Circuit locality from relativistic locality in scalar field mediated entanglement

1. Problem Statement

The paper addresses a fundamental conceptual gap between two distinct notions of locality in modern physics:

• Relativistic Locality (Spacetime): Rooted in Quantum Field Theory (QFT), this is defined by microcausality (Einstein causality), where field operators at spacelike separations commute ($[\hat{\phi}(x), \hat{\phi}(x')] = 0$). This prevents superluminal signaling.
• Circuit Locality (Subsystem): Rooted in Quantum Information Theory (QIT), this defines locality via the structure of quantum circuits. Information flows only through direct interactions or chains of gates connecting subsystems. A system is "local" if its evolution can be decomposed into unitaries acting on specific subsystems (e.g., $U = U_A U_B$).

The Core Question: Can relativistic locality (microcausality) mathematically enforce circuit locality? Specifically, if two quantum systems ($A$ and $B$) interact only via a relativistic scalar field $\phi$ (and not directly with each other), does the evolution of the total system decompose into a circuit where $A$ and $B$ only interact with $\phi$, but never directly with each other?

This is non-trivial because the Hilbert space of a QFT generally cannot be decomposed into a tensor product of spatial regions, making the mapping from spacetime locality to subsystem locality an open problem. This issue is critical for interpreting recent proposals to test the quantum nature of gravity via gravity-mediated entanglement (GME).

2. Methodology

The authors employ a rigorous derivation within the framework of a massive scalar field coupled to two quantum sources, utilizing the following steps:

• System Setup:

  • Two particles ($A$ and $B$) with internal degrees of freedom and center-of-mass motion.
  • Each particle is coherently controlled by a "qudit" (a $d$-level system), placing the particles in a quantum-controlled superposition of localized trajectories.
  • The particles couple to a massive scalar field $\phi$ via a charge density operator $\hat{\rho}$.
  • The Hamiltonian is $\hat{H} = \hat{H}_A + \hat{H}_B + \hat{H}_0 + \hat{H}_{int}$.

• Approximations:

  • Parametric Approximation: The authors assume the back-reaction of the field on the particles and the qudits is negligible. This allows the state to be treated as a superposition of branches, where in each branch ($|rs\rangle$), the particles act as classical, time-dependent sources for the field.
  • Localized Sources: Within each branch, the particles are assumed to be sufficiently localized such that their wavefunctions have negligible support outside specific spacetime regions.

• Mathematical Tools:

  • Magnus Expansion: Instead of a standard Dyson series, the authors use the Magnus expansion to solve the time-evolution of the field coupled to the classical sources. This is crucial for handling the non-commutativity of the interaction Hamiltonian at different times.
  • Baker-Campbell-Hausdorff (BCH) Formula: Used to disentangle the evolution operator of the field when two sources are present.
  • Microcausality: The commutator of field operators $[\hat{\phi}_I(t, x), \hat{\phi}_I(t', x')]$ is explicitly evaluated.

3. Key Contributions and Derivation

A. The Evolution Operator Structure

The authors derive the exact evolution operator for the field in the presence of two sources, $\rho_A$ and $\rho_B$. Using the Magnus expansion, they show that the interaction picture propagator $\hat{U}_I$ takes the form:
$$\hat{U}_I = e^{\Omega_2} e^{\hat{\Omega}_1}$$
where $\hat{\Omega}_1$ is linear in the sources and $\Omega_2$ is a c-number phase arising from the commutator of the Hamiltonian at different times.

When splitting the sources ($\rho = \rho_A + \rho_B$), the BCH formula reveals that the total evolution operator factorizes into individual source evolutions only up to a phase factor $\Omega_{cross}$:
$$\hat{U}_{\rho} = e^{i\Omega_{cross}} \hat{U}_{\rho_B} \hat{U}_{\rho_A}$$
The phase $\Omega_{cross}$ depends on the commutator of the field operators integrated over the spacetime supports of the two sources:
$$\Omega_{cross} \propto \int \int \rho_A(x) \rho_B(x') [\hat{\phi}(x), \hat{\phi}(x')]$$

B. The Role of Microcausality

The central result is the condition under which this phase vanishes:

• If the supports of the charge densities $\rho_A$ and $\rho_B$ are spacelike separated for the duration of the interaction, the commutator $[\hat{\phi}(x), \hat{\phi}(x')]$ vanishes due to microcausality.
• Consequently, $\Omega_{cross} = 0$.
• The evolution operator factorizes strictly: $\hat{U}_{total} = \hat{U}_{A\phi} \hat{U}_{B\phi}$.

This proves that relativistic locality enforces circuit locality: if the particles are spacelike separated, the field mediates the interaction in a way that allows the total evolution to be decomposed into a circuit where $A$ and $B$ never interact directly, only via the field.

C. Entanglement Generation

The paper analyzes the generation of entanglement between the control qudits (which control the particle positions).

• If the particles are in a superposition of trajectories, the field acquires different phases depending on the branch.
• The entanglement phase $\theta_{rs}$ is derived as the on-shell action of the field:
  $$\theta_{rs} = -\frac{1}{2} \int d^4x \, \rho_{rs}(x) \phi_{rs}(x)$$
• Crucially, this phase (and thus the entanglement) is non-zero only if there is sufficient time for a light signal to travel between the particles ($t_f - t_i > d_{min}$). This aligns perfectly with relativistic causality.

4. Results

1. Derivation of Field Mediation: The authors demonstrate that for systems in quantum-controlled superpositions of localized states, the evolution of a scalar field coupled to them can be written as a field mediation circuit (Eq. 5 and 6 in the paper) if and only if the particles remain spacelike separated.
2. Necessity of the Phase: They highlight that standard textbook treatments often neglect the global phase factor ($\Omega_2$) in field evolution. However, in the context of quantum superpositions, this phase is essential for the groupoid property of unitaries and for correctly capturing the causal structure.
3. Causal Constraint on Entanglement: The entanglement generated between the control systems is strictly bounded by the speed of light. Entanglement cannot be generated if the particles are spacelike separated for the entire duration of the experiment.

5. Significance

• Bridging QFT and QIT: The paper provides a rigorous mathematical link showing how the spacetime notion of locality (microcausality) implies the subsystem notion of locality (circuit decomposition) under specific physical conditions (localized states).
• Validation of GME Experiments: The results strengthen the theoretical foundation for recent proposals to test the quantum nature of gravity (e.g., the Bose-Marletto-Vedral proposal). These proposals rely on the premise that if gravity mediates entanglement, gravity must be non-classical. The authors show that in a relativistic QFT, a classical field can mediate entanglement, but only if the interaction respects relativistic causality (i.e., the field acts as a mediator in a circuit-local way). This clarifies that the "no-go" theorems for classical mediators rely on specific definitions of locality that are satisfied by relativistic fields.
• Limitations and Future Work: The authors acknowledge that their derivation relies on the parametric approximation and massive scalar fields. They note that for massless gauge fields (like electromagnetism or gravity), infrared divergences and gauge constraints (e.g., Coulomb gauge) complicate the direct factorization. However, they argue that the core physical intuition likely holds, suggesting a path for future research in linearized quantum gravity.

In summary, the paper establishes that microcausality is the mechanism that enforces circuit locality in relativistic quantum field theories, providing a precise condition under which field-mediated interactions can be modeled as local quantum circuits.