⚛️ phenomenology

Partonic Entropy of the Proton from DGLAP Evolution

This paper investigates the monotonic increase of proton partonic entropy under DGLAP evolution, demonstrating that saturation effects at small x are essential to tame this growth and proposing entanglement entropy as a testable observable within simplified saturation models.

Original authors: Krzysztof Golec-Biernat

Published 2026-01-27

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Krzysztof Golec-Biernat

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 the proton not as a solid marble, but as a bustling, chaotic city inside a tiny sphere. This city is filled with tiny messengers called partons (which include gluons and quarks). The paper by Krzysztof Golec-Biernat explores a fascinating question: How "messy" or "disordered" is this city as we look at it with higher and higher magnification?

Here is the story of the paper, broken down into simple concepts and analogies.

1. The Microscope and the City (DGLAP Evolution)

In particle physics, the "resolution" of your microscope is determined by a scale called $Q^2$ .

Low Resolution: You see the proton as a few big buildings.
High Resolution: As you zoom in (increase the energy), you start seeing that those buildings are actually made of smaller bricks, which are made of even smaller dust particles.

In the standard theory used here (called DGLAP), as you zoom in deeper, the number of these tiny partons explodes. It's like looking at a forest: from far away, it looks like a green blob. As you get closer, you see individual trees. As you get even closer, you see leaves, then veins in the leaves, then cells. The number of "things" you see keeps growing.

2. Measuring the "Mess" (Partonic Entropy)

The author wants to measure the entropy of this proton city. In everyday language, entropy is a measure of disorder or randomness.

The Analogy: Imagine a library.
- Low Entropy: All books are perfectly sorted by color and size. It's very ordered.
- High Entropy: Books are thrown everywhere, mixed up, and piled high. It's chaotic.

The paper defines a specific way to calculate this "messiness" based on how the partons are distributed. The key finding is that as you zoom in (increase the resolution), the proton's entropy goes up. The city gets messier and messier. The paper proves mathematically that this growth is steady and never stops within the current rules of the game.

3. The Problem: An Infinite Mess?

Here is the catch. In the standard "zoom-in" model, as you get infinitely close to the edge of the proton (a region called "small $x$ "), the number of partons seems to grow without limit.

The Analogy: Imagine the library keeps expanding. If you keep zooming in, you eventually find an infinite number of books. If the number of books is infinite, the "messiness" (entropy) becomes infinite.
The Paper's Claim: The math shows that without any limits, the entropy would grow forever. This is physically impossible because the proton is a finite object.

4. The Solution: The "Traffic Jam" (Saturation)

Nature has a safety valve. The paper argues that at some point, the partons get so crowded that they start bumping into each other and merging. This is called parton saturation.

The Analogy: Imagine a highway. At first, adding more cars increases the traffic flow (entropy). But eventually, the highway gets so full of cars that they can't move. They start merging lanes or stopping. The traffic density hits a maximum limit; it can't get any "denser" than a solid wall of cars.
The Result: This "traffic jam" stops the entropy from growing infinitely. It puts a cap on the messiness. The paper suggests that to get a realistic picture of the proton, we must include these "traffic jam" effects.

5. The Quantum Twist: Entanglement

The paper also touches on a very modern idea from quantum mechanics: Entanglement.

The Analogy: Imagine the proton is a giant puzzle. When you look at just one small piece (the part the probe sees), it looks random and messy. But that randomness isn't just chaos; it's because that piece is deeply connected (entangled) with the rest of the puzzle.
The Claim: The paper suggests that the "messiness" (entropy) we calculate might actually be a measure of how strongly the different parts of the proton are quantum-mechanically linked.
The Test: Interestingly, the authors mention that if you treat this entropy as "entanglement entropy," the predictions match real-world experimental data from particle accelerators. It's as if the proton is in a "maximally entangled state" when we look at it closely.

Summary of the Paper's Journey

Define the Mess: They created a formula to measure how disordered the proton's internal parts are.
Watch it Grow: They proved that as you look closer (higher energy), this disorder always increases.
Hit the Wall: They showed that without a limit, this disorder would become infinite, which doesn't make sense.
The Fix: They explained that "saturation" (partons crowding and merging) acts like a speed limit, stopping the disorder from growing forever.
The Deep Meaning: They propose that this disorder is actually a sign of deep quantum connections (entanglement) inside the proton, a theory that seems to match what scientists see in experiments.

In a nutshell: The proton is a chaotic city that gets messier the closer you look, but it has a built-in "crowd control" system that prevents the chaos from becoming infinite. This chaos might actually be the signature of the proton's quantum soul.

Technical Summary: Partonic Entropy of the Proton from DGLAP Evolution

Problem Statement
The paper addresses the theoretical definition and behavior of "partonic entropy" within the proton, specifically within the framework of collinear parton distribution functions (PDFs) evolving under the Dokshitzer–Gribov–Lipatov–Altarelli–Parisi (DGLAP) equations. While the rapid growth of gluon distributions at small Bjorken- $x$ (observed at HERA) is well-established, the information-theoretic properties of this dense partonic system remain a subject of active investigation. A central challenge is understanding how partonic entropy evolves with the resolution scale $Q^2$ and whether the unbounded growth predicted by linear evolution equations is physically realistic or requires modification via saturation effects. Furthermore, the paper seeks to clarify the connection between this partonic entropy and concepts from quantum information theory, specifically entanglement entropy.

Methodology
The authors employ a rigorous mathematical approach based on the standard DGLAP evolution scheme:

Definition of Probability Density: They define a probability density $P(x, t)$ on the interval $x \in [0, 1]$ by weighting the sum of singlet quark and gluon distributions by the longitudinal momentum fraction $x$ . This satisfies the momentum sum rule, interpreted here as a normalization condition ( $\int P dx = 1$ ).
Entropy Formulation: Using the Gibbs-Shannon formula, the partonic entropy $S(t)$ is defined as $S(t) = \int_0^1 dx P(x, t) \ln P(x, t)$ . The authors prove the non-negativity of this quantity using standard inequalities.
Evolution Analysis: The time derivative of the entropy with respect to the evolution variable $t = \ln(Q^2/Q_0^2)$ is derived directly from the DGLAP equations. The authors analyze the sign of this derivative by examining the properties of the splitting functions and the monotonicity of the PDFs at small $x$ .
Simplified Modeling: To isolate the mechanism driving entropy growth, the authors utilize a simplified parametric model of the PDFs (Beta-function form) where the small- $x$ $x$ behavior is controlled by a parameter $\alpha(t) \to 0$ $α (t) \to 0$ . They also analyze two specific models of parton saturation:
- A "reaction-diffusion" model (Markov process) including both splitting and recombination.
- A truncated model including only splitting (analogous to BFKL evolution).

Key Contributions and Results

Monotonic Growth in DGLAP: The primary result is a proof that within the standard linear DGLAP evolution scheme, the partonic entropy $S(t)$ increases monotonically with the evolution scale $t$ . This is demonstrated by showing that the derivative $dS/dt$ is strictly positive, driven by the behavior of the PDFs at small $x$ where $P(x)$ grows rapidly.
Relative Entropy Interpretation: The authors establish that the partonic entropy can be interpreted as the relative entropy (Kullback-Leibler divergence) of the evolving parton distribution $P(x, t)$ with respect to a uniform distribution $Q(x) \equiv 1$ . This implies that as $t$ increases, the parton distribution becomes increasingly distinguishable from a uniform state.
Origin of Unbounded Growth: Through the simplified model, the paper identifies the unbounded growth of entropy as originating specifically from the small- $x$ region ( $x \to 0$ ). In the limit where the small- $x$ exponent $\alpha \to 0$ , the entropy diverges as $1/\alpha$ .
Role of Saturation: The authors argue that this unbounded growth is an artifact of the linear DGLAP equations. They demonstrate that incorporating parton saturation effects (specifically gluon recombination) imposes an upper bound on the entropy. In a reaction-diffusion model including recombination, the entropy saturates to a finite value (approaching the entropy of a Poisson distribution) as rapidity $y \to \infty$ .
Entanglement Entropy Connection: In a model excluding recombination (pure splitting), the entropy grows linearly with rapidity ( $S \propto y$ ). The authors note that in this limit, the partonic microstates become equiprobable, leading to a conjecture (based on existing literature) that this partonic entropy corresponds to the entanglement entropy between the probed subsystem and the rest of the proton. This is linked to the idea that the proton is in a maximally entangled state at small $x$ .

Significance and Claims
The paper claims to provide a systematic derivation of partonic entropy starting from conventional collinear PDFs, rather than relying solely on simplified dipole models. It establishes a direct link between the DGLAP evolution dynamics and information-theoretic properties, showing that the linear evolution leads to a monotonic increase in entropy.

Crucially, the authors highlight that the unbounded growth of entropy is a signal that linear evolution breaks down at small $x$ , necessitating the inclusion of non-linear saturation effects to restore physical bounds. They propose that in models incorporating saturation, the partonic entropy stabilizes. Furthermore, they support the view that in the absence of saturation (pure splitting), the partonic entropy can be identified with entanglement entropy, a concept that has been tested against experimental data in other works. The paper concludes that while a precise formulation of saturation requires non-linear QCD evolution equations (beyond the scope of this work), the simplified models effectively illustrate how saturation tames the entropy growth and connects high-energy scattering to quantum information concepts.