🔢 mathematics

Regularity criterion for the 3D generalized Newtonian fluids

This paper establishes a regularity criterion ensuring that weak solutions to the Cauchy problem for 3D unsteady generalized Newtonian fluids become strong solutions within the range $\frac{5}{3} < p < \frac{11}{5}$ , provided the velocity gradient belongs to a specific critical Besov space $L^{\frac{2}{2-(3-p)a}}(0,T;\dot{B}^{-a}_{\infty,\infty}(\mathbb{R}^3))$ with parameters $a$ dependent on the power-law index $p$ .

Original authors: Qiao Liu, Xincheng Shi

Published 2026-03-23

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Qiao Liu, Xincheng Shi

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

Imagine you are watching a pot of thick soup, like a rich tomato bisque or a chunky stew, being stirred on a stove. This soup represents a generalized Newtonian fluid. Unlike water, which flows the same way no matter how fast you stir it, this soup changes its behavior. If you stir it gently, it might feel thick and gloopy (shear-thinning). If you stir it violently, it might suddenly feel stiff and resistant (shear-thickening).

Mathematicians and physicists want to predict exactly how this soup will move. They use a set of complex equations (the "governing equations") to describe the flow. However, there is a major problem: while they can prove that a "rough" solution exists (a solution that works on average but might have sudden, chaotic spikes or tears), they cannot always prove that the solution is "smooth" (perfectly predictable and continuous) for all time.

Think of a weak solution as a blurry, low-resolution video of the soup. You can see the general swirl, but if you zoom in, the pixels might break up, or the video might glitch. A strong solution is a crisp, 4K video where every drop of liquid is accounted for perfectly.

The Big Question

The paper asks: "What specific conditions must be met to ensure our blurry video becomes a crisp, 4K video?"

In the world of fluids, "smoothness" is often threatened by turbulence. If the fluid moves too erratically, the math breaks down. The authors are looking for a "Regularity Criterion"—a specific rule or warning sign that tells us, "As long as the fluid behaves this way, the video will stay crisp."

The New Rule: The "Smoothness Detector"

Previous rules for water (Newtonian fluids) and thick fluids were like checking the soup's temperature or its average speed. This paper introduces a more sophisticated detector.

The authors prove that if you look at the gradient of the velocity (which is a fancy way of saying "how quickly the speed of the fluid changes from one spot to the next") and check it against a specific mathematical "ruler" called a Besov space, you can guarantee the fluid stays smooth.

Here is the analogy:

Imagine the fluid's speed changes are like the texture of a fabric.
A "weak solution" might have a fabric that feels rough or has hidden holes (singularities).
The authors say: "If we measure the roughness of this fabric using a very specific, high-tech microscope (the Besov space $L^{2/(2-(3-p)a)}(0, T; \dot{B}^{-a}_{\infty,\infty})$ ), and the roughness stays within a certain limit, then the fabric is actually smooth silk underneath."

The "Goldilocks" Zone

The paper focuses on a specific range of fluid behaviors, defined by a number called $p$ .

If $p=2$ , it's normal water.
If $p < 2$ , it's a fluid that gets thinner when you stir it (like ketchup).
If $p > 2$ , it's a fluid that gets thicker when you stir it (like cornstarch and water).

The authors found a "Goldilocks zone" for these fluids (specifically when $5/3 < p < 11/5$ ). In this zone, they proved that if the "texture" of the fluid's speed changes fits a very precise mathematical pattern, the fluid will never glitch. It will remain a strong solution (smooth and predictable) forever.

How They Did It (The Secret Sauce)

To prove this, the authors had to solve a tricky puzzle. They needed to estimate how much energy the fluid's turbulence could generate.

The Old Way: Previous mathematicians used standard tools (like a hammer) to estimate this energy. It worked for some cases but missed the nuances of these specific thick/thin fluids.
The New Way: The authors used a "sharper scalpel." They developed a new inequality (a mathematical rule of thumb) that acts like a smart filter. This filter allows them to look at the fluid's roughness in a way that connects the "average" behavior to the "extreme" behavior.

They essentially showed that if the fluid's "roughness" doesn't get too wild in this specific mathematical space, the energy of the turbulence is kept in check, preventing the "video" from glitching.

Why This Matters

This isn't just about soup. These fluids are everywhere:

Blood: Understanding how blood flows in arteries (which can be shear-thinning) helps in designing better medical treatments.
Polymers: Making plastics and synthetic fibers requires knowing how these materials flow under heat and pressure.
Food Industry: Making yogurt, chocolate, or sauces requires precise control over flow.

By proving that these fluids stay smooth under these specific conditions, the authors give engineers and scientists a new safety net. They now know exactly what to look for to ensure that simulations of these fluids will remain accurate and stable, preventing the "mathematical glitches" that could lead to failed designs or incorrect predictions.

In short: The authors found a new, more sensitive way to measure the "roughness" of complex fluids. They proved that if this roughness stays within a specific, narrow band, the fluid will flow perfectly smoothly, never breaking down into chaos.

1. Problem Statement

The paper addresses the regularity problem for the 3D unsteady flows of generalized Newtonian fluids. These fluids are governed by a system of partial differential equations (PDEs) where the viscosity depends on the shear rate (characterized by the exponent $p$ ). The governing equations are:
$\begin{cases} u_t - \text{div}(|Du|^{p-2}Du) + (u \cdot \nabla)u + \nabla \pi = 0, \\ \text{div } u = 0, \\ u(\cdot, 0) = u_0, \end{cases}$
where $u$ is the velocity, $\pi$ is the pressure, and $Du = \frac{1}{2}(\nabla u + (\nabla u)^T)$ is the symmetric velocity gradient.

Context: For $p=2$ , this reduces to the classical Navier-Stokes (NS) equations. For $p \neq 2$ , it models shear-thinning ( $1 < p < 2$ ) or shear-thickening ( $p > 2$ ) fluids.
The Challenge: While the existence of weak solutions is established for $p > 1$ , the uniqueness and regularity (smoothness) of these solutions remain open problems, particularly for the range $5/3 < p < 11/5$ .
Goal: The authors aim to establish a regularity criterion (a sufficient condition) that ensures a weak solution becomes a strong (smooth) solution. Specifically, they seek to improve existing criteria by utilizing homogeneous Besov spaces with negative regular indices ( $\dot{B}^{-a}_{\infty, \infty}$ ), which are larger spaces than classical Lebesgue spaces.

2. Methodology

The proof relies on energy estimates, interpolation inequalities, and specific properties of Besov spaces.

A. Energy Estimates

The authors multiply the momentum equation by $\Delta u$ and integrate over $\mathbb{R}^3$ . This yields a differential inequality for the $L^2$ -norm of the gradient:
$\frac{1}{2}\frac{d}{dt}\|\nabla u\|_{L^2}^2 + S \leq c_1 \int_{\mathbb{R}^3} |\nabla u|^3 dx,$
where $S = \int |Du|^{p-2}|\nabla^2 u|^2 dx$ represents the dissipation term. The core difficulty lies in controlling the nonlinear term $\int |\nabla u|^3 dx$ .

B. Case-Specific Analysis

The authors split the analysis into two regimes based on the exponent $p$ :

Case 1: Shear-thinning ( $5/3 < p < 2$ ):
- They employ an interpolation between $L^{3p}$ and $L^q$ (where $q < 2$ ).
- They utilize Lemma 2.3, an improved Gagliardo-Nirenberg inequality in Besov spaces:
  $\|f\|_{L^m} \leq c \|f\|_{\dot{B}^{-a}_{\infty, \infty}}^\theta \|f\|_{L^n}^{1-\theta}.$
- By carefully choosing parameters $q$ and $a$ , they bound the $L^3$ norm of $\nabla u$ using the Besov norm $\|\nabla u\|_{\dot{B}^{-a}_{\infty, \infty}}$ and the $L^2$ norm, allowing the dissipation term $S$ to absorb the critical part of the estimate.
Case 2: Shear-thickening ( $2 \leq p < 11/5$ ):
- They use interpolation between $L^2$ and $L^q$ (where $q > 3$ ).
- Similar to Case 1, they apply the improved Gagliardo-Nirenberg inequality (Lemma 2.3) to estimate $\|\nabla u\|_{L^q}$ in terms of the Besov norm and $L^{3p}$ norm.
- Through Young's inequality and careful parameter selection, they again isolate the Besov norm term to control the growth of the energy.

C. Gronwall's Inequality

Once the nonlinear term is bounded by the Besov norm and the $L^2$ energy, the authors apply Gronwall's inequality to the differential inequality. This demonstrates that if the Besov norm of $\nabla u$ satisfies a specific integrability condition, the $L^2$ norm of $\nabla u$ remains bounded, preventing blow-up.

3. Key Contributions

Extension to Critical Besov Spaces: The paper unifies and extends previous Prodi-Serrin type criteria (which were typically in Lebesgue spaces $L^\alpha$ ) to critical homogeneous Besov spaces $\dot{B}^{-a}_{\infty, \infty}$ . Since $L^\alpha \subset \dot{B}^{-a}_{\infty, \infty}$ , this result covers a broader class of functions, including those with singularities that are not in $L^\alpha$ .
Unified Criterion for $5/3 < p < 11/5$ : The authors provide a single framework that covers both shear-thinning and shear-thickening regimes within the critical range where global strong solutions are not yet guaranteed by standard existence theory.
Novel Estimate Technique: Unlike previous works that used standard bilinear estimates, the authors utilize a specific improved Gagliardo-Nirenberg inequality (Lemma 2.3) tailored to handle the interpolation between different $L^p$ spaces and Besov spaces. This allows them to derive the necessary bounds on $\|\nabla u\|_{L^3}$ without requiring the stronger $L^\infty$ bounds that are difficult to obtain.

4. Main Results

Theorem 1.1: Let $5/3 < p < 11/5$ and $u_0 \in W^{1,2}_\sigma \cap W^{1,p}$ . If a weak solution $u$ satisfies:
$\nabla u \in L^{\frac{2}{2-(3-p)a}}(0, T; \dot{B}^{-a}_{\infty, \infty}(\mathbb{R}^3)),$
where the parameter $a$ satisfies:

$3/2 < a < \frac{2}{3-p}$ if $5/3 < p < 2$ ,
$1/p < a < \frac{2}{3-p}$ if $2 \leq p < 11/5$ ,

Then the weak solution $u$ is a strong solution on $\mathbb{R}^3 \times (0, T]$ .

Corollaries/Extensions:

The result implies previous criteria in Lebesgue spaces (e.g., by Sin and Baranovskii) as special cases.
Due to the equivalence of norms for divergence-free fields, the criterion can also be stated in terms of the vorticity $\nabla \times u$ or the velocity $u$ itself (with adjusted time integrability exponents).

5. Significance

Mathematical Rigor: The paper advances the understanding of the regularity of non-Newtonian fluids, a field crucial for modeling complex industrial and biological flows (e.g., blood, polymers).
Sharpness of Conditions: By working in Besov spaces with negative indices, the authors identify the "critical" scaling-invariant spaces for regularity. This is a significant step toward resolving the long-standing open problem of uniqueness and regularity for these systems.
Methodological Innovation: The application of the specific interpolation inequality (Lemma 2.3) to the generalized Newtonian fluid equations offers a new toolset for analyzing similar nonlinear PDE systems where standard Sobolev embeddings are insufficient.

In summary, this paper provides a robust regularity criterion that bridges the gap between weak and strong solutions for 3D generalized Newtonian fluids in the critical range $5/3 < p < 11/5$ , utilizing advanced harmonic analysis tools to refine existing conditions.