Compact Turnkey Soliton Microcombs at Microwave Rates via Wafer-Scale Fabrication

Yuanlei Wang, Ze Wang, Chenghao Lao, Tianyu Xu, Yinke Cheng, Zhenyu Xie, Junqi Wang, Haoyang Luo, Xin Zhou, Bo Ni, Kaixuan Zhu, Yanwu Liu, Xing Jin, Min Wang, Jianfei Liu, Xuening Cao, Ting Wang, Qihuang Gong, Bei‐Bei Li, Fangxing Zhang, Yunfeng Xiao, Qi‐Fan YangarXiv 2025 · DOI · arXiv

This research paper presents a significant leap in miniaturizing optical frequency combs, specifically "soliton microcombs," which are essential for bridging the gap between ultra-fast optical signals and the microwave electronics we use in everyday technology. By rethinking resonator geometry and utilizing wafer-scale fabrication, the team from Peking University has created a compact, "turnkey" system that produces microwave-rate signals with high stability and low noise.

TL;DR

Researchers have developed ultra-compact silicon nitride ($Si_3N_4$) microresonators that generate soliton microcombs at microwave repetition rates ($<10$ GHz) within a tiny $1 \times 1$ mm footprint. By using "finger" and "spiral" geometries and a turnkey self-injection locking mechanism, they’ve moved these complex systems from the lab bench into a portable module the size of a computer mouse.

What Problem Does This Paper Address?

Optical frequency combs (OFCs) act like "optical rulers," providing a series of perfectly spaced spectral lines. To be useful for microwave photonics (like radar or high-speed communications), these lines need to be spaced closely together—typically at frequencies below 10 GHz.

However, there is a physical trade-off: to get a low repetition rate (close spacing), you need a very long optical path (centimeters). In a standard "ring" resonator, a centimeter-long path results in a massive footprint (roughly $5 \times 5$ mm), which wastes valuable chip real estate and consumes too much power for integrated lasers to handle. This has historically made it difficult to integrate low-rate combs into dense photonic circuits.

Methods

The authors tackled the footprint and power issues through three main innovations:

  1. Compact Geometries: Instead of simple rings, they engineered finger-shaped and spiral-shaped resonators. These designs use "Euler bends"—specialized curves that minimize optical loss—to pack long waveguide lengths into a compact $1 \times 1$ mm area.
  2. Wafer-Scale Fabrication: They used a 4-inch wafer process with 777-nm-thick silicon nitride. The process included high-temperature annealing (1200°C) to reduce internal stress and minimize optical absorption, achieving intrinsic quality factors ($Q_0$) exceeding $10^7$ (meaning light can circulate tens of millions of times before dying out).
  3. Turnkey Operation via Self-Injection Locking: Generating a "soliton" (a stable, pulsing wave of light) usually requires complex, high-speed electronic tuning. Here, they directly coupled a standard distributed-feedback (DFB) laser to the chip. The backscattering of light from the resonator naturally "locks" the laser to the correct frequency, allowing the comb to start up automatically (turnkey) just by turning on the current.

Datasets

As an experimental physics paper, the "datasets" consist of performance metrics from the fabricated chips:

  • Device Variety: The study characterized three resonator types: 100 GHz (ring), 25 GHz (finger), and 10 GHz (spiral).
  • Yield: The wafer-scale process produced up to 48 photonic chips per 4-inch wafer.
  • Characterization: Data includes transmission spectra, integrated dispersion measurements, and phase noise plots.

Key Insights and Ideas

  • Footprint Scaling: While a ring resonator's footprint scales quadratically with the round-trip length ($L^2$), finger and spiral designs scale more linearly ($L$), allowing for much higher integration density.
  • Loss Management: The researchers discovered that for repetition rates above 25 GHz, finger shapes are superior, while spirals are better for rates below 25 GHz. This provides a design "rule of thumb" for future photonic integration.
  • The "Turnkey" Advantage: By eliminating the need for complex startup sequences or bulky optical isolators, the system becomes robust enough for use outside of a controlled laboratory environment.
  • Absorption vs. Scattering: Through detailed loss analysis, they found that while scattering (from sidewall roughness) is the main loss at narrow waveguide widths, material absorption (likely from hydrogen bonds introduced during fabrication) becomes a significant factor in the C-band.

Key Results

  • Repetition Rates: Successfully demonstrated single-soliton combs at 100 GHz, 25 GHz, and <10 GHz.
  • Phase Noise: The 10 GHz carrier achieved a phase noise of -130 dBc/Hz at a 100 kHz offset, which is among the lowest reported for fully integrated microcombs.
  • Stability: The system showed excellent long-term reliability, with power fluctuations under 5% over 5,000 seconds and a frequency drift of less than 2 kHz per hour.
  • Form Factor: The entire system—including the laser, chip, and control electronics—was packaged into a 10 cm x 7 cm x 4 cm module, comparable to the size of a computer mouse.

Limitations and Future Work

  • Absorption Losses: The authors noted that their absorption loss is higher than some state-of-the-art results, which they attribute to ultraviolet exposure during the PECVD process. Future iterations could use chemical-mechanical polishing (CMP) or additional annealing to further boost the Q-factor.
  • Higher Integration: While the current module is compact, the laser and chip are still separate components within the package. The next step is "monolithic" or "heterogeneous" integration, where the laser and resonator are fabricated on the same substrate.
  • Power Efficiency: While the turnkey system is efficient, further reducing the power threshold would allow for even smaller battery-operated devices.

Summary updated

Abstract

Soliton microcombs generated in nonlinear microresonators facilitate the photonic integration of timing, frequency synthesis, and astronomical calibration functionalities. For these applications, low-repetition-rate soliton microcombs are essential as they establish a coherent link between optical and microwave signals. However, the required pump power typically scales with the inverse of the repetition rate, and the device footprint scales with the inverse of square of the repetition rate, rendering low-repetition-rate soliton microcombs challenging to integrate within photonic circuits. This study designs and fabricates silicon nitride microresonators on 4-inch wafers with highly compact form factors. The resonator geometries are engineered from ring to finger and spiral shapes to enhance integration density while attaining quality factors over 10^7. Driven directly by an integrated laser, soliton microcombs with repetition rates below 10 GHz are demonstrated via turnkey initiation. The phase noise performance of the synthesized microwave signals reaches -130 dBc/Hz at 100 kHz offset frequency for 10 GHz carrier frequencies. This work enables the high-density integration of soliton microcombs for chip-based microwave photonics and spectroscopy applications.
BibTeX
@misc{wang2025compact,
  author = {Yuanlei Wang and Ze Wang and Chenghao Lao and Tianyu Xu and Yinke Cheng and Zhenyu Xie and Junqi Wang and Haoyang Luo and Xin Zhou and Bo Ni and Kaixuan Zhu and Yanwu Liu and Xing Jin and Min Wang and Jianfei Liu and Xuening Cao and Ting Wang and Qihuang Gong and Bei‐Bei Li and Fangxing Zhang and Yunfeng Xiao and Qi‐Fan Yang},
  title = {{Compact Turnkey Soliton Microcombs at Microwave Rates via Wafer-Scale Fabrication}},
  year = {2025},
  eprint = {2502.10941},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2502.10941},
  howpublished = {arXiv},
}
Figure