New Opportunities for Nonlinear Optics in Thin-film Lithium Niobate

1. TFLN platform and advantages

Thin-film lithium niobate (TFLN) has emerged as a versatile tool for integrated photonics. Nonlinear optical waveguides based on lithium niobate have long been studied as multifunctional tools for generating and manipulating optical fields. In principle, these devices can be used to replace many of the lasers and optical amplifiers operating between 400 to 4500 nm and can be used to determine resource states for quantum computing. However, nonlinear waveguides are power-hungry, which limits their use in the field where size, weight, power, and cost are critical considerations.

TFLN has led to improved performance in many applications since tightly confining structures greatly enhance interactions between fields and can modify the velocity of the interacting waves. Surface acoustic wave filters were the first devices to take advantage of these properties; the film thickness can be chosen to modify the velocity of the acoustic waves to achieve enhanced filter bandwidth. With the development of low-loss optical waveguides in TFLN, broadband electro-optic modulators soon followed. Wavelength-scale confinement enables modulators with half-wave voltages approaching 1 V, and engineering the velocities of the optical and radiofrequency fields has led to modulators with bandwidths beyond 100 GHz.

2. Dispersion-engineered nonlinear waveguides

Early demonstrations of nonlinear optics in TFLN focused on extending these design principles. The first nonlinear waveguides demonstrated in TFLN showed a 20-fold reduction in power requirements relative to diffuse waveguides based on reverse proton-exchange. We then used geometric dispersion engineering to control the velocities of the interacting waves, thus engineering the bandwidths of both frequency converters and optical parametric amplifiers. It was during this study that we discovered that dispersion-engineered nonlinear waveguides can exhibit qualitatively new dynamical regimes, which in turn can be used to determine new functionalities.

The freedom to engineer new dynamical regimes enables engineers to overcome the design constraints that typically limit device performance. One of the earliest examples of these performance enhancements was the discovery of a new approach to supercontinuum generation (SCG) based on second-order nonlinear interactions in TFLN waveguides. In conventional devices based on third-order nonlinearities, engineers must choose between using guided-wave devices with normal or anomalous dispersion. Anomalous dispersion enables large bandwidths to be achieved with relatively low power, but at the cost of introducing excess noise. Conversely, normal dispersion allows for low-noise operation, at the cost of less bandwidth and larger power requirements. SCG based on second-order nonlinearities in TFLN can overcome these tradeoffs. Multi-octave spectra can be achieved using two orders of magnitude less power than waveguides based on third-order nonlinearities, yet in principle these devices can still exhibit shot-noise-limited fluctuations.

3. Future applications in quantum photonics

Dispersion-engineered nonlinear waveguides operating with new dynamical regimes are now being explored as a platform to develop a new generation of components for quantum optics. TFLN devices can potentially go beyond the usual limits of nonlinear optics to reach saturated nonlinear interactions with relatively few photons. At this scale, new opportunities emerge including new approaches to quantum non-demolition measurements and deterministic sources of non-Gaussian states.