Gauge Invariance Kills the Floquet Superradiant Phase Transition — What You Actually Get Is Dynamical Casimir Emission

A predicted exotic phase transition in driven cavity QED turns out to be a gauge artifact. The real physics is vacuum pair creation with squeezed photons up to 10 dB.

Imagine you're tuning a guitar string by tightening the peg while someone else keeps plucking it. You think you're making the string resonate harder — coupling it more strongly to the sound hole — but what's really happening is the changing tension is creating new vibrations from the string's own zero-point motion. That's the core of this paper: modulating a magnetic field in a terahertz cavity doesn't enhance the light-matter coupling into a new phase of matter. It shakes the vacuum hard enough to produce real photon pairs. The committed claim: a Floquet superradiant phase transition (SRPT) predicted for periodically driven Landau polaritons does not exist. It is a gauge artifact arising from averaging the coupling constant inside a fixed Landau-level basis — a projection that violates the Thomas-Reiche-Kuhn (TRK) sum rule. When you work in the full, gauge-invariant minimal-coupling Hamiltonian and include the Faraday electric field induced by the modulated magnetic flux, Kohn's theorem collapses the problem to a single center-of-mass mode coupled to the cavity. A Krein stability theorem then guarantees the system cannot undergo an instability above a computable drive frequency. Below it, resonance tongues produce dynamical Casimir emission — not a phase transition. Where this lands on the ladder: the target is the 2019–2024 line of papers predicting Floquet-enhanced ultrastrong coupling and Floquet SRPTs in Landau polariton systems. The most prominent claim Hughes is contesting is from groups modeling magnetic-field modulation as a time-dependent coupling parameter while holding the Landau-level basis fixed. Hughes shows this is equivalent to silently dropping the diamagnetic (A-squared) term's time dependence — exactly the gauge-violation error that the no-go theorems of Rza̧żewski, Wódkiewicz, and others identified for static SRPTs decades ago. The paper doesn't just disagree; it derives the specific sum-rule violation responsible. Architecturally, this is analytic theory — no numerics, no simulation. The toolchain is Floquet-Bogoliubov analysis on a bosonic Hopfield Hamiltonian reduced via Kohn's theorem, stabilized via Krein's theorem from Hamiltonian dynamical systems. The generalized TRK sum rule for Kane nonparabolicity is a new derivation. The key structural insight is that the Hamiltonian must remain a 'sum of squares' (positive-semidefinite in field operators) at every instant, not just cycle-averaged. This is a pen-and-paper result; the compute dependency is zero. Integrity is unusually clean for a theory paper. The argument is a chain of exact theorems — Kohn, Krein, TRK — applied to a well-defined model. There is no fitting, no free parameters, no cherry-picked comparison. The falsifiability condition is explicit: if any gauge-invariant formulation of periodically driven Landau polaritons produces a photon condensate, the paper is wrong. The predicted observable — two-color squeezed photon pairs with 5–10 dB of two-mode squeezing at specific drive frequencies — is a concrete experimental target. The self-contained proof of the Krein bound is a nice touch; it means the stability claim doesn't rely on citing results from an unfamiliar mathematical literature. The milestone here is experimental. The dynamical Casimir effect has been observed in superconducting circuits (Wilson et al. 2011, ~1 GHz). Observing it in a Landau polariton system at THz frequencies with measurable two-mode squeezing would be a qualitative advance. The paper predicts 5–10 dB squeezing and identifies the resonance-tongue structure where emission should occur, giving experimentalists a map. The gap is fabricating a THz cavity with ultrastrong coupling AND a fast, clean magnetic-field modulation — nontrivial but within the trajectory of current 2DEG-in-split-ring-resonator platforms. The experiment Hughes did not run — and could not, being a theorist — is the actual THz measurement. But within theory, the obvious missing piece is a full numerical Floquet-Bogoliubov simulation including realistic dissipation (cavity loss, Landau-level broadening). The paper works in the lossless limit. This is almost certainly (a) — the analytic machinery is the contribution, and adding dissipation would require numerics that are a different paper. The gauge theorem itself is lossless and exact, so dissipation wouldn't change the no-go, but it would reshape the predicted squeezing spectrum and pair-production rates.