Generalized unitary evolution for symplectic scalar fermions

Technical Summary: Generalized Unitary Evolution for Symplectic Scalar Fermions

Problem Statement
The paper addresses the tension between the conventional spin-statistics theorem and the construction of local, Lorentz-invariant quantum field theories (QFT) where scalar fields obey fermionic statistics. Standard QFT postulates that physical operators, particularly the Hamiltonian, must be Hermitian to ensure real eigenvalues and unitary time evolution. This requirement, combined with Lorentz symmetry, enforces the spin-statistics theorem, which dictates that scalar fields must be bosonic. LeClair and Neubert (LN) previously constructed a theory of "symplectic scalar fermions" that evades this theorem by utilizing a pseudo-Hermitian Hamiltonian. However, the consistency of this theory in the presence of interactions, specifically regarding the definition of time evolution, inner products, and the unitarity of the S-matrix, requires further clarification. The central problem is to demonstrate that the LN model admits a consistent, generalized unitary evolution despite the non-Hermitian nature of its interacting Hamiltonian.

Methodology
The author employs the framework of pseudo-Hermitian quantum mechanics, originally developed by Bender, Boettcher, and Mostafazadeh, applied to the specific LN model of $N$ self-interacting scalar fermions.

1. Pseudo-Hermiticity Definition: The study defines a pseudo-Hermitian Hamiltonian $H$ via the relation $H^\# \equiv \eta^{-1}H^\dagger\eta = H$, where $\eta$ is a linear, Hermitian, invertible operator. Unlike standard Hermitian theories, $\eta$ is not necessarily positive-definite.
2. Field Definitions and $\eta$ Construction: The paper reviews the LN construction where the adjoint field $\bar{\phi}$ is distinct from the Hermitian conjugate $\phi^\dagger$. The author explicitly constructs the operator $\eta$ in terms of the anti-particle number operator: $\eta = \exp[-i\pi \int d^3p \, b^\dagger(p)b(p)]$. This operator ensures that $\bar{\phi} = \eta^{-1}\phi^\dagger\eta$, rendering the free Lagrangian and interaction terms pseudo-Hermitian.
3. Symmetry Analysis: The global symmetry of the free Lagrangian is analyzed. While the unconstrained system possesses $Sp(2, \mathbb{C})$ symmetry, the requirement that currents and charges be pseudo-Hermitian reduces this global symmetry to $SU(2)$. The author derives the corresponding conserved currents and charges, demonstrating they satisfy the $su(2)$ algebra.
4. Generalized Unitarity Formalism: To handle time evolution under a non-Hermitian Hamiltonian, the paper adopts the $\eta$-product inner product, $\langle \beta | \alpha \rangle_\eta \equiv \langle \beta | \eta | \alpha \rangle$. This preserves time translation symmetry. The S-matrix is defined using this inner product, and the condition for generalized unitarity is established as $S^\# S = I$, where $S^\# = \eta^{-1}S^\dagger\eta$.
5. Scattering Calculations: The author performs explicit calculations for the $N$-field LN model with quartic self-interactions. The analysis focuses on $2 \to 2$ scattering processes (particle-particle, particle-antiparticle, and antiparticle-antiparticle scattering) below the pair-production threshold. The generalized optical theorem is derived and verified at the one-loop level.

Key Contributions and Results

• Consistency of Pseudo-Hermiticity in the Heisenberg Picture: The paper clarifies that the definition of pseudo-Hermiticity is consistent across the Schrödinger and Heisenberg pictures. It demonstrates that if an operator is pseudo-Hermitian at $t=0$, it remains so at all times under the evolution of a pseudo-Hermitian Hamiltonian, provided the time evolution of the fields and their pseudo-Hermitian conjugates is consistent.
• Symmetry Reduction: A key finding is that demanding the currents and charges to be pseudo-Hermitian reduces the global symmetry of the free Lagrangian from $Sp(2, \mathbb{C})$ to $SU(2)$. The resulting charges map particle states to anti-particle states and vice versa, satisfying the $su(2)$ algebra.
• Generalized Unitarity Relation: The paper derives the generalized unitarity relation $S^\# S = I$. It highlights a critical distinction from standard unitary theories: the transition probability is not simply $|M|^2$. Instead, consistency requires that the number of anti-particles be conserved in the scattering process for the generalized unitarity to reduce to standard unitarity.
• Verification in the LN Model: For the specific case of $2 \to 2$ scattering in the LN model below the pair-production threshold, the number of anti-particles is conserved ($n_\alpha = n_\beta$). In this regime, $M^\#_{\alpha\beta} = M^\dagger_{\alpha\beta}$, and the generalized optical theorem is satisfied. The author explicitly calculates the one-loop amplitudes and cross-sections, showing that the imaginary parts of the amplitudes match the predictions of the generalized optical theorem.
• Limitations: The paper notes that unitarity is spoiled in processes where the number of anti-particles changes, such as $2 \to 4$ scattering (pair production) at high energies. Consequently, the theory does not satisfy standard unitarity ($S^\dagger S = I$) in all regimes, only the generalized relation.

Significance and Claims
The paper claims to provide a rigorous demonstration that the LeClair-Neubert model of symplectic scalar fermions admits generalized unitary evolution. By showing that the S-matrix satisfies the generalized unitarity relation $S^\# S = I$, the work extends the spin-statistics theorem, proving that local, Lorentz-invariant theories with fermionic scalar fields are possible if the Hamiltonian is pseudo-Hermitian.

The author posits that this framework is of fundamental importance as an extension of standard quantum mechanics. While the theory is generally non-unitary in the conventional sense due to the indefinite nature of the $\eta$ inner-product, the paper suggests potential applications in condensed matter physics where $2 \to 2$ scattering dominates below pair-production thresholds. Additionally, the paper mentions the relevance of symplectic fermions to the $dS/CFT$ correspondence, where strict unitary evolution may not be essential. The work also draws parallels between the quartic self-interaction of these scalar fermions and the Gross-Neveu and Nambu-Jona-Lasinio models, noting the unique feature of renormalizable quartic interactions in four dimensions for scalar fermions.

The paper remains modest regarding broader implications, acknowledging that the non-conservation of anti-particle number in general processes prevents the theory from being a fully unitary replacement for standard QFT in all contexts. It identifies the study of the massless LN model in the large $N$ limit and dynamical mass generation as a task for future investigation.