Silica glass, a high-performance material, is widely used in various applications, from lenses and glassware to optical fibers. Its exceptional properties, including thermal and chemical stability, hardness, and optical transparency across a broad wavelength range, make it highly desirable. However, fabricating three-dimensional (3D) silica glass objects with micrometer-scale features presents significant challenges due to the material's inherent brittleness and stability. Existing additive manufacturing techniques, such as stereolithography, direct ink writing, digital light processing, and multiphoton polymerization, often rely on sol-gel methods using silica nanoparticle-loaded composites. These methods typically require high-temperature sintering (~1200 °C) to remove organic components and achieve transparency and solid-state properties. This sintering process causes substantial structural shrinkage and severely restricts substrate material choices, limiting integration possibilities and applications. The current state-of-the-art methods achieve feature sizes at best in the tens of micrometers, except for a recent report achieving sub-micrometer resolution, yet still requiring high-temperature sintering. This research aims to address these limitations by developing a novel 3D printing method for producing solid, optically transparent silica glass structures with sub-micrometer resolution without the need for high-temperature sintering. The successful realization of such a process will significantly expand the potential applications of silica glass in micro- and nano-scale devices and systems, particularly in nanophotonics, nanoelectromechanical systems (NEMS), and nanofluidics, where precise control over three-dimensional structure and material properties is critical.

Previous attempts to 3D print silica glass have largely relied on sol-gel techniques, employing various methods like stereolithography, direct ink writing, and digital light processing. These approaches typically involve the use of organic components, acting as binders or photoinitiators, which are then removed through high-temperature sintering (around 1200 °C). This high-temperature step, while necessary to achieve the desired glass properties, presents several limitations. The significant shrinkage associated with sintering restricts the achievable precision and complexity of the structures. Furthermore, the necessity for high-temperature processing limits the choice of substrate materials, making integration with pre-existing structures challenging. Existing techniques, at best, achieve feature sizes in the tens of micrometers, with recent advancements achieving sub-micrometer resolution, albeit still requiring sintering. Hydrogen silsesquioxane (HSQ), an inorganic silica-like material, has been used as a high-resolution resist in two-dimensional patterning via electron beam, ion beam, or deep UV lithography. These techniques, based on linear absorption, have been adapted to create suspended structures by varying the crosslinking depth. However, creating free-form 3D structures with these methods remains infeasible. Recent work explored HSQ crosslinking using sub-picosecond lasers and nonlinear absorption, but the resulting structures were limited to suspended 2D beams, and the formation of true silica glass bonds was not fully demonstrated. This research builds upon these previous efforts, aiming to overcome their limitations and provide a new approach to 3D silica glass printing.

The researchers developed a three-step process for 3D printing silica glass: (1) drop-casting of HSQ dissolved in organic solvents onto a substrate, (2) tracing the desired 3D shape in the dried HSQ using a focused sub-picosecond laser beam (1040 nm wavelength, 298 fs pulse duration, 10 kHz repetition rate), and (3) dissolving the unexposed HSQ using a potassium hydroxide solution. Two different objectives were employed, one with a numerical aperture (NA) of 0.65 (for larger structures) and another with an NA of 1.4 (oil immersion, for sub-micrometer resolution). The laser power was carefully adjusted to ensure adequate crosslinking while preventing damage. The unexposed HSQ was removed using a 0.1 M potassium hydroxide solution with 0.05 vol% Triton X-100 surfactant added to minimize bubble formation. Subsequent annealing was performed in air at various temperatures (150 °C, 300 °C, 500 °C, 800 °C, 900 °C, and 1200 °C) for 1 hour each to investigate its effects on the material properties. The printed structures were characterized using a range of techniques. Scanning electron microscopy (SEM) and scanning transmission electron microscopy (STEM) were used to assess the morphology and resolution of the printed structures. Energy-dispersive X-ray spectroscopy (EDS) confirmed the elemental composition of the material. Raman spectroscopy was employed to analyze the chemical bonds and identify the presence of residual components like hydrogen, carbon, and different silicon-oxygen ring configurations. Photoluminescence spectroscopy was used to investigate the optical properties of the as-printed and annealed glass. Nanoindentation measurements determined the hardness and reduced elastic modulus. Finally, the optical performance of a 3D-printed optical microtoroid resonator was evaluated by measuring its transmission spectrum in the optical telecommunication S, C, and L bands. The shrinkage of the printed structures after annealing was measured using SEM, and a model was developed for analyzing the transmission spectrum of the resonator to extract key parameters.

The researchers successfully 3D printed silica glass structures with voxel dimensions as small as ~65 nm in width and 260 nm in height (aspect ratio of 4), demonstrating sub-micrometer resolution. Electron diffraction confirmed the amorphous nature (silica glass) of the printed material, and EDS analysis revealed minimal residual carbon (below 1%). The as-printed glass, while optically transparent, exhibited unique spectroscopic features compared to fused silica, including a higher ratio of 4-membered silicon-oxygen rings, photoluminescence, and residual hydrogenated and hydroxyl species. These features, attributed to the sub-picosecond laser exposure, were effectively removed by annealing at 900 °C, resulting in a material nearly identical to fused silica with only a 6.1% linear shrinkage. Annealing at 900 °C also improved the mechanical properties, increasing the hardness from 2.4 ± 0.2 GPa to 7.7 ± 0.6 GPa and the reduced elastic modulus from 40 ± 2 GPa to 75 ± 2 GPa, values comparable to fused silica. The 3D printing process allowed for the fabrication of complex, functional micro-optical components. A 3D-printed microtoroid resonator coupled to an integrated photonic bus waveguide demonstrated optical functionality both before and after annealing at 900 °C. The transmission spectra showed clear resonances in all cases, indicating that the annealing process did not significantly impair the resonator's performance. The ability to create suspended structures was shown by printing a suspended silica glass plate on the tip of an optical fiber, demonstrating compatibility with pre-existing structures and substrates that cannot withstand high temperatures.

This work presents a significant advancement in additive manufacturing of silica glass. The developed technique achieves sub-micrometer resolution and produces solid, optically transparent silica glass structures without the need for high-temperature sintering, addressing the limitations of existing methods. The low shrinkage observed after annealing at 900 °C is crucial for preserving the fidelity of the printed structures and enabling integration with diverse substrates. The successful fabrication and characterization of a functional microtoroid resonator highlight the technique’s potential for creating complex photonic devices. The ability to print on an optical fiber tip further underscores the technique's versatility and potential for applications requiring integration with existing systems. The findings have significant implications for various fields. In photonics, this method allows for the creation of high-precision micro-optical components with intricate designs. In biomedicine, it enables the production of complex 3D structures for drug delivery, biosensors, or microfluidic devices. In quantum optics, it opens possibilities for developing integrated quantum photonic circuits. Future research could focus on optimizing the laser parameters to further minimize the residual defects in the as-printed glass, potentially eliminating the need for annealing or reducing the annealing temperature. Investigating the incorporation of functional materials into the HSQ before printing could further enhance the properties and capabilities of the resulting 3D structures.

This research demonstrates a novel approach for 3D printing of silica glass with sub-micrometer resolution without the need for high-temperature sintering. The method utilizes nonlinear absorption of sub-picosecond laser pulses to crosslink HSQ, yielding optically transparent silica glass structures with excellent mechanical and optical properties after a low-temperature annealing step. The successful fabrication of functional micro-optical components and the demonstration of printing on an optical fiber tip highlight the technique's versatility and potential for a wide range of applications. Future work should focus on further optimizing the printing process and exploring the incorporation of other materials to expand the range of applications.

While this method offers significant advantages over existing techniques, certain limitations exist. The relatively high ratio of 4-membered silicon-oxygen rings and the presence of photoluminescence in the as-printed glass, although remedied by annealing, might limit applications requiring exceptional optical purity. Furthermore, the printing speed could be improved to enhance throughput. The current reliance on specific laser parameters and HSQ composition may also limit the generalizability of the method. Further research is required to address these points and fully explore the potential of this novel approach.