Temporal dynamics of GHz acoustic waves in chipscale phononic integrated circuits

A. Fahad Malik 2, Mahmut Bicer 2, Krishna C. Balram · 2025

Read Temporal dynamics of GHz acoustic waves in chipscale phononic integrated circuits by A. Fahad Malik 2, Mahmut Bicer 2, Krishna C. Balram (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2504-06959.txt.

Introduction

Fahad Malik, Mahmut Bicer, and Krishna Balram watch sound waves race through tiny chips, in work from 2025. Their chips guide sound at gigahertz rates through tracks a hair wide. Such phononic circuits could route microwave signals and sense small shifts, all in a very small space.

Sound in these chips is slow and lasts long, so lab tools can film its motion in time. The team shows a pulse that loops a ring and fades, and stray paths that add noise in a bent track. Their time-based method maps where waves bounce and leak, a check that grows vital as chips gain parts. No fast 3D sound camera exists yet, so time traces are the best guide chip makers have.

Catalog record

CREATORA. Fahad Malik 2, Mahmut Bicer 2, Krishna C. Balram
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE43 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2504-06959

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Temporal dynamics of GHz acoustic waves in chipscale phononic integrated circuits A. Fahad Malik 2, Mahmut Bicer 2, Krishna C. Balram Abstract Phononic integrated circuits, which manipulate GHz \text{\,}\mathrm{GHz} -frequency acoustic fields in µ ​ m \text{\,}\mathrm{\SIUnitSymbolMicro m} -scale waveguides, provide new degrees of freedom for routing and manipulation of microwaves in deeply sub-wavelength geometries with associated implications for chipscale sensing and signal processing. The combination of low propagation loss, long interaction lengths and slow speed of sound put together with the large measurement bandwidths and high frequency resolution available from modern vector network analyzers (VNA) makes it feasible to visualize the temporal dynamics of propagating acoustic fields in these devices and see the device in action . Two representative examples we discuss here are pulse circulation and ringdown in an acoustic microring resonator, and the observation of (parasitic) multipath interference effects in waveguide resonator geometries. In the absence of fast 3D acoustic field imaging modalities, such time domain reflectometry based methods provide the best alternative for mapping interface reflection and loss, which becomes increasingly critical as these devices start to scale in complexity. I Introduction Piezoelectric phononic integrated circuits (PnICs) [ 1 ] , platforms that can route \text{\,}\mathrm{G}\mathrm{H}\mathrm{z} -frequency acoustic fields in wavelength ( \text{\,}\mathrm{\SIUnitSymbolMicro m} -scale) waveguide geometries, are currently being actively explored across a variety of material platforms [ 2 , 3 , 4 ] for applications in sensing and signal processing. Like their well-established integrated photonics counterparts [ 5 ] , PnICs provide low-loss [ 6 ] waveguiding ( \approx 3.6\text{\,}\mathrm{dB}\text{\,}{\mathrm{mm}}^{-1} at 3.4\text{\,}\mathrm{G}\mathrm{H}\mathrm{z} ) with strong confinement and long interaction lengths in chipscale platforms. Given that the acoustic field is generated from a microwave signal in piezoelectric devices, PnIC platforms provide new degrees of freedom for manipulating microwaves in chipscale platforms at the deeply subwavelength-scale (acoustic wavelength ( \lambda_{a} ) \approx \text{\,}\mathrm{\SIUnitSymbolMicro m} , RF wavelength \approx \text{\,}\mathrm{cm} ). An illustrative example from our previous work [ 6 ] is the demonstration of compact spiral delay lines that achieve on-chip signal delays of \approx 2.5\text{\,}\mathrm{\SIUnitSymbolMicro s} , corresponding to a free space delay of \approx 750\text{\,}\mathrm{m} , while maintaining an on-chip footprint < 0.25\text{\,}{\mathrm{mm}}^{2} . A large part of the promise of such chipscale acoustics platform, like with their microelectronic and photonic counterparts, derives from the prospect of cascading multiple elements together and getting the field (either electrons, light, or microwaves via sound) to flow through them in a controlled sequence to realize complex functionality. A simple example would be to go beyond the single acoustic microring resonator [ 6 ] towards coupled ring-resonator [ 7 ] geometries with a view towards improving spectral filtering. But cascading elements brings with it the challenges of designing (multiple) interface transitions to minimize reflections and avoid mode conversion, as these parasitic reflections in a low-loss waveguide platform can lead to a variety of multipath interference effects, as we demonstrate below. These stray fields can cause spectral distortion, with the classic example being triplet transits in traditional surface acoustic wave (SAW) devices [ 8 ] . In the absence of (fast) 3D field imaging modalities that can map acoustic fields with nm-scale depth resolution [ 9 ] , time domain reflectometry (TDR) methods [ 10 ] provide the best alternative for mapping propagating acoustic fields in guided wave geometries. While TDR methods have long been used in the microwave domain [ 11 ] , they are particularly suited to studying acoustic fields in PnICs. The combination of low propagation loss, long interaction lengths and slow speed of sound combined with the large measurement bandwidths and frequency resolution of modern VNAs makes it feasible to visualize the temporal dynamics of propagating \text{\,}\mathrm{G}\mathrm{H}\mathrm{z} -frequency acoustic fields in these devices in exquisite detail, as we discuss below in Section III . In a way, the PnIC platform provides a nice playground for acoustic dynamics, especially multipath interference effects, and the VNA has sufficient sensitivity for us to resolve and identify these events. This sensitivity is the main advantage of using the time domain transform of VNA spectral data over doing the experiments in the time domain [ 6 ] , as insertion loss [ 1 ] is currently the main bottleneck in these chipscale devices and limits the achievable signal-to-noise ratio severely in practice. More broadly, visualizing the temporal dynamics allows us to see the device in action . Although PnICs have been developed in analogy with their photonic counterparts, the slower speed of sound coupled with direct access to the phase information in a VNA makes studying temporal dynamics in this platform much more accessible and provides additional insights that cannot be directly inferred from the spectral domain. Here, we use the side-coupled waveguide ring resonator geometry [ 4 , 6 ] as a model device to illustrate these ideas. II Acoustic Microring resonators Fig. 1: (a) Microscope image of a representative acoustic microring resonator device. Inset shows an FEM simulation of the z -displacement of the propagating waveguide mode. Focusing interdigitated transducers (IDTs) are used to convert microwave fields at \approx 3.4\text{\,}\mathrm{G}\mathrm{H}\mathrm{z} into propagating acoustic fields (launch port, L) which can be efficiently routed using on-chip waveguides. The acoustic field in the waveguide can couple into and circulate in an acoustic microring resonator. The power leaking out from the resonator can be measured in both the through (T) and drop (D) ports. (b) The transmission spectrum of the device in the T (red) and D (blue) ports. (c) Time domain transform of the data in (b) for the T (red) and D (blue) ports. The time-domain T port data is shifted by 20\text{\,}\mathrm{dB} for illustration clarity. Fig. 1 (a) shows an optical microscope image of a representative acoustic microring resonator device fabricated in a gallium nitride (GaN) on silicon carbide (SiC) platform [ 6 ] , with radius ( R ) of 115\text{\,}\mathrm{\SIUnitSymbolMicro m} side-coupled to acoustic waveguides with width ( w_{wg} ) of 1.8\text{\,}\mathrm{\SIUnitSymbolMicro m} and thickness ( t_{wg}) of 1.5\text{\,}\mathrm{\SIUnitSymbolMicro m} . The waveguide resonator gap ( g ) is \approx 600\text{\,}\mathrm{nm} . Focusing interdigitated transducers (FIDTs) are used to convert microwave fields at \approx 3.4\text{\,}\mathrm{G}\mathrm{H}\mathrm{z} into propagating acoustic fields that are focused at the waveguide entrance for efficient mode matching. A finite element method (FEM) simulation of the waveguide mode is shown in the inset of Fig. 1 (a). At the output port of the waveguides, the process is reversed and the FIDTs are used to reconvert the acoustic field exiting the waveguide into the microwave domain. The microwave signal is generated and detected in a standard swept-frequency scattering matrix (S-parameter) measurement using a vector network analyzer (Keysight E5063A). The resonator transmission spectrum ( S_{21} ) measured in the through (T, red) and drop (D, blue) ports are shown in Fig. 1 (b). This device has a coupling length ( l_{cp} ) of 6\text{\,}\lambda_{\mathrm{a}} where the acoustic wavelength ( \lambda_{a} ), which determines the IDT period, is 1.6\text{\,}\mathrm{\SIUnitSymbolMicro m} . The waveguide resonator gap in this case is 500\text{\,}\mathrm{nm} . Both transmission spectra show a periodic series of resonance peaks in D (dips in the T), that correspond to successive resonant modes of the ring resonator. The resonance condition can be stated approximately as m\lambda_{R}=2{\pi}R where m

Read the complete text