⚛️ high-energy theory

Monopoles, duality, and large charge in Chern-Simons QED $_3$

This paper computes the scaling dimensions of monopole operators in Chern-Simons QED $_3$ at large charge and large $N$ to provide strong quantitative evidence for proposed dualities with scalar QED $_3$ and to investigate the behavior of universal effective field theory terms in parity-breaking regimes.

Original authors: Shai M. Chester, Éric Dupuis

Published 2026-09-30

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Shai M. Chester, Éric Dupuis

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

In the microscopic world of quantum physics, particles do not always behave like the solid objects we see in daily life. Instead, they can exist as fields that ripple and interact across space. For decades, physicists have discovered that two completely different sets of rules can sometimes describe the exact same physical reality. This phenomenon, known as duality, suggests that a system made of one type of particle might be indistinguishable from a system made of a different type, provided they are viewed under the right conditions. This idea has been a powerful tool for understanding the universe, particularly in three-dimensional space where quantum effects are strong. However, proving these connections is notoriously difficult because the forces involved are often so intense that standard calculation methods break down. Researchers must rely on clever approximations or complex computer simulations to peer into these hidden relationships.

A team of physicists has now taken a significant step forward in verifying these dualities by calculating the properties of specific quantum objects called monopoles. In the context of this research, a monopole is not a particle in the traditional sense, but a special configuration of a field that carries a magnetic charge. Imagine a point where magnetic field lines all converge, similar to how a single electric charge creates an electric field, but for magnetism. In the specific theories studied here, these monopoles are not built from the basic ingredients of the theory but emerge from the collective behavior of the system. The researchers focused on two competing descriptions of nature: one involving fermions, which are the building blocks of matter like electrons, and another involving scalars, which are simpler field-like particles. Both descriptions include a mathematical twist called a Chern-Simons term, which acts like a background magnetic field that breaks the symmetry between left and right, a property known as parity.

The central question the team addressed was whether these two different descriptions truly lead to the same physical outcomes when the number of particle types and the strength of the magnetic twist are large. To answer this, they calculated the "scaling dimension" of the monopoles. In simple terms, this number tells us how the energy of the monopole changes as its charge increases. If the two theories are truly dual, they must predict the exact same energy values for monopoles of the same charge. The researchers performed a highly detailed calculation that went beyond the first, rough approximation to include subtle corrections. They computed these values for a wide range of charges and compared the results from the fermion-based theory against the scalar-based theory.

Their findings provide strong evidence that these dualities hold true, even in cases where the theories are complex and the symmetry between left and right is broken. For a specific case where the magnetic twist is set to a particular value, the team found that the lowest energy monopole in the fermion theory matched a value of approximately 2.09. This is remarkably close to the value of 2, which is predicted for a special type of conserved current that would appear if the symmetry of the system were enhanced to a higher level. For the next level of charge, their calculation yielded a value of about 5.28, which aligns closely with independent predictions made using a different, non-perturbative method known as the fuzzy sphere technique. For higher charges, the agreement between the two theories was even tighter, with the energy values matching to within just one percent. This level of precision suggests that the large-scale behavior of these theories is indeed identical, despite their microscopic differences.

The study also explored other variations of these dualities where the magnetic twist was adjusted to different values. In these cases, the agreement remained strong, with the energy values matching to within five or seven percent depending on the specific setup. However, the researchers noted that for one particular variation where the magnetic twist was very small, the two theories did not match well. In this scenario, the calculated values from the fermion theory were significantly different from the known values of the dual scalar theory. The authors suggest that this discrepancy might be due to the mathematical approximation they used becoming less reliable in this specific limit, or perhaps because the system behaves differently when the magnetic twist is minimal. Interestingly, they observed that if they ignored the main part of their calculation and looked only at the smaller correction terms, the results suddenly matched the dual theory very well, a curious coincidence that remains unexplained.

Beyond checking the dualities, the team investigated a universal property of these systems related to how energy scales with charge. In theories that preserve the symmetry between left and right, it is known that a specific constant term appears in the energy formula, regardless of the details of the system. The researchers found that for their fermion-based theory, even though it breaks this symmetry, this same constant term appeared with a value of approximately -0.0937. This suggests a surprising universality where the large-scale behavior of the system mimics that of a superfluid, a state of matter with zero viscosity. In contrast, when they performed the same analysis for the scalar-based theory, this constant term vanished completely, matching the behavior of a free particle. This difference highlights a subtle but important distinction between the two descriptions, even though they appear to describe the same physics in many other ways.

The work represents a rigorous test of theoretical physics, moving beyond simple checks to a detailed numerical comparison that covers a wide range of conditions. By calculating the energy of these exotic magnetic objects with high precision, the researchers have provided compelling evidence that the dualities proposed decades ago are correct. While some questions remain, particularly regarding the behavior at the smallest values of the magnetic twist, the overall picture is one of remarkable consistency. The results suggest that the universe may be far more interconnected than it appears, with different mathematical languages describing the same underlying reality. This deeper understanding of how quantum fields behave under extreme conditions could eventually help physicists unravel the mysteries of high-temperature superconductors and other complex materials where these quantum effects play a crucial role.

Technical Summary: Monopoles, duality, and large charge in Chern–Simons QED3

Problem Statement
The paper addresses the challenge of dynamically verifying non-supersymmetric dualities in three-dimensional gauge theories, specifically those relating Quantum Electrodynamics with fermions (QED3) to Scalar QED3 (sQED3) coupled to Chern-Simons (CS) terms. While particle-vortex dualities in 3d have been verified via lattice simulations, the presence of CS terms introduces a sign problem that prevents direct lattice comparison. The authors aim to test these dualities by computing the scaling dimensions of monopole operators in the large- $N$ , large- $k$ limit (where $N$ is the number of flavors and $k$ is the CS level, with $\kappa = k/N$ fixed) and extrapolating these results to the physically relevant regime of $N=1$ .

Methodology
The authors employ a saddle-point expansion of the thermal free energy on $S^2 \times \mathbb{R}$ in the presence of magnetic flux $4\pi q$ .

Large $N$ Expansion: They expand the free energy $F_q$ in powers of $1/N$ . The leading order ( $N^1$ ) term corresponds to the saddle-point evaluation, while the subleading order ( $N^0$ ) term arises from the functional determinant of fluctuations around the saddle point.
Saddle Point Configuration: For non-zero CS levels ( $\kappa \neq 0$ ), the saddle point involves a non-trivial holonomy $\alpha$ acting as a chemical potential. This requires "dressing" the bare monopole with charged matter modes. The authors identify the specific Landau levels that must be filled to satisfy the saddle-point equation, distinguishing between cases where the lowest Landau level is partially filled versus completely filled.
Subleading Calculation: The core technical contribution is the derivation of the subleading free energy term $\Delta^{(1)}_q$ $Δ_{q}^{(1)}$ . This involves:
- Constructing the thermal Green's function for fermions in the background of the monopole and the saddle-point holonomy.
- Calculating the "matter kernels" (fluctuation determinants) by expanding the gauge field fluctuations in vector spherical harmonics.
- Separating the kernels into CP-preserving and CP-violating components, the latter arising from the non-zero holonomy.
- Deriving a compact formula for the subleading scaling dimension expressed as an infinite sum and integral over angular momentum modes ( $\ell$ ) and frequency ( $\omega$ ).
Numerical Implementation: The authors implement a numerical algorithm using relativistic cutoffs and asymptotic subtraction (zeta regularization) to compute the divergent sums and integrals with high precision. They extrapolate the large- $N$ results to $N=1$ by assuming the subleading term captures the dominant finite- $N$ corrections.

Key Contributions

First Subleading Calculation for QED3: The paper generalizes previous subleading large- $N$ calculations (previously done for $k=0$ QED3 and $k \neq 0$ sQED3) to the case of QED3 with non-zero CS level ( $k \neq 0$ ).
Compact Formula for Arbitrary $\kappa$ : They derive a unified formula for the subleading scaling dimension that depends on the number of filled Landau levels and the filling fraction of the valence level, applicable for any fixed $\kappa$ .
Systematic Duality Checks: The authors systematically test a family of dualities labeled by an integer $m$ , relating QED3 with $N=1, k=-(2m+1)/2$ to sQED3 (or effective descriptions thereof) with specific parameters.

Results
The authors compare their extrapolated QED3 results against known data for the dual theories:

Case $m=1$ ( $N=1, k=-3/2$ QED3 $\leftrightarrow$ $N=1, k=2$ sQED3):
- The $q=1/2$ monopole scaling dimension is found to be $\Delta \approx 2.096$ , which is close to the expected value of $\Delta=2$ for the emergent SO(3) conserved currents.
- The $q=1$ monopole dimension is $\Delta \approx 5.275$ , matching closely with fuzzy sphere calculations ( $\Delta \approx 5.11$ or $5.34$) for spin-0 or spin-2 operators.
- For higher charges ( $q > 1$ ), the QED3 results match the corresponding sQED3 results with an average relative error of approximately 1%. The authors note that the QED3 extrapolation appears more accurate than the sQED3 extrapolation for low $q$ in this specific duality.
Cases $m=2, 3$ :
- For $m=2$ ( $k=-5/2$ ) and $m=3$ ( $k=-7/2$ ), the QED3 results match the dual sQED3 effective descriptions with average relative errors of 5% and 7%, respectively.
Case $m=0$ ( $N=1, k=-1/2$ QED3 $\leftrightarrow$ Critical O(2)):
- The extrapolated results do not match the known scaling dimensions of the critical O(2) model for any $q$ . The authors observe that the subleading term alone matches the O(2) data with high accuracy, while the leading term ruins the match, suggesting the large- $N$ expansion may diverge or behave poorly in this specific limit.
Large Charge EFT Analysis ( $q_0$ term):
- By fitting the large- $q$ behavior of the scaling dimensions, the authors extract the $q^0$ term.
- For QED3 with $k \neq 0$ , the $q^0$ term is found to be approximately $-0.0937$, matching the universal value predicted for parity-preserving superfluid EFTs, despite the theory breaking parity.
- For sQED3 with $k \neq 0$ , the $q^0$ term is found to vanish, consistent with the duality to a free fermion (which has a vanishing $q^0$ term).

Significance and Claims
The paper claims that the large- $N$ , large- $k$ expansion of monopole scaling dimensions provides a powerful, non-perturbative tool for checking 3d bosonization dualities, even when extrapolated to small $N$ and $k$ .

Evidence for Dualities: The close agreement (within ~1% for $m=1$ ) between the QED3 and sQED3 calculations for $m=1$ provides strong dynamical evidence for the duality between $N=1, k=-3/2$ QED3 and $N=1, k=2$ sQED3, including the enhancement of the U(1) symmetry to SO(3).
Universality of the $q_0$ Term: The results suggest a tension regarding the universality of the large-charge EFT $q^0$ term. While the term is universal for $k=0$ , the authors find it is non-zero for QED3 ( $k \neq 0$ ) but zero for sQED3 ( $k \neq 0$ ). This contradicts the expectation that the $q^0$ term should be identical for dual theories at finite $N, k$ , implying that the extrapolation from the large- $N$ limit to finite $N$ is subtle and theory-dependent.
Limitations: The authors modestly note that the failure of the $m=0$ duality check and the worsening accuracy for filled Landau levels indicate that the large- $N$ expansion is not uniformly reliable across all parameters, particularly when the lowest Landau level is completely filled.

The work does not propose new experimental setups but rather refines the theoretical toolkit for understanding strongly coupled 3d CFTs and their dualities.

Technical Summary: Monopoles, duality, and large charge in Chern–Simons QED3

More like this