A 16GHz epitaxial aluminium nitride thin-film bulk acoustic wave resonator has been achieved by Hemant Gulupalli and colleagues at the Rensselaer Polytechnic Institute in collaboration with Yale University and Cornell University. A physics-based model predicts quality-factor limits at cryogenic temperatures. Small-signal RF measurements, conducted from 6.5 K to 300 K, reveal a consistent decrease in loaded quality factor with rising temperature, ranging from approximately 1589 at 6.5 K to 363 at 294K. The validated model offers a transferable set of tools for understanding and optimising low-loss resonators, ultimately enabling the development of cryogenic microwave filter elements for superconducting quantum hardware.
High-frequency resonator performance linked to temperature-dependent loss modelling
A peak loaded quality factor of 1589, equating to 24.79THz, was achieved in a 16GHz aluminum nitride thin-film bulk acoustic wave resonator at 6.5K. Previously, maintaining such high performance at cryogenic temperatures necessitated complex hybrid designs or materials with limited integration potential. The AlN resonator, epitaxially fabricated on silicon carbide, exhibited a consistent decline in quality factor with increasing temperature, falling to 363 at room temperature. This behaviour aligns with predictions from a new physics-based model incorporating both intrinsic and extrinsic loss mechanisms. The significance of this result lies in the potential to create highly efficient and stable microwave components for emerging technologies.
Thin-film bulk acoustic resonators (FBARs) operate by converting electrical energy into mechanical vibrations, and vice versa. Their performance is critically dependent on the quality factor (Q), which represents the energy stored in the resonator compared to the energy dissipated per cycle. A high Q factor is essential for minimising signal loss and achieving precise frequency control. Aluminium nitride (AlN) is a particularly promising material for FBAR fabrication due to its high acoustic velocity, relatively low dielectric loss, and compatibility with standard microfabrication techniques. Epitaxial growth on silicon carbide (SiC) further enhances the material quality and reduces defect density, contributing to improved performance. The observed decrease in Q factor with increasing temperature is attributed to increased acoustic dissipation, arising from various loss mechanisms within the material and device structure.
The model developed surpasses computationally intensive finite-element simulations, providing a practical framework for optimising low-loss resonators intended for demanding applications such as superconducting quantum hardware and advanced 6G communications systems. The AlN FBAR maintains a comparatively high quality factor when compared with other acoustic devices, including contour-mode resonators and surface acoustic wave devices, despite this predictable decline. This performance is linked to the resonator’s geometry and the metal-insulator-metal stack used in its construction. Specifically, the choice of materials and layer thicknesses within the stack influences the acoustic impedance and reduces unwanted parasitic modes, thereby improving the Q factor. The model accounts for these geometric and material parameters, allowing for precise prediction and optimisation of resonator performance.
Validating their new physics-based model, the team tested a 23GHz high-overtone bulk acoustic resonator, or HBAR, using previously published data, confirming its broad applicability beyond the initial 16GHz device. Furthermore, they demonstrated the potential for monolithic integration with gallium nitride (GaN) electronics, enhancing the platform’s appeal for future applications. Monolithic integration offers significant advantages in terms of size, cost, and performance, as it eliminates the need for discrete component assembly and reduces parasitic effects. GaN is a wide-bandgap semiconductor with excellent high-frequency and high-power capabilities, making it an ideal companion for AlN FBARs in advanced communication systems. Current measurements do not yet demonstrate long-term durability or address the challenges of achieving consistently high yields in mass production, representing areas for future work.
Predicting cryogenic energy dissipation in 23GHz aluminium nitride film bulk acoustic resonators
Building ever more sensitive quantum devices and high-frequency communication systems requires components that minimise energy loss at extremely low temperatures. This work provides a validated framework for interpreting performance limits and offers a detailed understanding of how film bulk acoustic resonators, or FBARs, behave when cooled. Fabricating a 16GHz aluminium nitride resonator and carefully measuring its behaviour as temperature decreased allowed scientists to establish a physics-based model to predict energy loss. The model incorporates both inherent material properties and external factors like device support, moving beyond reliance on complex computer simulations. This approach offers a practical means of optimising resonator design and potentially extending to other materials and frequencies.
The physics-based model developed in this study distinguishes between intrinsic and extrinsic loss mechanisms. Intrinsic losses originate from within the AlN material itself, including thermoelastic damping (TED) and dielectric loss. TED arises from the conversion of acoustic energy into heat due to the material’s viscoelastic properties, while dielectric loss is associated with the energy dissipation within the material’s dielectric structure. Extrinsic losses, on the other hand, are caused by factors external to the AlN film, such as clamping losses due to the device support structure and acoustic radiation losses. By accurately modelling these different loss mechanisms, the team was able to predict the temperature dependence of the Q factor with high fidelity. The model’s parameters were extracted from the experimental data obtained from 6.5 K to 300 K, ensuring its accuracy and reliability.
The implications of this research extend beyond fundamental understanding of acoustic dissipation. The ability to accurately predict and mitigate energy loss in FBARs is crucial for developing high-performance filters and resonators for a wide range of applications. In the context of 6G communications, these devices can enable higher data rates and improved spectral efficiency. For superconducting quantum hardware, low-loss resonators are essential for maintaining qubit coherence and improving the fidelity of quantum computations. The model’s versatility allows for adaptation to different resonator geometries, materials, and operating frequencies, making it a valuable tool for researchers and engineers working in these fields. Further research will focus on improving the model’s accuracy by incorporating additional loss mechanisms and exploring novel materials with even lower acoustic dissipation.
The research demonstrated a 16GHz aluminium nitride resonator achieved a maximum loaded quality factor of 1589 at 6.5 K, decreasing to 363 at 294 K. This understanding of temperature-dependent acoustic dissipation is important because it limits the performance of devices used in technologies like 6G communications and quantum computing. Researchers developed a physics-based model to quantify these losses, distinguishing between intrinsic material properties and extrinsic factors like device support. The validated model provides a framework for optimising resonator design and interpreting quality factor limits across various materials and frequencies.
👉 More information
🗞 Cryogenic Loss Limits in Microwave Epitaxial AlN Acoustic Resonators
🧠ArXiv: https://arxiv.org/abs/2604.13364
