⚛️ lattice

Pion Transition Form Factor in Lattice QCD

Using the blending method and distillation framework on an $N_f=2+1$ lattice ensemble, this study confirms that the disconnected contribution to the neutral pion transition form factor is approximately 1% of the connected part and shares the same sign, thereby enabling constructive interference.

Original authors: Shihao Su, Liuming Liu, Peng Sun

Published 2026-07-31

📖 4 min read🧠 Deep dive

CC BY 4.0

Original authors: Shihao Su, Liuming Liu, Peng Sun

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 universe is built from a giant, invisible LEGO set. The smallest bricks in this set are particles called quarks, which snap together to form protons, neutrons, and the pions that zip around inside atomic nuclei. But these bricks don't just sit there; they are constantly jiggling, interacting, and swapping energy with invisible messengers called photons. Sometimes, a neutral pion (a specific type of particle made of quarks) can turn into two photons, or vice versa. Scientists call this a "transition," and to understand exactly how likely it is to happen, they need to measure something called a "form factor." Think of the form factor as a detailed map or a recipe card that tells you exactly how the pion and the photons are holding hands.

Why does this matter? Because this tiny interaction is a key piece of a much bigger puzzle: understanding why the muon (a heavy cousin of the electron) spins a little bit differently than our current best theories predict. This difference, known as the "muon g-2," is one of the hottest mysteries in physics right now. If we get the recipe card for the pion wrong, our predictions for the muon's behavior will be off, and we might miss a clue about new, undiscovered physics hiding in the shadows. To get this recipe right, scientists use a super-powerful computer simulation called "Lattice QCD," which treats space and time like a giant 3D grid to calculate how these particles behave from first principles.

In this study, a team of researchers set out to solve a specific disagreement in the scientific community about how the pion's "recipe" is put together. When calculating the pion's transition form factor, the math breaks down into two main types of diagrams: "connected" and "disconnected." You can think of the connected part as a direct highway where the quarks flow smoothly from one end of the interaction to the other. The disconnected part is more like a side road where the quarks loop back on themselves, creating a temporary, ghostly bubble before rejoining the main flow.

For a long time, many scientists believed these two parts worked against each other, like two people pushing a car from opposite sides (destructive interference). However, a recent study suggested they might actually be pushing in the same direction (constructive interference). This paper steps in to settle the score using a clever new trick called the "blending method."

The researchers used a supercomputer to simulate the universe on a grid (specifically an $N_f = 2 + 1$ lattice ensemble). They faced a tricky problem: calculating the "disconnected" part is notoriously noisy and expensive, like trying to hear a whisper in a hurricane. To fix this, they used the blending method, which is like having a high-resolution camera for the important parts of the picture (the low-energy "ground state") and a smart, statistical guess for the blurry background (the high-energy modes). This allowed them to get a clear, unbiased view of both the connected and disconnected contributions without breaking the bank on computing power.

What did they find? The simulations showed that the disconnected part has the same sign as the connected part. In our car analogy, this means both groups are pushing in the same direction, helping each other out rather than canceling each other out. This confirms the recent observation that the two contributions interfere constructively.

The numbers tell the full story. On their specific lattice setup (the C24P29 ensemble), the disconnected contribution is tiny—about 1% of the size of the connected part. For example, at a specific momentum configuration where the virtuality is $-Q^2$ , the connected part contributes roughly 209.5(2.0) (in units of $10^{-3} \text{ GeV}^{-1}$ ), while the disconnected part adds a small but positive 1.63(0.63). Because they share the same sign, they add up to make the total effect slightly stronger, rather than weaker.

This result is a big deal because it clears up a confusion that has plagued previous calculations. By proving that the signs are the same, the team has helped tighten the theoretical constraints on the muon's magnetic moment. While the disconnected part is small, knowing exactly how it behaves is essential for reducing the tiny errors in our understanding of the Standard Model. The paper doesn't claim to have solved the muon mystery entirely, but it has definitely sharpened the lens through which we are looking at it, confirming that the pion's internal structure is a bit more cooperative than some had feared.

Technical Summary: Pion Transition Form Factor in Lattice QCD

Problem Statement
The neutral pion transition form factor, $F_{\pi^0\gamma^*\gamma^*}(q_1^2, q_2^2)$ , describes the coupling of a neutral pion to two off-shell photons. This quantity is critical for precision tests of the Standard Model, particularly in reducing the theoretical uncertainty of the hadronic light-by-light (HLbL) scattering contribution to the muon anomalous magnetic moment ( $g-2$ ). The $\pi^0$ -pole term constitutes the dominant source of uncertainty in HLbL calculations.

A significant theoretical ambiguity exists regarding the relative sign between the "connected" and "disconnected" contributions to this form factor. Previous lattice QCD studies have reported conflicting conclusions: some suggest the contributions carry opposite signs (destructive interference), while a recent study [20] suggests they carry the same sign (constructive interference). Resolving this discrepancy is essential for accurate HLbL predictions, especially since the double-virtual form factor required for dispersive analyses has not been measured experimentally.

Methodology
The authors perform a lattice QCD calculation to extract the form factor and definitively determine the sign relation between the connected and disconnected diagrams. The study utilizes an $N_f = 2+1$ lattice ensemble (C24P29) with clover fermions.

Key methodological features include:

Blending Method: The authors employ a recently proposed "blending" technique to estimate all-to-all propagators. This method combines the ground-state filtering properties of the distillation framework (using low-mode eigenvectors of the Laplace operator) with a stochastic estimate of high-frequency modes using random vectors. This approach provides unbiased estimators at a moderate computational cost.
Hybrid Treatment of States and Currents:
- External Pion States: Treated within the distillation framework (low-mode subspace $L_1$ ) to suppress excited-state contamination.
- Electromagnetic Currents: Evaluated in the full blending space ( $L_1 \oplus L_2$ ), where $L_2$ is approximated by a set of orthonormal random vectors.
Unbiased Estimators: The paper details the construction of unbiased estimators for the identity operator and bilinear operators within the blending basis. It addresses the treatment of non-diagonal operator matrices by dividing the eigenvector space into multiple parts and using block/interlace schemes with compression matrices to maintain statistical unbiasedness.
Correlation Functions: The three-point correlation functions are decomposed into connected and disconnected topological classes via Wick contractions. The authors compute both contributions separately using the blending method to compare their signs directly.

Key Results
Numerical tests were performed on the C24P29 ensemble with pion mass $m_\pi \approx 293$ MeV. The results for the function $A(\tau)$ , from which the form factor is extracted, show:

Sign Agreement: Within statistical precision, the disconnected contribution to the neutral pion transition form factor has the same sign as the connected contribution.
Magnitude: The disconnected contribution is approximately 1% of the magnitude of the connected contribution.
Interference: The identical signs imply constructive interference of the probability amplitudes between the connected and disconnected parts.
Validation: The paper validates the unbiased nature of the blending method by calculating the conserved vector current matrix element, which remains consistent with unity within a 1% error margin.

Table II presents the extracted form factor values $F_{\pi^0\gamma^*\gamma^*}(-Q^2, -Q^2)$ for different momentum configurations, showing the separate contributions of the connected (BLD-con and hybrid con) and disconnected (dis) parts.

Significance and Claims
The paper claims to definitively resolve the sign ambiguity regarding the connected and disconnected contributions to the $\pi^0\gamma^*\gamma^*$ form factor. By confirming that these contributions share the same sign, the study supports the findings of the recent work [20] and contradicts earlier literature [1–3, 17, 21–23] that suggested opposite signs.

The authors emphasize that this result is crucial for:

Reducing Theoretical Uncertainty: A correct determination of the relative sign is necessary to minimize systematic errors in lattice QCD calculations of the HLbL contribution to the muon $g-2$ .
First-Principles Access: Since the double-virtual form factor is inaccessible to current experiments, this lattice calculation provides the only first-principles tool to access this kinematic regime.
Methodological Validation: The successful application of the blending method to this complex three-point function demonstrates its viability for calculating hadronic matrix elements involving all-to-all propagators with unbiased estimators.

The paper concludes that the constructive interference observed enables a more precise determination of the transition form factor, which is a prerequisite for reducing the theoretical error in the Standard Model prediction of the muon anomalous magnetic moment.

More like this