Operator product expansion on the lattice: a numerical test in the two-dimensional non-linear sigma-model

Sergio Caracciolo, Andrea Montanari, Andrea PelissettoJournal of High Energy Physics 2000 · DOI

TL;DR

This paper presents a numerical test of the Operator Product Expansion (OPE) in the two-dimensional non-linear sigma-model. The authors demonstrate that the OPE provides an accurate description of correlation functions on relatively small lattices, paving the way for its application in more complex theories like Quantum Chromodynamics (QCD).

What Problem Does This Paper Address?

In quantum field theory, understanding the behavior of physical quantities at short distances is crucial. The Operator Product Expansion (OPE) is a powerful theoretical tool that allows us to analyze these short-distance behaviors by expressing products of local operators as a sum of local operators with coefficients that depend on the distance. However, applying OPE to numerical simulations, especially in the context of lattice field theory, presents significant challenges. These include dealing with lattice artifacts, finite-size effects, and the need for nonperturbative renormalization of operators. This paper aims to bridge this gap by numerically verifying the validity of the OPE in a well-understood, yet non-trivial, model – the two-dimensional non-linear sigma-model – and assessing its accuracy in a realistic lattice setting.

Methods

The authors employ a combination of numerical simulations and theoretical calculations. They simulate the O(3) non-linear sigma-model using a Swendsen-Wang cluster algorithm with Wolff embedding. The core of their method involves:

  1. Computing Lattice Correlation Functions: They calculate four-point correlation functions, which are the lattice analogues of the matrix elements needed for OPE.
  2. Using Perturbative OPE Predictions: They compare these numerical results with predictions derived from the Operator Product Expansion, utilizing RG-improved perturbation theory for the Wilson coefficients.
  3. Nonperturbative Renormalization: They address the challenge of renormalization by focusing on conserved currents, which do not require renormalization, and by carefully considering the definition of renormalized operators.
  4. Finite-Size Scaling: They use multiple lattice sizes to study finite-size effects and extrapolate to the continuum limit.
  5. RG-Improved Perturbation Theory: They leverage Renormalization Group (RG) improved perturbation theory to obtain more reliable predictions for the running coupling constant and Wilson coefficients.

Datasets

The study utilizes simulated data from the O(3) non-linear sigma-model on three different lattice sizes:

  • Lattice (A): 128 × 64
  • Lattice (B): 256 × 128
  • Lattice (C): 512 × 256

These simulations were performed with varying bare coupling constants ($g_L$) to probe different regimes. No new benchmark datasets were introduced; the focus was on testing the OPE against simulated data from a well-established model.

Key Insights and Ideas

  • OPE Works on Small Lattices: A key finding is that the OPE provides a remarkably accurate description of correlation functions even on relatively small lattices (where lattice spacing $a$ is not negligible compared to the physical scale). This is a crucial result for applying OPE to more complex theories where large lattices are computationally prohibitive.
  • Window for OPE Validity: The authors identify a "window" of distances where the OPE is reliable, typically for $2a \lesssim |x| \lesssim m^{-1}$, where $m$ is the mass gap. This window is broad enough to be practically useful.
  • Systematic Errors are Manageable: While statistical errors are significant, the authors demonstrate that systematic errors, particularly those arising from the perturbative truncation of Wilson coefficients, can be controlled. Using RG-improved perturbation theory and higher-loop calculations significantly improves the agreement.
  • Nonlinear Sigma-Model as a Testbed: The two-dimensional non-linear sigma-model serves as an excellent testbed due to its asymptotic freedom and nonperturbative mass gap, properties it shares with QCD, making the results highly relevant.

Key Results

  • The OPE provides an accurate description of correlation functions with an error of 5-10% in a large window of distances.
  • The numerical results show good scaling behavior, with lattice artifacts being negligible compared to statistical and systematic errors on the simulated lattices.
  • The running coupling constant can be determined with high precision using various methods, with errors being practically negligible for the OPE applications.
  • The study highlights the importance of RG-improved perturbation theory for obtaining reliable OPE predictions.

Limitations and Future Work

  • Statistical Errors: The authors acknowledge that statistical errors are a significant limitation, especially when trying to resolve finer details or higher-order effects in the OPE.
  • Perturbative Truncation: While RG improvement helps, the perturbative truncation of Wilson coefficients remains a primary source of systematic error.
  • Future Work: The paper suggests that this successful demonstration of OPE on the lattice opens doors for its application to more complex theories like QCD. Further work could involve extending these studies to include more complex operators and exploring different renormalization schemes. The authors also mention the possibility of using improved actions to further reduce lattice artifacts.

Summary updated

BibTeX
@article{caracciolo2000operator,
  author = {Sergio Caracciolo and Andrea Montanari and Andrea Pelissetto},
  title = {{Operator product expansion on the lattice: a numerical test in the two-dimensional non-linear sigma-model}},
  journal = {Journal of High Energy Physics},
  year = {2000},
  doi = {10.1088/1126-6708/2000/09/045},
}

Loading related content...

Figure