Substrate-blind photonic integration based on high-index glass materials

Hongtao Lin, Lan Li, Yi Zou, Qingyang Du, Okechukwu Ogbuu, Charmayne Smith, Erick Koontz, David Musgraves, Kathleen Richardson, Juejun HuProceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE 2014 · DOI

TL;DR

This paper introduces a "substrate-blind" photonic integration approach using high-index non-silicate glasses. This allows for the creation of complex photonic devices on a wide variety of unconventional substrates, overcoming the limitations of traditional substrate-dependent methods and opening doors for new applications in sensing, flexible electronics, and beyond.

What Problem Does This Paper Address?

Traditional photonic integration technologies are heavily reliant on the substrate material. This means that the choice of substrate dictates the fabrication methods, material compatibility, and ultimately, the design possibilities for photonic devices. For instance, silicon-on-insulator (SOI) and silicon nitride are standard for high-index-contrast photonics, while III-V semiconductors are used for other applications. This substrate specificity creates a significant barrier to integrating photonic devices with emerging materials and platforms like plastics, textiles, metals, or even 2D materials like graphene. These unconventional substrates offer unique functionalities for applications in areas such as renewable energy, advanced imaging, and flexible electronics, but current photonic integration methods are not well-suited for them. The authors aim to break free from these substrate constraints by developing a versatile and compatible integration platform.

Methods

The core of this work lies in the use of high-index non-silicate glasses, specifically amorphous chalcogenides (ChGs) and transition metal oxides (like TiO₂ and ZrO₂). These materials are chosen for several key reasons:

  • Epitaxial-Free Deposition: They can be deposited without requiring lattice-matching to the substrate, eliminating a major fabrication hurdle.
  • Tailorable Optical Properties: Their refractive indices can be precisely tuned by composition, allowing for the creation of high index contrast needed for efficient waveguiding.
  • Low Processing Temperature: Deposition and fabrication can be performed at temperatures below 250°C, making them compatible with a wide range of substrates, including those that cannot withstand high temperatures (like plastics).
  • Compositional Flexibility: ChGs offer a vast range of compositional alloying, allowing for fine-tuning of optical properties and transparency windows.

The fabrication process involves depositing these glass films using techniques like thermal evaporation or spin coating. Device patterning is achieved through methods such as lift-off or thermal nanoimprint for waveguides, and standard UV lithography followed by plasma etching for other structures. This approach enables monolithic integration, meaning all layers are deposited and patterned sequentially on the target substrate.

Datasets

Not discussed in the paper.

Key Insights and Ideas

  • "Substrate-Blind" Integration is Achievable: The central idea is that by using materials with low-temperature processing and tunable optical properties, photonic devices can be integrated onto virtually any substrate without compromising performance. This is a significant departure from traditional methods.
  • Unlocking Unconventional Platforms: This technology opens up photonic integration for substrates previously considered incompatible, such as flexible plastics, textiles, and even atomically thin materials like graphene.
  • Enabling Complex 3D Structures: The ability to stack multiple layers of these glasses allows for the creation of complex 3D photonic structures, which are difficult or impossible to achieve with conventional methods.
  • High-Performance Mid-IR Devices: The materials and methods are well-suited for mid-infrared (mid-IR) applications, demonstrating high-quality resonators with excellent Q-factors.
  • Direct Graphene Integration: The approach allows for direct deposition of photonic structures onto graphene, bypassing the need for transferring graphene onto pre-fabricated photonic circuits, which simplifies fabrication and enables more complex designs.

Key Results

The paper showcases the versatility of their approach through three main examples:

  • Mid-IR Resonator Sensors on Fluoride Crystals: They demonstrated high-performance mid-IR micro-disk resonators on CaF₂ substrates. These resonators achieved an intrinsic quality factor (Q-factor) of 4 × 10⁵, which they claim is the highest reported for mid-IR on-chip resonators. They also showed the potential for chemical sensing, demonstrating the ability to detect ethanol with high sensitivity.
  • Flexible Photonic Devices on Plastic Substrates: They fabricated flexible photonic devices using ChGs and TiO₂ on plastic substrates. These devices exhibited high Q-factors (up to 4.6 × 10⁵ for ChG micro-disk resonators) and demonstrated remarkable mechanical robustness, sustaining up to 5,000 bending cycles with minimal optical performance degradation.
  • Monolithic Photonic Structures on Graphene: They successfully deposited ChG films directly onto CVD-grown graphene and patterned optical gratings. Raman analysis confirmed that the graphene's structure remained largely unchanged, indicating a gentle integration process.

Limitations and Future Work

  • Q-factor Reduction in Complex Structures: While high Q-factors were achieved in simpler structures, the loaded Q-factor of a horizontal slot resonator was 40,000, slightly lower than that of GAST resonators without the slot (80,000). This suggests that the complexity of multi-layer structures can impact performance.
  • Fatigue Analysis: While flexible devices showed good resilience, the minor loss increase observed after bending cycles (~0.5 dB/cm) was attributed to potential layer delamination or sub-micron cracks not visible under optical microscopy. Further investigation into the long-term mechanical stability and failure mechanisms would be beneficial.
  • Future Work: The authors anticipate extending this multi-material integration strategy to other technically important substrates, including functional oxides, metals, and ceramics. They also envision further development towards multi-functional integration, combining optical functionalities with other electronic or sensing capabilities.

Summary updated

Abstract

Conventional photonic integration technologies are inevitably substrate-dependent, as different substrate platforms stipulate vastly different device fabrication methods and processing compatibility requirements. Here we capitalize on the unique monolithic integration capacity of composition-engineered non-silicate glass materials (amorphous chalcogenides and transition metal oxides) to enable multifunctional, multi-layer photonic integration on virtually any technically important substrate platforms. We show that high-index glass film deposition and device fabrication can be performed at low temperatures (< 250 °C) without compromising their low loss characteristics, and is thus fully compatible with monolithic integration on a broad range of substrates including semiconductors, plastics, textiles, and metals. Application of the technology is highlighted through three examples: demonstration of high-performance mid-IR photonic sensors on fluoride crystals, direct fabrication of photonic structures on graphene, and 3-D photonic integration on flexible plastic substrates.
BibTeX
@inproceedings{lin2014substrate,
  author = {Hongtao Lin and Lan Li and Yi Zou and Qingyang Du and Okechukwu Ogbuu and Charmayne Smith and Erick Koontz and David Musgraves and Kathleen Richardson and Juejun Hu},
  title = {{Substrate-blind photonic integration based on high-index glass materials}},
  booktitle = {Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE},
  year = {2014},
  doi = {10.1117/12.2073972},
}
Figure