Finite Weyl Polynomials Push Quantum Field Vacuum to the Edge of Tsirelson’s Bound

Physicists studying the deepest structure of quantum fields have long known that the vacuum of relativistic quantum field theory is not as empty as it seems. In a new theoretical study published in The European Physical Journal C, researchers J. G. A. Caribé, M. S. Guimaraes, I. Roditi and S. P. Sorella have constructed explicit, finite families of quantum observables for a free massive scalar field in 1+1 dimensions that violate the Bell-CHSH inequality in the vacuum, and they prove that these observables can drive the Bell correlation all the way up to Tsirelson’s bound of 2√2, the famous quantum maximum that no theory with stronger-than-quantum correlations can exceed. The work turns an abstract existence result into a concrete recipe with explicit numbers.

The context goes back to landmark results by Summers and Werner, who showed that in free bosonic and fermionic field theories, measurements performed in suitably chosen space-like separated regions can exhibit maximal Bell correlations in the vacuum. That was a striking statement: the vacuum state, seemingly featureless, encodes correlations as strong as quantum mechanics permits. But the Summers-Werner analysis was existential. It established that bounded local operators with maximal Bell correlations exist, without supplying simple bosonic observables or a finite set of coefficients that a working theorist could write down. The new paper aims to close exactly that gap for the massive scalar field.

A natural first attempt immediately runs into trouble. The most obvious bounded observables built from a scalar field are functions of a single smeared field, such as the cosine, hyperbolic tangent, or spectral projections of a field quadrature. The authors prove a restriction: if all four settings of a Bell-CHSH test are bounded functions of individual field quadratures in a centered quasifree state, then all their correlations derive from one common Gaussian probability measure on the four amplitudes. Importantly, the Hadamard cross-correlations between the regions are not discarded; they appear as off-diagonal entries of the Gaussian covariance matrix. Yet the existence of this single, context-independent positive measure is precisely the classical situation, and the CHSH value cannot exceed 2. The vacuum correlations are there, but this class of observables cannot reveal them.

The way around the obstruction lies in the full local Weyl algebra, whose elements do not reduce to functions of one quadrature. Weyl operators, the exponentials of smeared fields, carry a symplectic phase: the product of two Weyl operators picks up a phase determined by the Pauli-Jordan distribution of their smearing directions. A finite Hermitian sum of Weyl operators along two noncommuting directions is therefore not a function of any single quadrature and admits no common Gaussian representation. The authors exploit this by building, in each of two complementary space-time wedges, pairs of bounded axis observables made from trigonometric polynomials containing only odd harmonics. When the underlying one-particle directions are symplectically normalized, the two axis observables in each wedge anticommute exactly.

Exact anticommutation is the technical key that keeps everything bounded. If two bounded Hermitian operators S and T anticommute, then their normalized sum and difference, each divided by √2, are again bounded by one, because their squares equal half of S squared plus T squared. Alice’s two measurement settings are the two axis observables in the right wedge; Bob’s are the normalized sum and difference of the axis observables in the left wedge. Because the wedges are space-like separated, Alice’s operators commute with Bob’s, satisfying the locality requirement, while the noncommutativity within each wedge opens the door to violations above the classical bound of 2.

Localization in the wedges is achieved through modular theory, a cornerstone of algebraic quantum field theory. By the Bisognano-Wichmann construction, the modular operator of a wedge is the exponential of the boost generator, and the authors choose a normalizable one-particle packet with flat spectral density supported in a compact band of the boost spectrum. This choice yields every needed inner product in closed, exact form at nonvanishing bandwidth, avoiding the improper, zero-width objects that sometimes haunt such constructions. In the narrow-band limit the covariance data reduce to those of a familiar two-mode squeezed vacuum, providing a consistency check and connecting the construction to earlier parity-pseudospin approaches.

The resulting Bell-CHSH correlator collapses to a finite double sum, and the numbers are striking. An economical polynomial with only six Weyl operators per axis, three odd harmonics each, already yields a violation of 2.14885 at an exactly normalizable finite-band point with parameters λ = 0.99 and d = 0.001. Pushing further with a normalized Fejér polynomial of degree 511, containing 512 Weyl operators per axis observable, at parameters λ = 0.99999 and d = 10⁻⁶, the authors compute a CHSH value of 2.80027, within a hair’s breadth of Tsirelson’s bound of about 2.82843. Every entry in their tables uses normalizable, finite-bandwidth packets, no improper modular eigenvectors required.

The central theorem is sharper still. Using normalized Fejér approximants of the square-wave-like function sgn(cos x), which converge to the sharp sign function only in the infinite-degree limit, the authors show that the supremum of the vacuum CHSH value over their entire finite-polynomial family equals exactly 2√2, even though no finite member ever attains it. For every tolerance ε, however small, some finite polynomial and finite-bandwidth packet delivers a value strictly between 2√2 minus ε and 2√2. The proof of strictness is elegant: the vacuum spectral distribution of any smeared field is a nondegenerate Gaussian with full support, so the vacuum mean square of any continuous, nonconstant, bounded polynomial is strictly less than one, keeping the correlator strictly below the contraction bound at every finite stage. Convergence happens in the vacuum mean square, not in operator norm.

The approach to the maximum also carries conceptual weight. The parameter λ is tied to the boost frequency ω₀ through λ² = e^(−2πω₀), so sending λ toward 1 means concentrating the spectral packet near the zero of the modular, or boost, generator. The authors emphasize that this is not merely a convenient numerical regime: it singles out the modular spectral region near eigenvalue one, naturally associated with the type III₁ factor structure of local algebras that underlies the Summers-Werner results. In other words, the explicit construction traces maximal vacuum Bell correlations back to the modular architecture of relativistic quantum theory, while both the polynomial degree and the one-particle norms must grow as the bound is approached.

The authors also note that their finite Weyl polynomials are unproblematic from the standpoint of measurement theory in quantum field theory, since within the probe-based framework of Fewster and Verch every local observable admits asymptotic measurement schemes. Looking forward, they suggest that alternative harmonic regularizations, such as Abel-Poisson summation, which preserves the odd-harmonic structure needed for exact anticommutation while damping Fourier modes geometrically, might yield more efficient routes to maximal violation at fixed polynomial degree. For now, the message is clear: the vacuum of a relativistic quantum field can host correlations as strong as quantum mechanics permits, and with the right finite, explicit, bounded observables, it will say so at nearly the maximal volume quantum mechanics allows.

Subject of Research: Explicit bounded Weyl-polynomial observables that approach maximal Bell correlations in the vacuum of a 1+1 dimensional free massive scalar quantum field theory

Article Title: Finite Weyl polynomials and the approach to Tsirelson’s bound in relativistic scalar quantum field theory

Article References: Caribé, J. G. A., Guimaraes, M. S., Roditi, I., & Sorella, S. P. (2026). Finite Weyl polynomials and the approach to Tsirelson’s bound in relativistic scalar quantum field theory. The European Physical Journal C, 86(10), Article 1224. https://doi.org/10.1140/epjc/s10052-026-16508-z

Image Credits: AI Generated

DOI: 10.1140/epjc/s10052-026-16508-z

Keywords: Bell-CHSH inequality, Tsirelson bound, quantum field theory, Weyl operators, modular localization, scalar field, vacuum entanglement, algebraic QFT, Bisognano-Wichmann, quasifree state, Fejér approximants, space-like separation