Fabrication of microfluidic cavities using Si-to-glass anodic bonding

Prior research has shown that microfluidic cavities are commonly fabricated using wafer bonding followed by dicing. This method increases the risk of structural failure and complicates post-processing. The need for alternative fabrication methods that avoid these limitations remains unmet. It was already known that anodic bonding could join silicon and glass without requiring additional structural supports. However, the role of such supports in influencing phase boundaries in confined systems like helium-3 had not been fully resolved. No prior work had resolved the impact of eliminating these supports on cavity stability and performance. This gap motivated the exploration of anodic bonding as a standalone method for cavity fabrication. The absence of a clear alternative to full wafer bonding for large cavities presented a key challenge. The potential for smoother surfaces and optical access in new designs suggested a shift in fabrication priorities. The need for robust, pressure-tolerant cavities in experimental physics remained a driving factor.

The aim of the study was to develop a reliable fabrication method for microfluidic cavities using anodic bonding without requiring structural posts. This approach addresses the limitations of traditional full wafer bonding, which is prone to failure and requires dicing after bonding. The motivation stemmed from the need to eliminate potential pinning centers that could interfere with phase boundary dynamics in confined systems. The researchers sought to define cavity height through etched silicon rather than relying on physical spacers. The study also aimed to improve surface smoothness and eliminate pinning sites to enhance experimental accuracy. The use of transparent glass as a cover was intended to allow optical access for further analysis. The inclusion of pressure-capable ports was designed to facilitate fluid access in high-pressure environments. The overall goal was to create a cavity suitable for torsional oscillator experiments requiring high-quality joints.

The fabrication process involved anodic bonding of a pre-cut Hoya SD-2 glass piece to a patterned silicon substrate. The silicon was etched to define the cavity depth and dimensions. Anodic bonding was performed at 425 °C with a voltage of 200 V to ensure a strong bond. The absence of structural posts was a key innovation, as it eliminated potential pinning centers for phase boundaries. The bonding process was optimized to avoid bond separation under high pressure. The cavities were tested by pressurizing them beyond 30 bars to assess structural integrity. The edges of the cavities were carefully prepared to ensure smooth surfaces and eliminate defects. The junction between the coin silver fill line and the silicon base was designed to minimize internal friction, supporting high-quality torsional oscillations.

The fabricated cavities reached a depth of approximately 1.08 μm with dimensions up to 7 mm × 11 mm or 11 mm in diameter. Anodic bonding at 425 °C with 200 V produced stable structures without the need for posts. Pressurization tests showed that the glass broke before the bond failed, indicating strong adhesion. The absence of structural spacers eliminated potential pinning centers for phase boundaries. The smooth surfaces achieved through careful preparation reduced pinning sites, improving experimental accuracy. The use of transparent glass allowed optical access for monitoring internal processes. The coin silver fill line formed a low internal-friction joint with the silicon base, enhancing torsional oscillator performance. The overall design supported high-quality factor (Q) measurements, which are essential for frequency resolution in superfluid inertial experiments.

The authors propose that anodic bonding without structural posts is a viable method for fabricating microfluidic cavities. The elimination of posts reduces interference with phase boundary dynamics in confined systems. The bonding process at 425 °C with 200 V ensures structural integrity under high pressure. The use of transparent glass as a cover provides optical access for experimental monitoring. The smooth surfaces achieved through surface preparation minimize pinning sites, improving measurement accuracy. The coin silver fill line forms a low internal-friction joint, which supports high-quality torsional oscillations. The design supports mass coupling experiments requiring high Q factors for frequency resolution. The authors suggest that these features may be beneficial in other applications requiring optical access and pressure tolerance.