Nonlinear Optics: High-efficiency nanophotonic frequency conversion could enable room-temperature quantum computing

A chip-integrated lithium niobate microring resonator created by researchers at Stevens Institute of Technology (Hoboken, NJ) in a thin-film lithium niobate on insulator (LNOI) medium shows high-efficiency frequency conversion, which could lead to classical applications such as on-chip parametric amplification, and quantum applications such as room-temperature quantum computing chips.¹

The researchers, led by Yuping Huang, an associate professor of physics and director of the Center for Quantum Science and Engineering, are moving closer to these goals with a nanoscale chip that facilitates nonlinear photon interactions with much higher efficiency than any previous system. The new method works at very low energy levels, suggesting that it could be optimized to work at the level of individual photons—the holy grail for room-temperature quantum computing and secure quantum communication. Its efficiency is very high at 10^-6 per single photon.

“We’re pushing the boundaries of physics and optical engineering in order to bring quantum and all-optical signal processing closer to reality,” says Huang.

The researchers designed and fabricated a racetrack-shaped microcavity in LNOI, with one 300 μm straight side of the waveguide racetrack periodically poled (by applying high-voltage electrical pulses) to achieve quasi-phase-matching between fundamental cavity modes, with the quasi-TE mode profiles shown in modeling to have a greater than 90% overlap for the fundamental and second-harmonic modes at 1545.6 nm and 772.8 nm, respectively. To get the resonances of the fundamental and second-harmonic modes in sync, the researchers placed the chip on a variable-temperature surface, allowing tuning via the different thermo-optic responses of the two modes.

Normalized conversion efficiency of 230,000%/W

Unlike silicon, lithium niobate is difficult to chemically etch with common reactive gases, so Huang’s group used an ion-milling tool to etch the racetrack, whose waveguide has a top width of 1.8 μm and a loaded quality factor (Q_L) of about 3.7 × 10⁵. At a pump optical power of only 5.6 μW, the absolute efficiency of the device is 1.3%, which equates to a normalized conversion efficiency of 230,000%/W (see figure). A comparison by the researchers of their device with six other nanophotonic resonators designed for nonlinear frequency conversion shows that the Stevens device has an order of magnitude or more conversion efficiency (the researchers scaled all conversion efficiencies to a 35 μW pump power to enable a direct comparison).

Chen explains that to both etch the racetrack on the chip and optimize the way photons move around it requires dozens of delicate nanofabrication steps, each requiring nanometer precision. “To the best of our knowledge, we’re among the first groups to master all of these nanofabrication steps to build this system; that’s the reason we could get this result first,” he says.

Moving forward, Huang and his team aim to boost the crystal racetrack’s Q_L, which in essence is its ability to confine and recirculate light. The team has already identified ways to increase Q_L by a factor of at least 10, but each level up makes the system more sensitive to imperceptible temperature fluctuations of a few thousands of a degree and requires careful fine-tuning.

Still, the Stevens team say they’re closing in on a system capable of generating interactions at the single-photon level reliably, a breakthrough that would allow the creation of many powerful quantum-computing components such as photonic logic gates and entanglement sources, which in a quantum a circuit can canvass multiple solutions to the same problem simultaneously, conceivably allowing calculations that could take years via classical computers to be solved in seconds. “It’s the holy grail,” said Chen, another member of the Stevens research team. “And on the way to the holy grail, we’re realizing a lot of physics that nobody’s done before.”

REFERENCE

1. J.-Y. Chen et al., Optica (2019); https://doi.org/10.1364/optica.6.001244.