a crystal-clear view of how the early universe's "soup" flowed. They found that right at the moment the universe changed from a solid-like state to a fluid, it became the most slippery substance in existence, only to get slightly thicker again as it got hotter, while its resistance to being squeezed simply vanished as it heated up.

Technical Summary

1. Problem Statement

The transport properties of the Quark-Gluon Plasma (QGP), specifically the shear viscosity ($\eta$) and bulk viscosity ($\zeta$), are critical for understanding the collective dynamics of heavy-ion collisions. While phenomenological models and hydrodynamic simulations suggest small viscosities (near the AdS/CFT lower bound $\eta/s \ge 1/4\pi$), first-principles calculations are challenging.

• Perturbative QCD: Calculations for shear viscosity show slow convergence between Leading Order (LO) and Next-to-Leading Order (NLO). Bulk viscosity calculations are even more complex and currently available only at LO.
• Lattice QCD Challenges: Extracting real-time transport coefficients from Euclidean lattice correlators is an ill-posed inverse problem. Previous lattice studies (using the Multilevel algorithm) were limited to pure gauge theory, restricted temperature ranges ($0.9–1.5 T_c$), and often lacked controlled continuum extrapolations due to coarse lattices. Furthermore, the Multilevel algorithm is not straightforwardly applicable to full QCD with dynamical quarks.

2. Methodology

The authors employ a high-precision lattice QCD approach using Gradient Flow (GF) combined with a Blocking technique to overcome signal-to-noise issues and discretization errors.

A. Lattice Setup and Simulation

• Theory: Pure $SU(3)$ Yang-Mills theory (gluon plasma).
• Temperature Range: $0.76 T_c$ to $2.25 T_c$, covering the confined phase, the transition region, and the deconfined phase.
• Ensembles: At each temperature, simulations were performed on three large, fine lattices to enable controlled continuum extrapolations.
  • Maximal spatial extent: $L = 3.31$ fm.
  • Minimal lattice spacing: $a = 0.01164$ fm.
  • Statistics: $\sim 5000$ configurations per ensemble.
• Algorithms: Heat Bath and Over-Relaxation algorithms were used for generation.

B. Gradient Flow and Blocking

• Gradient Flow: The gauge fields are evolved in a fictitious flow time ($\tau_F$) to suppress UV fluctuations. This regularizes the Energy-Momentum Tensor (EMT) operators ($T_{\mu\nu}$) without introducing new renormalization constants beyond known perturbative factors ($c_1, c_2$).
• Blocking Technique: A recently developed method is applied to further improve the signal-to-noise ratio by a factor of 3–7. It replaces noisy spatial correlators at large distances with fitted values, utilizing correlations between spatial distances.
• Renormalization: The renormalization constants $c_1$ and $c_2$ are determined semi-nonperturbatively using the gradient flow coupling, extrapolated to the continuum limit.

C. Spectral Reconstruction

The core challenge is inverting the integral transform relating Euclidean correlators $G(\tau)$ to the spectral function $\rho(\omega)$:
$$G(\tau) = \int_0^\infty \frac{d\omega}{\pi} \frac{\cosh[\omega(1/2T - \tau)]}{\sinh(\omega/2T)} \rho(\omega, T)$$

• UV Constraint: The high-frequency (UV) behavior of the spectral function is constrained using the best available NLO perturbative spectral functions for thermal $SU(3)$.
• IR Modeling: The low-frequency (IR) transport peak is modeled using a Lorentzian ansatz.
• Systematic Uncertainty: Since the width of the transport peak ($C$) is unknown, the authors bracket the dominant modeling uncertainty by varying $C$ between physically motivated limits ($C=1$ for a sharp peak, $C=5$ for a broad peak).
• Below $T_c$: For temperatures below the transition, a smoothing function is introduced to handle structural changes in the spectral function not captured by the perturbative UV form.

3. Key Contributions

1. Extended Temperature Range: The study covers $0.76 T_c$ to $2.25 T_c$, significantly broader than previous lattice studies ($0.9–1.5 T_c$), allowing for a comprehensive view of the phase transition.
2. Controlled Continuum Extrapolation: By using three large, fine lattices at each temperature, the authors perform rigorous continuum extrapolations, eliminating discretization errors that plagued earlier studies using coarse lattices.
3. High-Precision Correlators: The combination of Gradient Flow and Blocking achieves percent-level precision (3%–12% relative uncertainty) for the EMT correlators, providing stringent constraints for spectral reconstruction.
4. Quantified Systematics: Instead of assuming a fixed transport peak, the authors explicitly quantify the systematic uncertainty arising from the unknown peak width by scanning a physically motivated range.

4. Key Results

Shear Viscosity ($\eta$)

• Normalized by $T^3$: $\eta/T^3$ exhibits a minimum (dip) near the transition temperature $T_c$. Below $T_c$, it shows weak temperature dependence; above $T_c$, it increases with temperature.
• Normalized by Entropy ($\eta/s$): The ratio $\eta/s$ reaches a minimum near $T_c$ (close to the AdS/CFT bound $1/4\pi$) and rises for $T > T_c$. At high temperatures, the results approach the NLO perturbative QCD prediction.
• Comparison: The results are consistent with previous lattice studies using the Multilevel algorithm (e.g., Meyer, Astrakhantsev) but provide tighter constraints due to better continuum control.

Bulk Viscosity ($\zeta$)

• Normalized by $T^3$: $\zeta/T^3$ develops a peak near $T_c$ and decreases monotonically in the deconfined phase ($T > T_c$). Below $T_c$, it shows weak temperature dependence.
• Normalized by Entropy ($\zeta/s$): The ratio $\zeta/s$ decreases monotonically across the entire studied temperature range.
• Comparison: The magnitude and trend agree well with other lattice determinations and sum rule analyses, though values below $T_c$ differ from some analytic fits due to differences in entropy density determination.

Systematic vs. Statistical Uncertainty

The study highlights that the systematic uncertainty arising from the spectral model (specifically the transport peak width) dominates the total error budget (typically >70%), rather than statistical noise. This underscores the difficulty of the inverse problem even with high-precision data.

5. Significance

• First-Principles Benchmark: This work provides the most rigorous, continuum-extrapolated determination of gluon plasma viscosities to date, serving as a critical benchmark for hydrodynamic models of heavy-ion collisions.
• Validation of Perturbation Theory: The agreement between lattice results and NLO perturbative QCD at high temperatures ($T > 1.5 T_c$) validates the use of perturbative methods in the deconfined regime.
• Methodological Advancement: The successful application of Gradient Flow + Blocking to pure gauge theory demonstrates a viable path for future calculations in full QCD (with dynamical quarks), where Multilevel algorithms are difficult to implement.
• Insight into Phase Transition: The distinct behaviors of $\eta$ and $\zeta$ across $T_c$ (dip vs. peak) offer deep insights into the changing degrees of freedom and interaction strength of the QGP as it transitions from a hadronic gas to a deconfined plasma.