Spectro-temporally tailored Non-Gaussian Quantum Operations in Thin-Film Waveguides
Peter Namdar, Patrick Folge, Carlos E. Lopetegui, Silia Babel, Benjamin Brecht, Christine Silberhorn, Valentina Parigi
TL;DR
The paper addresses the challenge of implementing spectro-temporal mode selective non-Gaussian operations at telecom wavelengths, focusing on mode selective single-photon addition and subtraction. It develops an inverse-design optimization framework that links waveguide geometry and pump spectra to the Joint Spectral Amplitude $J(\omega_s,\omega_i)$ and Transfer Function $T(\omega_s,\omega_{up})$, applying it to both metallic waveguide approximations and thin-film lithium niobate platforms. The results demonstrate high mode purity with Schmidt numbers near $K \approx 1$ and substantial heralded non-Gaussianity for both SPA and SPS, with purities $\mu^-$ and $\mu^+$ consistently exceeding 0.9 in many configurations; thin-film LN dispersion engineering further enhances performance. This work provides a path toward scalable, spectro-temporally multiplexed quantum photonic networks operating in the telecom band, leveraging heralded non-Gaussian operations in integrated platforms.
Abstract
Advancements in photonic platforms have enabled the precise control of light's spectral and temporal degrees of freedom, a capability crucial for the development of scalable quantum information systems. In this work, we address the challenge of implementing spectro-temporal mode-selective non-Gaussian quantum operations, specifically single-photon subtraction (SPS) and addition (SPA), in the telecom wavelength regime. Building on prior experimental demonstrations of mode-selective near-infrared SPS, we present the first design framework for achieving mode-selective SPA and SPS using thin-film lithium niobate nonlinear waveguide platforms. We introduce an inverse-design optimization scheme by modeling the quantum-optical response via the Joint Spectral Amplitude and Transfer Function, in order to identify optimal waveguide and pump parameters that maximize mode selectivity and state purity. This approach is first tested on a metallic waveguide design. We then exploit the dispersion engineering capabilities of thin-film waveguides, which offer enhanced nonlinear interactions through tighter light confinement. Our findings demonstrate that tailored nonlinear processes, particularly parametric down-conversion and frequency up-conversion, can support high-fidelity non-Gaussian operations essential for next-generation quantum photonic networks.
