arXiv:2609.20559
High-order correlations and ultrafast Wigner negativities in bright-squeezed-vacuum-driven high-harmonic generation
Sebastián de-la-Peña, Heiko Appel, Marcelo F. Ciappina, Ofer Neufeld, Angel Rubio
Abstract
High-harmonic generation (HHG) is a prototypical strong-field process in which intense light drives matter to emit radiation at integer multiples of the driving frequency. Extending HHG into the quantum-optical regime offers new opportunities to probe and control strongly nonlinear light-matter interactions using nonclassical states of light. Yet describing this regime requires a fully quantum treatment of the correlated electron-photon dynamics, which becomes computationally challenging for broadband, strongly squeezed fields. Here we solve the quantum-electrodynamical dynamics of a two-level system driven by bright squeezed vacuum in a converged multimode Hilbert space. Both the driving field and emitted harmonics are fully quantized, with the light-matter interaction treated nonperturbatively. This enables direct access to the multimode quantum state and its higher-order correlations beyond semiclassical sampling or perturbative descriptions. We show that squeezed-vacuum driving produces harmonic emission with qualitatively distinct second- and third-order photon correlations compared with coherent excitation. Moreover, back-action from the driven emitter strongly reshapes the incident squeezed field, generating pronounced Wigner-function negativities that evolve on attosecond timescales. Our results establish a fully quantum framework for broadband strong-field dynamics with squeezed light and provide a route to predicting and interpreting quantum-HHG experiments and their extension to more complex emitters.
AI Summary
The paper asks whether bright-squeezed-vacuum-driven high-harmonic generation produces quantum effects beyond those captured by sampling classical driving fields, particularly when the pump photon number becomes large. It propagates a two-level system coupled nonperturbatively to one quantized squeezed driving mode and a discretized continuum of quantized emission modes. A squeezed-frame transformation makes the initial pump vacuum and permits the joint limit $|\xi|\to\infty$, $\Delta\omega\to0$ at fixed driving strength $\Omega_R=\kappa e^{|\xi|}$, where $\kappa\propto\sqrt{\Delta\omega}$. Calculations examine nominal gap-to-frequency ratios $\Delta/\omega_L=1$ and $5$, with spectral coupling $g\approx5\times10^{-4}$. Photon occupations show a stronger cutoff dependence on driving strength and less resolved spectral structure under squeezed-vacuum excitation than under coherent driving. Directly calculated second- and third-order photon moments, including three-cycle calculations at $\Omega_R/\omega_L=2$, exhibit harmonic-dependent resonance patterns. Tracing out matter and emission modes yields the pump state, whose reconstructed Wigner function develops negative regions. Its integrated negativity, invariant under the squeezing transformation, evolves on attosecond timescales, peaks near minima of the initial pump-field variance, and develops additional negative lobes with successive half-cycles. This pump-state negativity is the stated signature beyond positive phase-space sampling. The matter model is explicitly phenomenological; extensions to realistic multilevel or collective systems remain prospective. Emission calculations retain selected harmonic-frequency windows to reduce computational cost. Broader spectral ranges or stronger coupling would require more careful treatment of level-gap renormalization.
Error found: The claimed numerical convergence is insufficiently documented because the supplied material gives no squeezed-Fock or emission-photon cutoff convergence tests. The abstract's claimed coherent-versus-squeezed differences in second- and third-order correlations are not demonstrated in the supplied textual results and captions. The squeezed pump is single-mode, so the calculation does not establish the broader applicability to broadband squeezed drivers suggested by the framing. The printed TLS Hamiltonian has gap 2Δ, inconsistent with interpreting Δ as the gap and Δ/ω_L = 1 as one-photon resonance. The supplementary excited-state population formula uses the wrong Pauli operator and omits the projector's factor of 1/2, making the printed benchmark definition inconsistent with its stated initial condition and density matrix.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.