Subthreshold parameters of $ππ$ scattering revisited

This paper utilizes experimental data, lattice QCD calculations, and Roy equation-based dispersive representations to compute the subthreshold parameters of $\pi\pi$ scattering, employing Monte Carlo sampling to quantify uncertainties and analyzing the impact of theoretical correlations between scattering lengths.

The Gist

Imagine the subatomic world as a giant, chaotic dance floor. On this floor, particles called pions (think of them as the "social butterflies" of the atomic world) are constantly bumping into each other, bouncing off, and swirling around. Physicists call this $\pi\pi$ scattering.

The goal of this paper is to figure out exactly how these pions interact when they are moving very slowly—so slowly that they are just about to touch, but haven't quite started their full dance yet. This specific moment is called the "subthreshold" region.

Here is the breakdown of what the authors did, using some everyday analogies:

1. The Mystery of the "Hidden Rules"

Physicists have a set of mathematical rules (called Chiral Perturbation Theory) that predict how these pions should behave. These rules have "knobs" or parameters that need to be turned to match reality.

• The Problem: In the past, some experiments suggested these knobs were turned to very extreme positions. This implied that the pions were behaving in a way that broke other parts of physics (like suggesting the pions were much lighter than they actually are).
• The Goal: The authors wanted to measure these knobs with extreme precision to see if the "extreme" theories were right or if they were just looking at the data wrong.

2. The Detective Work: Using "Roy Equations" as a Map

To solve this, the authors didn't just guess. They used a powerful mathematical tool called Roy Equations.

• The Analogy: Imagine you are trying to figure out the shape of a hidden object inside a foggy room. You can't see the object directly, but you can see how light bounces off the walls (the scattering lengths). The Roy Equations are like a super-advanced map that tells you: "If the light bounces off the wall at this specific angle, the hidden object must be shaped exactly like this."
• The authors took the known "bounces" (experimental data) and used the map to reconstruct the hidden shape (the subthreshold parameters).

3. Gathering the Evidence: Two Sources of Truth

To make sure their map was accurate, they gathered clues from two very different places:

1. The Experimentalists (NA48/2): These are the people who actually built giant machines to watch pions collide. They measured how pions behave in a specific decay process (like watching a slow-motion crash).
2. The Simulators (Lattice QCD/ETM): These are supercomputers that simulate the entire universe from scratch, calculating how pions interact based on the fundamental laws of physics, without needing a physical lab.

4. The "Monte Carlo" Dance Party

The authors didn't just calculate one single answer. They knew that every measurement has a tiny bit of error (like trying to measure a table with a ruler that has slightly worn-out markings).

• The Method: They ran a computer simulation 100,000 times (a "Monte Carlo" method).
• The Analogy: Imagine you are trying to guess the average height of a crowd. Instead of measuring one person, you randomly pick 100,000 people, measure them, and average the results. By doing this, they created a "probability cloud" that showed not just the most likely answer, but how much wiggle room there was in the answer.

5. The Big Reveal: Smoothing Out the Rough Edges

When they put all the data together, they found something important:

• The Old Theory was Wrong: Previous studies (by a group called DFGS) suggested the "knobs" were turned to extreme values. The authors found that those results were likely due to a specific assumption they made about how the data correlated.
• The New Reality: When they removed that assumption and used the new, high-precision data, the "knobs" turned out to be much closer to 1.
• Why this matters: A value of 1 is what the simplest version of the theory predicts. This means the theory is working beautifully! It suggests that the pions are behaving exactly as the "standard model" of particle physics expects them to, without needing weird, complicated corrections.

6. The "Tension" Resolution

There was a previous conflict (tension) between different experiments. Some said the pions were heavy; others said they were light.

• The Resolution: The authors showed that the conflict wasn't because the pions were actually behaving strangely. It was because some researchers were relying on a specific theoretical shortcut (the "CGL correlation"). When they used the raw data from the supercomputers and the new experiments, the conflict disappeared. The pions are just behaving normally.

Summary

Think of this paper as a group of detectives who finally solved a cold case.

• The Case: Why did some measurements of particle collisions look so weird?
• The Clues: New, ultra-precise measurements from a lab in Europe and supercomputer simulations.
• The Tool: A mathematical map (Roy Equations) that connects the dots.
• The Verdict: The "weirdness" was an illusion caused by a bad assumption. The particles are behaving exactly as the laws of physics predict. The "knobs" on the theory are set to the perfect, simple values, confirming our understanding of the subatomic dance floor.

Technical Summary

1. Problem Statement

The paper addresses the determination of subthreshold parameters ($\alpha_{\pi\pi}, \beta_{\pi\pi}, \lambda_{1-4}$) and coefficients ($\bar{b}_i$) describing the $\pi\pi$ scattering amplitude. These parameters are crucial for:

• Testing the convergence of Chiral Perturbation Theory ($\chi$PT), particularly in the 3-flavor framework.
• Resolving tensions between different experimental and theoretical analyses regarding the value of $\alpha_{\pi\pi}$.
• Determining the leading-order 3-flavor low-energy constant (LEC) $B_0$, which relates to the pion mass and the chiral condensate.

The Core Tension:
Previous analyses by Descotes-Genon et al. (DFGS) [5], based on BNL-E865 data, suggested a large value for $\alpha_{\pi\pi} \approx 1.38$. This implies a large next-to-leading order (NLO) correction and a suppressed leading-order pion mass ($Y^{(3)} \sim 0$), which conflicts with expectations from 2-flavor $\chi$PT and $\eta \to 3\pi$ data. Conversely, Colangelo, Gasser, and Leutwyler (CGL) [7] and recent lattice QCD results suggest a value closer to unity ($\alpha_{\pi\pi} \approx 1.08$), indicating better convergence. The authors aim to resolve this discrepancy using modern data and rigorous dispersive methods.

2. Methodology

The authors employ a dispersive approach based on Roy equations, which enforce unitarity, analyticity, and crossing symmetry. The methodology involves the following steps:

1. Amplitude Representation:

   • The scattering amplitude $A_{\pi\pi}(s, t, u)$ is represented in two ways:
     • Phenomenological: Based on Roy equations with subtraction constants identified as the S-wave scattering lengths $a_0^0$ and $a_2^0$.
     • Subthreshold: An expansion around the subthreshold point (Eq. 1) containing the parameters of interest ($\alpha_{\pi\pi}, \beta_{\pi\pi}, \lambda_i$).
   • A bridge is established between these representations via the $b_i$-representation (Bijnens et al.) and chiral coefficients $c_i$.

2. Input Data:
   The authors utilize a Monte Carlo sampling approach ($10^5$ entries) to propagate uncertainties from various inputs:

   • Experimental: Scattering lengths from the NA48/2 collaboration (Model C), derived from $K_{e4}$ decay and the $K \to 3\pi$ cusp.
   • Lattice QCD: Results from ETM (for $a_2^0$) and RBC/UKQCD.
   • Theoretical Constraints: The phase shifts at the matching point ($\sqrt{s_0} = 800$ MeV) and background moments ($I_n^I, H$) from ACGL [21] and DFGS [5].

3. Handling Correlations:
   A critical aspect of the analysis is testing the dependence on the theoretical correlation between $a_0^0$ and $a_2^0$ (Eq. 41), which relies on the scalar pion radius and 2-flavor $\chi$PT.

   • Scenario A: Use NA48/2 Model C (which assumes the correlation).
   • Scenario B: Use NA48/2 data combined with the independent ETM lattice calculation of $a_2^0$ to remove the theoretical correlation assumption.
   • Scenario C: Use ETM $a_2^0$ combined with the CGL correlation to test independence from specific experimental datasets (BNL-E865 vs. NA48/2).

4. Numerical Procedure:

   • Generate random scattering lengths based on input distributions.
   • Reconstruct phase shifts using the Schenk parametrization (extended by DFGS to include matching point uncertainties).
   • Calculate dispersion integrals (moments) numerically.
   • Transform moments into phenomenological coefficients ($p_i$), then to chiral coefficients ($c_i$), and finally to subthreshold parameters ($\alpha_{\pi\pi}, \beta_{\pi\pi}, \lambda_i$).

3. Key Contributions

• Re-evaluation of Subthreshold Parameters: The paper provides updated, high-precision values for $\alpha_{\pi\pi}$ and $\beta_{\pi\pi}$ using the most recent experimental (NA48/2) and lattice (ETM, RBC/UKQCD) data.
• Resolution of the $\alpha_{\pi\pi}$ Discrepancy: The authors demonstrate that the high value of $\alpha_{\pi\pi}$ reported by DFGS is likely driven by the specific dataset (BNL-E865) rather than the theoretical correlation between scattering lengths.
• Independence from Theoretical Bias: By combining NA48/2 data with ETM lattice results, the authors successfully extracted parameters without relying on the CGL correlation (Eq. 41), confirming that the lower values of $\alpha_{\pi\pi}$ are robust.
• Statistical Rigor: The use of Monte Carlo sampling allows for a comprehensive modeling of probability distributions, capturing correlations (e.g., between $\alpha_{\pi\pi}$ and $\beta_{\pi\pi}$) and non-Gaussian uncertainties (specifically for ETM's $a_2^0$).

4. Key Results

The authors present results for several input combinations. The most significant findings are:

Main Result (NA48/2 Model C):

• $\alpha_{\pi\pi} = 1.053 \pm 0.071$
• $\beta_{\pi\pi} = 1.115 \pm 0.008$
• Correlation $\rho_{\alpha\beta} = -0.16$

Robust Result (NA48/2 + ETM Refit - No CGL Correlation):

• $\alpha_{\pi\pi} = 1.084 \pm 0.034$
• $\beta_{\pi\pi} = 1.111 \pm 0.008$
• Correlation $\rho_{\alpha\beta} = 0.66$

Comparison with Literature:

• CGL [7]: $\alpha_{\pi\pi} = 1.08 \pm 0.07$. The new results are highly compatible.
• DFGS [5] (Extended/Global fit): $\alpha_{\pi\pi} = 1.384 \pm 0.267$. The new results exclude the high mean value suggested by DFGS with high significance.
• GKPRY [12]: Results show larger uncertainties due to broader error bars in their scattering length determination.

Sensitivity Analysis:

• The results for $\alpha_{\pi\pi}$ and $\beta_{\pi\pi}$ are found to be negligibly sensitive to the uncertainties in the phase shifts at the matching point ($\theta_0, \theta_1$).
• However, other parameters ($\bar{b}_3, \bar{b}_5, \lambda_1, \lambda_3$) show high sensitivity to these phase shift uncertainties.

5. Significance

• Validation of 3-Flavor $\chi$PT: The obtained values of $\alpha_{\pi\pi}$ and $\beta_{\pi\pi}$ are close to unity (the Leading Order value). This supports the expectation of good convergence for these parameters in 3-flavor $\chi$PT, contradicting the large NLO corrections implied by the DFGS analysis.
• Implications for $Y^{(3)}$: The lower value of $\alpha_{\pi\pi}$ avoids the prediction of a suppressed leading-order pion mass ($Y^{(3)} \sim 0$). This resolves the tension with $\eta \to 3\pi$ data, which favors a larger chiral condensate ($Y^{(3)} \approx 1.44$).
• Future Applications: The authors intend to use these refined parameters to update the extraction of the 3-flavor LEC $B_0$ and to extend analyses of NLO LECs ($L_4, L_5$), providing a more consistent picture of low-energy QCD.

In conclusion, the paper successfully reconciles recent experimental and lattice data with theoretical expectations, confirming that the subthreshold parameters are consistent with standard 2-flavor $\chi$PT predictions and that the previously reported high values of $\alpha_{\pi\pi}$ were likely artifacts of specific data inputs rather than fundamental theoretical issues.