Piezoelectric transducer and manufacturing method therefor

Abstract

According to an example aspect of the present invention, there is provided piezoelectric transducer comprising silicon beam with a first piezoelectric thin film at a first end of the silicon beam and a second piezoelectric thin film at a second end of the silicon beam, the first and second piezoelectric films being separated from each other by a gap, and each of the first and second piezoelectric films comprising a layer of aluminium nitride between metal electrode layers.

Description

PIEZOELECTRIC TRANSDUCER AND MANUFACTURING METHOD THEREFORFIELD

[0001] The present disclosure relates to piezoelectric transducers.BACKGROUND

[0002] One advantage of piezoelectric micro-electro-mechanical systems, MEMS, PiezoMEMS, is its scalability, overcoming issues commonly associated with alternative transduction methods. However, the breadth and depth of studies on scaling PiezoMEMS to nano-electro-mechanical systems, NEMS, remain limited. Effective MEMS miniaturization not only improves energy efficiency and surface area reduction but also requires coupling of nanoscale physics, particularly Casimir forces, with the NEMS design, device operation, and fabrication opportunities.SUMMARY

[0003] According to some aspects, there is provided the subject-matter of the independent claims. Some embodiments are defined in the dependent claims.

[0004] According to a first aspect of the present disclosure, there is provided a piezoelectric transducer comprising silicon beam with a first piezoelectric thin film at a first end of the silicon beam and a second piezoelectric thin film at a second end of the silicon beam, the first and second piezoelectric films being separated from each other by a gap, and each of the first and second piezoelectric films comprising a layer of aluminium nitride between metal electrode layers.

[0005] According to a second aspect of the present disclosure, there is provided a method of producing a piezoelectric transducer according to the first aspect, comprising formation of the silicon beam on a silicon-on-insulator wafer, deposition of a first metal electrode using stage tilting sputtering or stage tilting evaporation, deposition of a piezoelectric aluminium nitride film using atomic layer deposition, deposition of a secondmetal electrode using stage tilting sputtering or stage tilting evaporation, etching of the second metal electrode using a photoresist mask, etching of the piezoelectric aluminium nitride film using argon and chlorine, etching of the first metal electrode using a photoresist mask separating the second metal electrode on the ends of the silicon beam.BRIEF DESCRIPTION OF THE DRAWINGS

[0006] FIGURE 1A is an overview of the NEMS AlN-based in-plane transducer in isometric view;

[0007] FIGURE IB is an overview of the NEMS AlN-based in-plane transducer in cross-sectional view;

[0008] FIGURE 2 illustrates an operational principle of a NEMS-based latching switch;

[0009] FIGURE 3A illustrates modeling of a NEMS in-plane AIN latching switch, displacement distribution at an equilibrium point;

[0010] FIGURE 3B illustrates modeling of a NEMS in-plane AIN latching switch, pull-in and pull-out;

[0011] FIGURE 4 illustrates a nanofabrication process of the in-plane AlN-based nanotransducer, fabrication flow and SEM images of the key steps;

[0012] FIGURE 5A illustrates a cross-sectional STEM micrograph of a nano AlN- based transducer cross-section;

[0013] FIGURE 5B illustrates a zoomed-in STEM cross-section showing AIN columnar growth on the sidewall in a nano AlN-based transducer;

[0014] FIGURE 5C illustrates a SAED pattern of an Si / Mo / AIN / Al structure of a nano AlN-based transducer, and

[0015] FIGURE 6 comprises TEM EDS maps of the SiO2-Si-Mo-AlN-Al material composition of a fabricated nanotransducer.EMBODIMENTS

[0016] Effective MEMS miniaturization not only improves energy efficiency and surface area reduction but also requires coupling of nanoscale physics, particularly Casimir forces, with the NEMS design, device operation, and fabrication opportunities. Here, combined with the finite element method, FEM, modeling, experimental aluminum nitride, AIN, nanotransducer fabrication is performed to address the fabrication challenges with beam-based in-plane nanostructures and to develop the piezoelectric NEMS design utilizing a controllable stiction behavior. The FEM modeling demonstrates that the designed structure featuring a 220 nm thick silicon, Si, beam with AIN layers located on vertical sidewalls exhibits both pull-in and pull-out behaviors. Based on modeling results, a complete piezoelectric sidewall transducer structure is fabricated. The final transducer represents a patterned Si nanobeam with the Molybdenum-AIN -Aluminum transducer located on the vertical sidewalls. The presented findings not only open new opportunities in the piezoelectric MEMS scaling but also establish a platform for the design of innovative "more- than-a-Moore" devices, such as in-plane mechanical latching switches.

[0017] For the past few decades, MEMS devices have been established as powerful platforms for high-precision sensing and actuation in the semiconductor industry.[1,2]Yet, one of the main trajectories of further MEMS development is closely linked with effective device scaling leading to a more extensive integration of NEMS technology?"'5Scaling from the microscale to the nanometer range is mainly driven by dual objectives: enhancing MEMS performance and exploitation of nanoscale physics phenomena in MEMS design. For the first target, MEMS miniaturization aims to lower the device’s power consumption and expedite response time. Simultaneously, the coupling of nanoscale phenomena with the MEMS scaling allows the integration of the additional functionalities associated with its smaller size, such as quantum mechanical electrodynamic forces.^ Nevertheless, MEMS / NEMS scaling also presents a unique set of challenges that must be overcome. As the dimensions decrease, size-dependent effects can have a more significant impact on device fabrication and even behavior.[6,7]Size-dependent effects such as surface contamination, mass-stiffness decoupling, and contact roughness can drastically change the nanofabrication methodologies, potentially impacting important device parameters such as effective mass and quality factor.

[0018] Amid these challenges, piezoelectricity emerged as one of the most promising approaches for sensing and actuation in M EMS J8Recent advances in piezoelectric miniaturization constitute a wide variety of applications: nanoelectromechanical relays?9'10energy harvesters11] and RF resonators?12] Beyond applications, the fabrication of PiezoMEMS has notably advanced to encompass complex geometries, exemplified by successful fabrication on complex structures like the vertical surfaces of cantilevers. Still, the current trend in piezoelectric MEMS development remains focused on out-of-the-plane bending devices, where a deposited layer of piezoelectric material on a flat surface plays a pivotal role in the device’s output. Consequently, substantial scaling of the surface area becomes unavoidably interrelated with a compromise in the overall performance of the device?13,14] Simultaneously, exploring alternative designs for more effective in-plane motion not only streamlines effective device scaling but also offers significant benefits from an application perspective.

[0019] Herein, we investigate the scaling potential of Piezoelectric NEMS devices by conducting modeling, fabrication, and structural characterization of an AlN-based in-plane NEMS transducer. Aluminum nitride, AIN, is stated as a highly promising piezoelectric material due to its compatibility with CMOS processes and the availability of various deposition techniques?15,16] Moreover, advanced deposition methods such as atomic layer deposition, ALD, and metal-organic-chemical-vapor-deposition, MOCVD, allow the implementation of AIN layer on vertical sidewalls, which plays a pivotal role in reaching the desired in-plane transducer behavior?I 5'16

[0020] However, some uncertainties arise towards the full integration of highly efficient in-plane piezoelectric devices at the nanoscale. These scaling challenges can be divided into two major categories. The first crucial aspect of the scaling transition lies in the shift of influence from volume -based interactions to surface-based interactions?17] As distances approach the sub-micron scale, surface forces such as Van der Waals and, in specific cases, Casimir forces, become increasingly prominent, giving rise to effects like stiction, adhesion, and surface tension, which are ultimately capable of dominating the behavior of nanoscale devices?18] Even though nanoscale forces have already been measured experimentally, the practical usage of Casimir Forces has still not been widely exploited in NEMS design. The proposed transducer concept is aimed to use these forces to imitate a latching relay with controllable stiction in pull-in and pull-out regimes. Subsequently, the coupling of in-plane motion, AIN deposition, and utilization of surface interactions placesstrict NEMS design constraints such as the necessity of parallel motion and the presence of contact area. Thus, the double-clamped beam structure emerges as the most viable approach that combines scaling opportunities with the potential for parallel motion. Nevertheless, the introduction of piezoelectric materials, such as AIN, brings forth additional challenges, such as the crystallographic orientation and the piezoelectric response. The unpattemed AIN layer on the fixed silicon beam proves ineffective in generating meaningful in-plane displacement, as the material lacks the ability to contract along the sidewall. Another group of challenges lies in the nanofabrication techniques themselves, as conventional MEMS processing methods may not be directly applicable to nanoscale dimensions19] For instance, the deposition of high-aspect-ratio layers on vertical sidewalls prohibits the use of more traditional high-resolution patterning techniques, such as lift-off.

[0021] In this work, by focusing on the nanofabrication of a piezoelectric multilayer sidewall structure of the AlN-based in-plane transducer, we aim to investigate the fundamental mechanisms behind the scaling process. To unleash the potential of the proposed structure, comprehensive finite-element modeling, FEM, was performed in the COMSOL Multiphysics, COMSOL, software package. Modeling results consider nanoscale interactions such as Casimir Forces and utilize them in the latching switch device realization. Furthermore, the AIN nanotransducer structure was fabricated using EBE lithography and ALD deposition techniques in the dimension ranges based on the modeling results. Finally, the fabricated structure was characterized by focused ion beam / scanning electron microscope (FIB / SEM) techniques with the subsequent transmission electron microscopy (TEM) analysis of the deposited and patterned thin films.

[0022] FIGURE 1A is an overview of the NEMS AlN-based in-plane transducer in isometric view. FIGURE IB is an overview of the NEMS AlN-based in-plane transducer in cross-sectional view.

[0023] Piezoelectric Transducer Design.

[0024] In order to comprehensively study the scaling prospects of MEMS piezoelectric devices, the proposed transducer structure based on piezoelectric thin film on vertical sidewalls is introduced. The potential device’s configuration is based on a piezoelectric unimorph actuator and consists of a double-clamped silicon beam with the sandwiched AIN layer between the top and bottom metal electrodes. As depicted in FIGURES 1A and IB the structure breakdown reveals that AIN vertical thin films areisolated from each other by means of the etched area in the middle of the beam. This separation of piezoelectric thin films is pivotal in facilitating in-plane parallel motion, a critical aspect of the device's operation with a beam structure. The deposition of aluminum and molybdenum for the upper and lower electrodes, respectively allows potential differences across the AIN layers when a voltage is applied. Two distinct design solutions were implemented to investigate the challenges that emerge during the transition of fabrication processes to the nanoscale. The first one involved the utilization of a doubleclamped beam structure. While the traditional piezoelectric cantilever design with a single fixed end has been widely established and proven effective, it lacks the capability to facilitate parallel plate motion, which is essential for studying precise nanoscale surface interactions?20Therefore, with the unpattemed double-clamped beam structure, the only contribution to the displacement is a strain generated from the non-fixed areas which could be only on the top and bottom of the beam. Ultimately it results in almost the absence of the desired parallel plate motion and necessitates a modification, involving the removal of the central portion of the beam, thereby leaving two AIN areas near the fixed ends. The second feature of the proposed design concept is a flat stiction contact area separated from the transducer beam with a narrow gap of 20-200 nm. Notably, this structure allows us to capture and study the complexities of the surface interaction phenomenon. While AIN thin films on one side of the beam drive the structure towards a fixed flat area, it is possible to observe the impact of the stiction or even Casimir Forces on the transducer behavior. Ultimately, the utilization of surface forces can lead to completely novel design approaches in NEMS. For instance, FIGURE 2 demonstrates a proposed design of a NEMS switch device. Initially, the switch is considered to be in the “OFF” position: no voltage is applied, and the gap between the fixed and movable area shows the open circuit condition. Then, by increasing the applied voltage on AIN sidewall layers, the beam deflects towards the contact area. At close proximity, <10 nm, pull-in effect occurs: surface interactions are strong enough to overcome the beam’s elastic restoration force resulting in stiction contact even without applied voltage on AIN. Eventually, to bring the beam back to the initial state from the “On” position, the opposite voltage signal is applied overcoming the stiction with the piezoelectric actuation and elastic restoration force together.

[0025] Results, modelling

[0026] In order to lay the groundwork for further transducer fabrication and understand the set of scaling challenges, a comprehensive FEM modeling was conducted.The simulations were performed in the COMSOL multiphysics software package. FIGURE 3 depicts the model breakdown with the introduced boundary conditions. As shown, the silicon beam is fixed from both ends with the two separated AIN thin film-based transducers. In turn, to reach the in-plane motion behavior of the transducer it is required to orientate the piezoelectric layers perpendicularly to vertical surfaces. This can be reached by means of setting additional rotated coordinate systems for each dedicated AIN layer, where the z and y axes are flipped on 90 degrees to imitate the correct growth direction of AIN of vertical sidewalls. The AIN wurtzite -type structure's elastic-stiffness coefficients, acquired through experimentation, are represented as a stiffness tensor matrix. With the stress-piezoelectric constants, these anisotropic parameters are utilized to define modelled material as a piezoelectric.[21,22]The model contains a multilayer structure Si-Mo-AIN-Al, where the top surface of Mo serves as ground potential with the voltage applied to the bottom border of Al. Surface interactions in the simulations are introduced by adding a contact boundary condition between moving and fixed areas, while the Casimir force is applied as a pressure to the contact area. Casimir force is set up with a distributive behavior through the contact area, having a maximum at the tipping point and a minimum at the starting point of the contact. It has already been reported that starting from the fundamental Casimir force equation (1) and culminating with the Lipschitz theory of electrodynamic interactions, including material-specific corrections, it is possible to reach relatively accurate estimations of the Casimir force?2’26Similarly, as in equation (1), the stiction force in the model behaves as a gap-dependent pressure.

[0028] Simulation results showed that it is possible to couple piezoelectric transduction with the surface physics at the nanoscale with the optimal choices of dimensions. Through the transducer’s operation, the structure is balanced between three different forces: surface attraction expressed as Casimir force, elastic restoration of the deformed beam, and the piezoelectric actuation. According to simulated results in FIGURE 3A, with the initial 20 nm gap, the pull-in point, where Casimir forces are strong enough to create stiction, can be observed at 21 V. Moreover, when the gap distance comes closer to 2-4 nm, the Casimir force becomes strong enough to overcome the elastic restoration of the beam. At this point, we can assume that the moving beam's surface will be in contact with the fixed electrode even without applied voltage. In contrast, to validate the capability of theproposed piezoelectric transducer in harnessing the latching feature of NEMS and ensuring controlled "pull-in" conditions, the opposite FEM study was conducted. Specifically, we reduced the gap to the point of contact and estimated the requisite voltage amplitude necessary to overcome stiction effects. The results, as depicted in FIGURE 3B, reveal that a voltage of 5 V is sufficient to break the Casimir stiction. Notably, this simulation represents the worst-case scenario, as it does not account for elastic restoration effects. Ideally, the predeformed cantilever will require less voltage because elastic force will have the same direction as the piezoelectric force.

[0029] Fabrication

[0030] The piezoelectric sidewall in-plane nano transducer was fabricated using a CMOS-compatible nanofabrication process, including a combination of electron-beam lithography and dry etching methods, as depicted in Figure 4. Traditionally the patterning of metal electrodes is mostly concentrated around an out-of-the-plane design approach of Piezoelectric NEMS devices. f9,27] This design allows to leverage the lift-off technique in the patterning of metal and piezoelectric layers. However, this method is not suitable, where high-aspect-ratio sidewall structures play a crucial role. Instead, in the proposed transducer each patterning step requires a complete etching process with the photoresist mask.

[0031] Based on modelling results for transducer fabrication Silicon-on-insulator substrate with 220 nm device layer of Silicon and 2 pm of the buried oxide (BOX) layer was selected. Since the sidewall structure involves the pattering of several layers with multiple masks, precise alignment marks formation plays a pivotal role in the nanotransducer process. There are two strategies to utilize alignment in EBL: 1) to have a sufficiently high material (atomic number) contrast, and 2) to exhibit a sufficiently high vertical surface topography (at least 2 pm to be visible at 100 kV electrons). The fabrication started with the second method to fully take advantage of the utilization of SOI structure as depicted in Figure 4: a) Precise alignment marks cavities in Si device layer were etched in cryogenic mode with SFe plasma through AR-P 6200 photoresist mask; b) SiO2 BOX etching was performed in BHF solution to reach high vertical surface difference; c,j) Nano-beam formation by cryogenic plasma etching of Si device layer; d,k) 50 nm of the Mo was deposited by stage tilting sputtering process to form uniform step coverage of the bottom electrode; e,l) c-axis orientated AIN film was deposited through atomic layer deposition (ALD) with in-situ atomic layer annealing which previously showed significant crystal quality improvement ofAIN on vertical sidewalls

[0015] ; f) 50 nm of Al was evaporated with stage tilting process to form top electrode; g,m) etching of Al layer with the AR-P 6200 photoresist mask by combination of BCh and Ch plasma; h) etching of AIN in Ar + Ch plasma chemistry; i) bottom Mo layer was etched by means of CF4 + O2 plasma process to form a complete transducer structure.

[0032] Characterization

[0033] To systematically study the structure of the fabricated nanotransducer, transmission electron microscopy, TEM, scanning transmission electron microscopy, STEM, and selected area electron diffraction, SAED, were used. FIGURES 5A - 5C depict both bright field, BF, STEM micrographs as well as SADP. STEM analysis confirmed that the combination of high-resolution EBL and dry etching techniques allows the complete multilayer transducer structure on the vertical sidewalls. AIN thin films on vertical sidewalls as a key structure in the proposed nanoscale transducer design can be characterized mainly by the following figures of merit: conformity and crystal orientation. Firstly, the non- uniform AIN film on the vertical surface may result in parasitic cross-talk between nonlateral motion, compromising the overall efficiency of the device. As shown in Figure 5b, conformal deposition allows the piezoelectric thin film to coat and, furthermore, pattern complex sidewall surfaces uniformly. When it comes to the AIN crystal quality, the deposited thin film must keep the c-axis orientation perpendicular to the underlying surface for the highest electromechanical coupling. Figure 5b shows that implemented ALD deposition with in-situ atomic layer annealing allows to reach crystalline AIN with a columnar structure in the growth direction. Previously reported results have shown that in addition to improvements by in-situ plasma annealing in the ALD process, AIN crystal quality also strongly relies on the careful material selection of the underlying layer. Thus, earlier observations revealed that the best crystal quality of AIN was achieved on the Al layer with a (111) orientation. 'I 5Ultimately, the presented STEM results demonstrate that the integration of molybdenum into the nanotransducer process could be driven not only by the fact that it is a traditional electrode material but also by being a promising candidate for achieving c-axis orientated AIN on vertical sidewalls. The SADP seen in Figure 5c shows that the AIN grown on the sidewall exhibits strong c-axis preferential orientation perpendicular to the growth surface, necessary for good piezoelectric response. The considerable thickness of AIN in relation to the Si device layer (-120-140 nm vs 220 nm)played a vital role in the presence of both equal components of 002 direction from the bottom and sidewall surfaces respectively.

[0034] Energy-dispersive X-ray spectroscopy, EDS, analysis was undertaken to perform accurate mapping of the material composition on the sidewalls of the nanotransducer interface. FIGURE 6 depicts the color maps of all deposited and patterned materials. Results revealed that the planned Si-electrode-AlN-electrode structure is reached on vertical sidewalls. However, despite the fact that Al top electrode thickness is not uniform around the patterned surface, the key sidewall area, which ultimately will be utilized in the in-plane actuation, is present.

[0035] Conclusion

[0036] In summary, it is proposed to unlock the scaling potential of MEMS through the innovative design, modeling, and fabrication of a nanoscale in-plane-based transducer structure. To demonstrate the coupling of nanoscale physics and the double-clamped beam structure with vertical piezoelectric sidewalls we introduce the latching mechanism with the controllable pull-in phenomena. The conducted modeling validates that the right balance between the active piezoelectric actuation, elastic restoration, and Casimir forces culminates in the mechanical switch behavior featuring controllable pull-in / pull-out conditions. Based on the modelling results, the nanotransducer was fabricated utilizing a Si / Mo / AIN / Al material stack. The nanofabrication process reveals the successful realization of a multilayer piezoelectric structure on vertical sidewalls on 220 nm device layer SOI chips. The multilayer patterning involved separate steps for each dedicated layer due to the incompatibility of the lift-off technique with the in-plane structure. The deposition of AIN was conducted using the ALD method with the in-situ plasma annealing, while by means of physical vapor deposition the top and bottom electrodes were fabricated, Al and Mo respectively. STEM characterization further underscores the promising quality of the c-axis orientation of the ALD AIN layer with the distinguishable growth directions perpendicular to the sidewall and bottom surfaces and demonstrates the potential of Mo as an underlying layer. Ultimately, the adaptability of our beam-based design, utilizing a SOI structure, extends to free-standing release combined with the possibility of device-level characterization. This study provides a promising approach towards broadening the scope to full-scale in-plane NEMS integration.

[0037] Experiments

[0038] FEM Simulation: FEM simulation was performed in the COMSOL multiphysics software package with the solid mechanics and electrostatics modules. The AIN elasticity and coupling matrices were assumed based on experimental data,[21,22]the mass density was p = 3300 kg m-3, and the relative permittivity was & = 9. In-plane AIN orientation was implemented through the built-in base vector system (XZ-plane). The fixed boundary condition was applied to both sides of the beam, while the contact pair was defined between the beam’s middle area and the fixed contact surface. The potential difference was applied to the AIN layer to create parallel motion and the Casimir force was set up as a gapdependent pressure expressed by equation 1.

[0039] NanoTransducer Fabrication: 400 nm of AR-P 6200 EBL resist was spin- coated to a SOI chip with 220 nm of device layer. The alignment marks were patterned by a 100 keV EPBG5000pES EBL system with a 400 uC cm-2. Exposure was followed by a development in the AR 600-546 solution for 1 min and IPA for 30 sec. Si etching was performed by inductively coupled plasma (ICP) cryogenic process with SFe and O2 gases. Then 2 pm of buried oxide layer was removed in the 1:10 HF mixture. For the beam formation, the same lithography + Si etching steps were repeated. Then 50 nm layer of Mo was deposited by magnetron sputtering with a 70 deg tilt, followed by ALD deposition of AIN. ALD was done following the procedure described in previously reported studies.

[0015] In turn, for the top electrode 50 nm of Al were thermally evaporated. All further patterning steps utilized the same AR-P 6200 photoresist mask with the mentioned before the EBL process. Finally, three subsequent etching processes took place: 1) ICP etching of Al in Ch and BCh chemistry, 2) AIN layer was removed in Ch and Ar ICP process, 3) RIE etching of Mo in CF4+O2 plasma.

[0040] STEM Characterization: First, the sample was prepared in the dual-beam Focused ion beam SEM system JEOL JIB-4700F aiming at electron-transparent lamella. Preliminary the additional layer of Platinum was deposited upon crucial structure to enhance protection during the Ga-ion milling process. Obtained lamella then was imaged in the high- resolution JEOL JEM-2800 microscope.

[0041] It is to be understood that the embodiments of the invention disclosed are not limited to the particular structures, process steps, or materials disclosed herein, but are extended to equivalents thereof as would be recognized by those ordinarily skilled in the relevant arts. It should also be understood that terminology employed herein is used for thepurpose of describing particular embodiments only and is not intended to be limiting.

[0042] Reference throughout this specification to one embodiment or an embodiment means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the present invention. Thus, appearances of the phrases “in one embodiment” or “in an embodiment” in various places throughout this specification are not necessarily all referring to the same embodiment. Where reference is made to a numerical value using a term such as, for example, about or substantially, the exact numerical value is also disclosed.

[0043] As used herein, a plurality of items, structural elements, compositional elements, and / or materials may be presented in a common list for convenience. However, these lists should be construed as though each member of the list is individually identified as a separate and unique member. Thus, no individual member of such list should be construed as a de facto equivalent of any other member of the same list solely based on their presentation in a common group without indications to the contrary. In addition, various embodiments and example of the present invention may be referred to herein along with alternatives for the various components thereof. It is understood that such embodiments, examples, and alternatives are not to be construed as de facto equivalents of one another, but are to be considered as separate and autonomous representations of the present invention.

[0044] Furthermore, the described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the preceding description, numerous specific details are provided, such as examples of lengths, widths, shapes, etc., to provide a thorough understanding of embodiments of the invention. One skilled in the relevant art will recognize, however, that the invention can be practiced without one or more of the specific details, or with other methods, components, materials, etc. In other instances, well-known structures, materials, or operations are not shown or described in detail to avoid obscuring aspects of the invention.

[0045] While the forgoing examples are illustrative of the principles of the present invention in one or more particular applications, it will be apparent to those of ordinary skill in the art that numerous modifications in form, usage and details of implementation can be made without the exercise of inventive faculty, and without departing from the principles and concepts of the invention. Accordingly, it is not intended that the invention be limited, except as by the claims set forth below.

[0046] The verbs “to comprise” and “to include” are used in this document as open limitations that neither exclude nor require the existence of also un-recited features. The features recited in depending claims are mutually freely combinable unless otherwise explicitly stated. Furthermore, it is to be understood that the use of "a" or "an", that is, a singular form, throughout this document does not exclude a plurality.

[0047] As used herein, “at least one of the following: ” and “at least one of ” and similar wording, where the list of two or more elements are joined by “and” or “or”, mean at least any one of the elements, or at least any two or more of the elements, or at least all the elements.INDUSTRIAL APPLICABILITY

[0048] At least some embodiments of the present invention find industrial application in nanoconstruction.CITATION LIST[1] A. S. Algamili, M. H. M. Khir, J. O. Dennis, A. Y. Ahmed, S. S. Alabsi, S. S. Ba Hashwan, M. M. Junaid, Nanosc. Res. Lett. 2021, 16, 1.[2] Yole Intelligence, Status of the MEMS Industry 2023, https: / / www.yolegroup.com / product / report / status-of-the-mems-industry-2023, accessed: November, 2023.[3] J. Zhu, X. Liu, Q. Shi, T. He, Z. Sun, X. Guo, W. Liu, O. Bin Sulaiman, B. Dong, C. Lee, Micromachines 2019, 11, 7.[4] T. Cao, T. Hu, Y. Zhao, Micromachines 2020, 11, 694.[5] L. Wei, X. Kuai, Y. Bao, J. Wei, L. Yang, P. Song, M. Zhang, F. Yang, X. Wang, Micromachines 2021, 12, 724.[6] F. W. Delrio, M. P. De Boer, J. A. Knapp, E. D. Reedy, P. J. Clews, M. E.Dunn, Nat. Mater. 2005, 4, 629.[7] H. Sadeghian, J. F. L. Goosen, F. Van Keulen, C. K. Yang, A. Bossche, P. J. French Sens. Actuators A Phys. 2010, 2, 162[8] X. Ee, Q. Shi, P. Vachon, E. J. Ng, C. Fee, J. Micromech. and Microeng.2021, 32, 014005.[9] U. Zaghloul, G. Piazza, IEEE Trans. Electron. Devices 2014, 61, 3520.

[0010] R. B. Karabalin, M. H. Matheny, X. L. Feng, E. Defa, G. Le Rhun, C. Marcoux, S. Hentz, P. Andreucci, M. L. Roukes, Appl. Phys. Lett. 2009, 95, 103111.

[0011] G. Zhu, R. Yang, S. Wang, Z. L. Wang, Nano. Lett. 2010, 10, 3151.

[0012] R. B. Karabalin, M. H. Matheny, X. L. Feng, E. Defa, G. Le Rhun, C. Marcoux, S. Hentz, P. Andreucci, M. L. Roukes, Appl. Phys. Lett. 2009, 95, 103111.

[0013] N. Sinha, G. E. Wabiszewski, R. Mahameed, V. V. Felmetsger, S. M.Tanner, R. W. Carpick, G. Piazza, Appl. Phys. Lett. 2009, 95, 53106.

[0014] M. Rinaldi, C. Zuniga, G. Piazza, Proc. IEEE Int. Conf, on Micro Electro Meeh. Sys. 2009, 916.

[0015] E. Osterlund, H. Seppanen, K. Bespalova, V. Miikkulainen, M. Paulasto- Krockel, J. Vac. Science & Tech. A 2021, 39, 32403.

[0016] K. Bespalova, T. Nieminen, A. Gabrelian, G. Ross, M. Paulasto-Krockel, Adv. Electron. Mater. 2023, 9, 2300015.

[0017] S. Baglio, S. Castorina, N. Savalli, Scaling Issues and Design ofMEMS, John Wiley & Sons, 2007.

[0018] R. Yang, J. Qian, P. X-L Feng, R. Yang, J. Qian, P. X-L Feng, Small 2020, 16, 2005594.

[0019] P. van Assenbergh, E. Meinders, J. Geraedts, D. P. Dodou van Assenbergh, D. Dodou, E. Meinders, J. Geraedts, Small 2018, 14, 1703401.

[0020] M. Wang, L. Tang, C. Y. Ng, R. Messina, B. Guizal, J. A. Crosse, M. Antezza, C. T. Chan, H. B. Chan, Nat. Comm. 2021, 12, 1.

[0021] V. Bougrov, M. E. Levinshtein, S. L. Rumyantsev, A. Zubrilov, Properties of Advanced Semiconductor Materials: GaN, AIN, InN, BN, SiC, SiGe, John Wiley & Sons, 2001

[0022] A. F. Wright, J. Appl. Phys. 1997, 82, 2833.

[0023] B. V. Derjaguin, I. Abrikossova, J. Phys, and Chem. of Solids 1958, 5, 1.

[0024] S. G. Johnson, Numerical methods for computing Casimir interactions, Leet. Notes Phys. Springer, Berlin, 2011, 834, p.175-218.

[0025] I. E. Dzyaloshinskii, E. M. Lifshitz, L. P. Pitaevskii, Adv. Phys. 1961, 10, 165.

[0026] H. B. Casimir, in Proc. Kon. Ned. Akad. Wet. 1948, 51, p. 793.

[0027] B. D. Davidson, D. Seghete, S. M. George, V. M. Bright, Sens. Actuators A Phys. 2011, 166, 269.

Claims

CLAIMS:

1. A piezoelectric transducer comprising:- silicon beam with a first piezoelectric thin film at a first end of the silicon beam and a second piezoelectric thin film at a second end of the silicon beam;- the first and second piezoelectric films being separated from each other by a gap, and- each of the first and second piezoelectric films comprising a layer of aluminium nitride between metal electrode layers.

2. The piezoelectric transducer according to claim 1, wherein the gap separating the first and second piezoelectric films is between 20 and 200 nanometres long.

3. The piezoelectric transducer according to claim 1 or 2, wherein the metal electrodes include aluminium, titanium, molybdenum, tantalum, platinum, titanium nitride and / or tantalum nitride.

4. The piezoelectric transducer according to claim 3, wherein the metal electrodes comprise molybdenum facing the silicon beam and aluminium facing away from the silicon beam.

5. The piezoelectric transducer according to claim 4, wherein the metal electrodes are each 50 nanometres thick.

6. The piezoelectric transducer according to claim 4, wherein the metal electrodes are each between 10 and 100 nanometres thick.

7. The piezoelectric transducer according to any of claims 1 - 6, wherein the layer of aluminium nitride is 100 or 120 - 140 nanometres thick.

8. The piezoelectric transducer according to any of claims 1 - 6, wherein the layer of aluminium nitride is between 50 and 200 nanometres thick.

9. A method of producing a piezoelectric transducer according to one of claims 1 - 8, comprising:- formation of the silicon beam on a silicon-on-insulator wafer;- deposition of a first metal electrode using stage tilting sputtering or stage tilting evaporation;- deposition of a piezoelectric aluminium nitride film using atomic layer deposition; - deposition of a second metal electrode using stage tilting sputtering or stage tilting evaporation;- etching of the second metal electrode using a photoresist mask;- etching of the piezoelectric aluminium nitride film using argon and chlorine;- etching of the first metal electrode using a photoresist mask separating the second metal electrode on the ends of the silicon beam.

10. The method according to claim 9, wherein the first metal electrode is a molybdenum electrode and the second metal electrode is an aluminium electrode layer, both first and second metal electrode layers being 50 nanometres thick.

Images