Fabrication and characterization of interdigital transducer structures as temperature sensors by two-photon lithography

1. Introduction

The application-based technological developments with customer requirements accelerated the development of sensor technology in many fields. Sensors are the main mechatronic system components that detect changes in the environment through mechanical or non-mechanical stimulations such as pressure, stiffness, roughness, and temperature [1-2]. The preferred manufacturing technology, selected materials, and the required response time are the most important factors in sensor technology. These factors are important in sensor fabrication.

Developments in industrial applications have started to require more micro and nano-scale devices. This need requires sensor systems to be produced in smaller sizes for a wide variety of industrial applications [3-5]. Micro and Nano Electro-Mechanical Systems (MEMS/NEMS) are used in transportation sectors such as automotive [6] and aviation [7], energy [8], robotics [9-10], optics [11], biomedical [12] and geophysics [13]. MEMS devices function as biosensors in medical research and contribute to many promising studies [14]. These sensors can reduce very wet laboratories to a microchips [15]. In addition to the solutions produced for the electrical [16-17], and construction [18] sectors, they also have applicability in knowledge and labor-intensive fields such as mechanics and chemistry. Although these devices are a single unit, they also have electrical and mechanical components. These devices can measure changes in physical quantities such as temperature, pressure, viscosity, stress, and mass [19-25]. Devices called laboratories on chips have become increasingly preferred in biomedical and bioengineering fields [26]. These devices with reduced geometric dimensions can react much faster and will continue to be more attractive as more efficient and portable devices [27-28]. Of course, in addition to these superior features, there are also some limitations. The most important of these is the need for simultaneous perception and execution in an environment while associating it with mutually exclusive mechanisms [29-30]. To overcome these limitations, single-platform devices capable of detecting acoustic waves have been proposed for simultaneous sensing and actuation. Such devices are known as Surface acoustic waves (SAW) sensors. These devices are widely used not only as sensors and actuators but also as filters, oscillators, and transformers. The applications here are made possible by the use of piezoelectric components as the central backbone of SAW sensors [31-34].

As is known, SAW-based sensors can be fabricated using a cleanroom facility and conventional [35] or next-generation mask lithography. These complex several-step processes [36] are difficult to follow; Graphenic sensitive layers can be used in the manufacturing of new generation surface acoustic wave (SAW) sensors developed with a graphene sensitive layer. When this type of sensor is produced, low cost, easy manufacturability and superior performances can be achieved. In fact, within the capabilities of production technology, producing nano-sized products can increase the advantages mentioned. Studies on graphenic precision layer SAW sensors are ongoing. [37]. It is also known that there are individual layers that come together spontaneously. Therefore, with the rapid development of microprocessors and the development of software tools, SAW designs with complex structures in terms of both geometry and materials can be proposed. This revealed the need for new manufacturing technologies that could be used quickly to produce the proposed new designs. In this context, direct writing techniques have emerged as a promising approach in sensor fabrication. Au et al. [38] compared the cost of a sensor produced traditionally, that is, through lithography, with that of 3D printing technology. In stereolithography, the difference was found to be 15 USD [39]. The fact that the 3D printing technique is both low-cost and simple to produce has made it very attractive to reproduce the same structure with minimum human effort. Additionally, the sensitivity and accuracy of the printed sensor are affected compared to traditional methods [40-41]. Among the direct writing methods, Ink-jet printing provides improved flexibility in terms of both design customization and substrate size [42]. However, the need for miniaturized devices limits the use of this method as the line resolution is in the order of a few tens of micrometers. To improve spatial resolution, direct laser writing methods (DLW) are particularly interesting as they allow the fabrication on demand of customized complex structures with submicron resolution [43, 44].

DLW using a UV beam was introduced to create micro-supercapacitor devices onto a graphite oxide film. Other DLW-based techniques were reported in the literature. They allow the fabrication of sensors by laser fusion of silver laser reduction of graphene oxide, and laser micromachining. However, these techniques are limited to some specific materials [45].

In that context, Two-Photon Lithography (TPL) in positive photoresists combined with material deposition and lift-off is an interesting alternative approach as it allows the rapid fabrication of complex designs in a large choice of materials. In addition, TPL allows an increase in spatial resolution due to the non-linear absorption process and some photochemical effects [46-49]. Fabrication using TPL [50, 51] evolved in the 90s [52] and has found its application in many different fields including but not limited to biosensing [53-55], micro and nano-optics [56-61], robotics [62, 64], security [65].

This work focuses on the use of TPL combined with physical vapor deposition of Gold for the fabrication of an Interdigital Transducer (IDT) based passive temperature sensor. As compared to the conventional lithography techniques, Two-Photon Lithography (TPL) stands out from traditional fabrication methods by eliminating the need for physical masks in the creation of IDTs for acoustic wave sensors. This mask-free approach offers greater flexibility, precision, and customization options, making TPL particularly advantageous for rapidly prototyping and tailoring IDT designs to specific sensor applications [66-69]. Moreover, combined with physical or chemical vapor deposition methods and lift-off, TPL in positive photoresists offers the advantage of fabricating IDT structures with a large number of materials (metals and metal oxides) [70-73].

2. Materials and methods

2.2. Sensor design

The dimensions of the proposed single port IDT structure, presented in Figure 1, are provided in Table 1.

Fig. 1Design and design parameters of the proposed IDT sensing structure. COMSOL was used during the designing phase

Table 1Summary of geometric parameters for the proposed IDT structures

Parameter	Size in µm
Thickness of the design	2
Electrode finger width ($W$)	3
Distance between electrodes ($P$)	6
IDT length (reflectors) ($L_2$)	80
IDT length ($L_1$)	100
Distance between IDT and reflectors ($D$)	3
Distance between two IDTs (${\lambda }_0$)	21
Aperture ($A$)	92.5
Number of reflectors ($N_r$)	15
Number of IDT pairs ($N_i$)	30

Fig. 1 describes the structural design parameters for the proposed resonator-based temperature sensor. As depicted in Fig. 1, the IDT structure is designed with a certain pitch among them and the resulting surface acoustic wave is well-founded when the pitch of the IDT fingers and the wavelength of the surface wave are equal to each other. Eq. (1) is employed to determine the resonant frequency ($f$), which is a function used for measuring the temperature, of the proposed structure and its relation to the propagation velocity ($V_r$) of the resulting surface wave:

1

$f=\frac{Vr}{\lambda },$

where $\lambda$ is the wavelength of the SAW. The IDT structure needs to be quantified to develop its model. Fig. 1 and Table 1 show the labeled IDT structure and its explanation respectively. To achieve the strongest IDT activation and phase superposition of SAW, the IDT pitch, which is $W+P$, should be equal to half of the SAW wavelength, as described by the wave interference principle. Therefore:

2

$W+P=N*\frac{\lambda }{2}.$

The distance ($D$) between the IDT and the adjacent reflectors should also satisfy equation 3 to make sure that IDT receives the standing wave on its peak:

3

$D=\left(N-\frac{1}{2}\right)*\frac{\lambda }{2}.$

The aperture width of the IDT fingers also plays an important role in the performance of the SAW resonator. Normally, it is between 50 and 100 times the SAW wavelength.

2.3. Fabrication

Nanoscribe GmbH’s Photonic Professional GT was utilized for the fabrication of the sensing structure. The fabrication setup was based on a DLW system using TPL in a positive photoresist. This system implemented a femtosecond fiber laser source which was operating at 780 nm wavelength. The power of the deployed laser source was between 50-150 mW with 80 MHz of repetition rate. The laser source produced a pulse length of 100-200 femtoseconds.

The fabrication process of IDT structures made from gold on a glass substrate is shown in Fig. 2. The dimension of the glass substrate used for this fabrication process was size 22 mm×22 mm with a thickness of 170 µm. The substrate was cleaned before the realization of the printing process. Later on, 60 % v/v of MICROPOSIT S1813 positive photoresist was deployed on the surface of the substrate. The sample was then spin-coated onto the substrate at 4000 RPM for 30 seconds using a Karl Suss CT62 spin-coating machine (Fig. 2(a)). After the completion of the spin coating process, a layer of approximately 500 nm in thickness was achieved. The sample was then exposed using Nanoscribe GmbH’s Photonic Professional GT (Fig. 2(b)). The magnification and the numerical aperture of the objective were 63x and 1.4 respectively. After the TPL sept, the sample was developed using the MF-319 developer solution for 30 seconds at room temperature. Then the sample was cleaned by immersing it in deionized water. After development, 5 nm of chromium and 30 nm of gold were successively deposited on the sample by physical vapor deposition (Fig. 2(c)). The chromium was used to improve the adhesion of gold on the glass substrate. To remove the remaining photoresist from the sample, a lift-off process was employed by immersing the prepared sample in acetone and keeping it there for 10 hours (Fig. 2(d)). Finally, the sample was rinsed using de-ionized water and blow-dried using air.

Fig. 2Fabrication workflow for the printing of IDT structures using TPL followed by physical vapor deposition of gold and lift-off. (Renderings are done through the software Blender)

2.4. IDT structures characterizations

– Electron scanning microscopy; by using a sputter coater SC7640 from Quorum Technologies, the samples were metalized. A Field Emission Gun (FEG Hitachi SU8030) operating at 10 kV was used to perform scanning electron microscopy (SEM).

– Optical microscopy; A Nikon Eclipse LV100 upright microscope was used to capture optical microscopy images.

– Resonance frequency measurements.

The fabricated IDT structures were tested using a network analyzer to measure their frequency response and to determine the resonance frequency. The general representation of the setup is shown in Fig. 3.

Fig. 3Schematic diagram of the setup for temperature response

3. Results. Printed IDT

Following the workflow shown in Fig. 2 several IDT structures, shown in Fig. 4, were fabricated using the DLW technique. Scanning Electron Microscopy (SEM) images of one of the fabricated IDT structures are shown in Fig. 5.

Fig. 4The Fabricated IDT structures using DLW technique

Fig. 5SEM images of the fabricated IDT structure

Fig. 6Frequency response of the fabricated IDT structures

The frequency response of the developed sensor is shown in Fig. 6. The resonance frequency of the designed structure was predicted to be 424.01 MHz by the theoretical calculations and simulations. From the experiments, the value of resonance frequency was measured as 426.2 MHz. Comparing these two results, the deviation of 0.5 % is well within the acceptable range.

The illustration of the test setup is shown in Fig. 3. The optical microscope image of the IDTs used for temperature response analysis is given in Fig. 7. The frequency response of the fabricated IDTs was measured from 20 °C to 100 °C with a step size of 5 °C. The results are presented in Fig. 8.

Fig. 7Optical microscope image of the IDTs used for temperature response analysis

Fig. 8Temperature response of the five analyzed IDT structures. Each IDT structure was analyzed 10 times. Each point in the figures a)-e) corresponds to the average frequency measured for a given temperature, f) is the average of the responses of the 5 IDT structures. The red line is a linear fit

The fabricated IDT structures show very good linearity when tested against temperature. The average sensitivity is 0.123 MHz/°C. The average R2 values from 10 tests for each IDT structure along with the combined average value are shown in Fig. 8. The results show that the IDT structures exhibit a linear response towards the temperature changes.

The standard deviation and variance plot for each IDT is presented in Fig. 9. This variation can be due to more than one reason. Even though the conditions during the testing were mostly the same, it is possible that external noise and interference were present. Factors such as electromagnetic interference or acoustic noise can introduce additional variability in the observed response.

Fig. 9Standard deviation and variance plot for the fabricated IDTs

4. Conclusions

In this study, the micro-sized heat sensor is produced using the layered manufacturing technique. The 2PP method was used as a mask in the layered production technique. Usable results were obtained in the tests and analyzes of the heat sensor produced with this technique. The most important advantage in producing the IDT structure with the additive manufacturing technique is that a very small-sized structure is produced error-free and efficiently

The electron microscope images of the fabricated IDT structures showed consistency of the employed printing method with the proposed design. The frequency response of the fabricated IDT structures was measured using a test setup and the peak frequency was found to be 426.2 MHz. The simulations gave only a slightly deviating resonant frequency of 424.01 MHz. The results of those tests showed their response to be very much linear with the temperature changes. The average R2 value of all the responses was > 0.99.