I. Abstract of the Paper’s Core Research

• Ultra-precise printing, with a resonant frequency that closely matches the theoretical value.

Stable fabrication of 9–10 μm metal interconnects with corresponding electrode periods of 36–128 μm yields device‑measured center frequencies that deviate from theoretical predictions by no more than 2%, thereby addressing the longstanding issue of severe frequency drift in aerosol‑based SAW devices. Moreover, the highest third harmonic at 123 MHz represents the highest operating frequency achieved to date in aerosol‑printed SAW devices.

• Differential RF Characteristics of Single- and Dual-IDT Structures

A 2‑pole single‑electrode configuration exhibits high reflection intensity for short‑load conditions, while a 4‑pole dual‑electrode design offers superior harmonic utilization. By precisely tuning the number of electrode periods, the bandwidth can be controlled, validating the theoretical relationship NBW = 2fc/Np and enabling the on-demand design of narrow‑ or broadband RF SAW filters.

• Harmonic multiplexing enables high-precision temperature sensing.

Fundamental frequency and third-harmonic signals are simultaneously acquired to compensate for temperature drift; the measured TCF of YZ lithium niobate is −95.9 ppm/°C, in excellent agreement with the material’s standard value, while dual‑channel high‑ and low‑frequency temperature measurement significantly enhances detection robustness.

• A Novel Time-Domain Reflectometry-Based Impedance Measurement Scheme

The dual‑port SAW sensor distinguishes changes in load resistance via time‑domain reflections measured at port S11, while compensating for temperature variations and RF‑induced disturbances. It can accurately detect short circuits up to kilo‑ohm‑level loads, enabling passive impedance sensing.

• Fully digital, maskless, rapid iteration

No lithography or vacuum deposition is required; CAD‑based direct modification of interdigital transducer (IDT) line widths and geometries enables simultaneous fabrication of multiple SAW device specifications in a single batch, significantly shortening the RF sensor R&D cycle.

II. Complete SAW Device Fabrication Process and Core Advantages for Aerosol Jet Printing

(1) Complete SAW Printing Process

• YZ‑LiNbO₃ piezoelectric wafer cutting and cleaning, ensuring a pristine substrate free of impurities;
• Clariant silver nanoparticle ink is atomized by ultrasonic atomization, with a carrier gas flow of 21 sccm and a sheath gas flow of 29 sccm, using a 100 μm nozzle at a printing speed of 3 mm/s.
• Precise direct writing of 9–10 μm interdigital electrodes, with two IDT array configurations—2p and 4p—being fabricated.
• Low-temperature baking removes the solvent and enhances the electrical continuity of the silver film.
• RF probe stations and temperature-controlled platforms are used to test S-parameters, temperature characteristics, and impedance responses.

(II) Five Core Process Advantages (Comparison with Lithography/Electro‑Jet Printing)

• Approaching the 10 μm limit in feature size, low-cost fabrication of high-frequency SAW devices.

Conventional inkjet printing achieves a minimum feature size of 20 μm, while electrostatic inkjet equipment is costly. Aerosol‑based techniques optimize airflow to suppress material overflow, enabling stable fabrication of micron‑scale fine interdigitated electrodes and supporting RF devices operating at the hundreds‑of‑megahertz range.

• Non-contact conformal printing, compatible with piezoelectric irregular substrates.

3–5 mm safe spray distance; curved surfaces and small piezoelectric chips require no planarization, and are compatible with miniature passive sensing chips.

• A single- or dual-electrode IDT structure compatible with one set of equipment.

By simply switching the printed layout, two interdigitated‑finger configurations can be fabricated, each tailored to either filtering or multi‑parameter sensing applications, resulting in high fab‑line utilization.

• Maskless digital manufacturing, zero retooling costs.

Eliminates the need for photomask fabrication and wet etching, enables same-day prototyping of design modifications, and is ideal for universities and enterprises seeking rapid validation of RF sensing solutions.

• Integrated fabrication of multi-physical‑quantity devices

A single printing process can fabricate a multifunctional SAW chip capable of simultaneous temperature and impedance measurement, eliminating the need for multi‑layer thin‑film lamination.

III. Five Core Technological Innovations of This Study

Innovation 1: A stable forming process achieving aerosol‑level resolution

Innovative Aerosol SAW Harmonic Multiplexing Temperature Measurement Technology

Innovative 3-Dimensional Time-Domain S11 Impedance Sensing Architecture

Innovative Four-Band Digital Precision Control Method

Innovative Low-Cost Printed SAW Standardized Design Paradigm

IV. Existing Process Challenges and Corresponding Solutions for Overcoming Them

Challenge 1: Ultra-fine lines are prone to overspray and breakage, resulting in significant frequency offset errors.

Challenge 2: Conventional aerosol SAW devices can only operate at low frequencies, making it difficult to realize devices in the hundreds of megahertz range.

Challenge 3: Single-frequency temperature measurement is susceptible to interference from RF signal fluctuations.

Challenge 4: Conventional printed SAW devices offer only a single function—temperature sensing.

Challenge 5: Photolithography production lines require substantial capital investment, making small-batch R&D economically unviable.

V. Future Diversified Industrial Application Scenarios

• Industrial High-Temperature Passive Temperature-Sensing Tag

• Miniature RF Front-End Filter Components

• Aeroengine Structural Health Monitoring

• Food / Cold-Chain Passive Sensor Tags

• University RF MEMS Research Platform

Figure 1. (a) Conceptual schematic of a dual-port surface acoustic wave delay-line device; (b) “single-ended” or “2p” IDT structure; (c) “dual-ended” or “4p” IDT structure.

Figure 2. Based on the device dimensions and substrate materials reported in the literature, the relationship between the frequency characteristic (fc) of AJ‑printed SAW devices—experimentally measured at operating frequencies exceeding 10 MHz—and the νSAW/3IDT ratio is compared.

Table 1. Structural Parameters of Selected AJ-Printed SAW Devices

Figure 3. Microphotographs of SAW devices A–F, with annotations. The inset on the right shows a high-magnification image of the IDT structure at the second port of these devices, with the trace width and pitch labeled.

Figure 4 shows the S21 response curve of the printed SAW dual‑port structure presented in Table 1 and Figure 3. As predicted by Equation (1), the local maximum of S21 occurs near the center frequency fc, as illustrated in Figure 2. S21 represents the fraction of the RF power applied to the IDT at Port 1 that is ultimately transmitted to Port 2.

Figure 5 shows examples of zero‑bandwidth (NBW) control for two SAW filters (A and B in Table 1). Apart from the number of IDT finger pairs, these two devices have identical structures; their bandwidth values are proportional to Np and closely match the approximate formula NBW = 2fc/Np.

Figure 6. Temperature response curve of Device D. The left panel shows an example of broadband S21 measurements conducted at a single temperature, with f0 and f3 regions labeled; the right panel illustrates how the peak frequencies at f0 and f3, as well as the thermometer (TC) temperature readings, vary with the number of measurements throughout the entire temperature‑response test.

Figure 7. For the measurement data shown in Figure 6, a linear regression was performed to correlate the temperature values recorded by the thermometer (TC) with those from device Df3. The observed TCF value closely matches the accepted TCF value for this substrate material, with a discrepancy of approximately 2%.

Figure 8 shows the S11 magnitude response curves of devices E and F in the time domain (displayed in the upper and lower portions of the figure, respectively), corresponding to different values of the load resistance at port 2. A counter‑intuitive variation in the reflection coefficient at port 2 is observed when the load resistance is approximately 3.5 µs; this phenomenon is dictated by the designed IDT structure.

Figure 9. S22 and S12 scattering parameters of devices E and F: Both devices exhibit S22 and S12 resonance peaks near 85 MHz; however, device F also shows multiple S22 resonance peaks down to approximately 28 MHz, which corresponds precisely to the fundamental resonant frequency predicted by Equation (1). As expected, only a negligible S12 resonance peak is observed around 28 MHz for device F, since port 1 of this device operates at the fundamental harmonic frequency fc ≈ 85 MHz, while port 2, leveraging a 4p IDT structure, achieves efficient operation at the third harmonic of 85 MHz.