Capacitive Micromachined Ultrasonic Transducers: Theory and Technology

Since the proposing of capacitive micromachined ultrasonic transducer by Khuri Yakub group in 1994 that this kind of transducer occupying the advantages of wide bandwidth, impedance matching well with the propagation medium especially in fluid and air and high sensitivity, has shown a great potential for wide ranges of applications. This Letter reports kind of micro‐electro‐mechanical systems (MEMS) capacitive ultrasonic transducer with the novel cavities embedded in the device layer of silicon on insulator wafer bonded with a glass substrate. The optimum geometric dimensions are confirmed by both mechanical vibrating of the membrane and the electrical characteristics analysis. Finite‐element analysis is adopted to determine the operation mode. The safety and reliability of the proposed device is ensured by the obtained deflections and equivalent stress under operation/collapse voltage. The bottom electrodes of the proposed transducer are fabricated on the top surface of the glass substrate. The parallel parasitic capacitance is reduced, thus improving the electromechanical coupling coefficient. The test results show that the electromechanical coupling coefficient is 69.65%, which demonstrates that this proposed MEMS capacitive ultrasonic transducer structure can enhance the performance significantly.

Since the proposing of capacitive micromachined ultrasonic transducer by Khuri Yakub group in 1994 that this kind of transducer occupying the advantages of wide bandwidth, impedance matching well with the propagation medium especially in fluid and air and high sensitivity, has shown a great potential for a wide range of applications such as medical imaging, medical ultrasound diagnosis, therapy and underwater ultrasonic imaging. This paper presents a kind of MEMS capacitive ultrasonic transducer with the novel cavities embedded in the device layer of Silicon On Insulator. Firstly, the geometric dimensions are derived from the theoretical analysis of the vibrating membrane and the electrical analysis of the transducer. Secondly, Finite Element Analysis is adopted to confirm the operation mode and the deflection and equivalent stress under operation/collapse voltage are obtained. Then, the fabrication process is depicted and the fabricated transducer is shown. The test results show that the electromechanical coupling coefficient is 69.65%.

Capacitive Micromachined Ultrasonic Transducers (CMUTs), introduced about few decades ago, have been shown to be a good alternative to conventional piezoelectric transducers in various aspects, such as sensitivity, transduction efficiency, and bandwidth. It is used in a wide range of applications including Non destructive testing applications, medical imaging, flow metering, micro/nanoelectronics and industrial cleaning etc. The advent of MEMS technology has lead to intensive study of micro machined ultrasonic transducers. They are similar to other capacitance transducers as they employ a vibrating membrane to send and receive ultrasound in air and in water. The advantage of hexagonal Capacitance micro machined ultrasonic transducers (CMUTs) is that they can be operated in the high temperature range and the transducer array can be fabricated at a lower cost. Presented in this paper is a study on some aspects of the design of a modified CMUT, which the authors believe would lead to a CMUT with many advantages, including less loading effect and higher packing density. The focus of discussion in the paper is FEM simulation of a single cell hexagonal CMUT using polysilicon as membrane material. Also discussed is the simulation study of the critical parameters which influence the operation of the CMUTs. The simulations were carried out using MATLAB and IntelliSuite 8.0. The study indicates that in the range of study, the best performance is achieved at the natural resonance frequency, which is very desirable.

Major steps used in fabricating surface micromachined capacitive ultrasonic immersion transducers are investigated. Such steps include membrane formation and cavity sealing under vacuum. Three transducer membrane structures are evaluated: a nitride membrane with an LTO sacrificial layer; a polysilicon membrane with an LTO sacrificial layer; and a nitride membrane with a polysilicon sacrificial layer. The major differences in the three processes are the conductivity, dielectric constant and residual stress of the membrane. Three vacuum sealing mechanisms are compared, each of which requires a different degree of lithographic sophistication, and results in a sealed cavity. Submicron via sealing requires sophisticated lithography, but is amenable to LPCVD nitride, LTO and other sealing procedures. Standard g-line lithography results in vias which seal only with high sticking coefficient species, such as LTO. A novel etch channel structure, which results in lateral sealing and requires neither sophisticated lithography nor a particular sealing is demonstrated. The experiments in the paper are guided by theoretical analysis and computer simulations when applicable. The optimized process based on a nitride membrane with a polysilicon sacrificial layer results in devices which have a broad band 50 /spl Omega/ real part impedance in the megahertz range. In addition, a transducer dynamic range in excess of 100 dB is achieved with an untuned bandwidth of 50%. The fabrication techniques and results herein reported indicate that surface micromachined capacitive ultrasonic transducers are an alternative to piezoelectric transducers in immersion applications.

Investigations into the characteristics of water-coupled ultrasonic capacitance transducers have been undertaken for a range of transducer configurations. The radiated fields have been scanned in water using a miniature hydrophone detector, and the results compared to theory based on a plane piston approach. Micromachined backplates in conjunction with thin Mylar and mica membranes have been investigated, together with aperture modifications such as an annulus and Fresnel zone plate. The measured results agree well with theory, thus demonstrating that wideband predictable performance (>8 MHz) is obtainable with such transducers. Additionally, pulse-echo C-scans of a Plexiglas plate containing an artificial defect have been undertaken with the capacitance transducer, in order to demonstrate that the capacitance transducer has sufficient sensitivity to allow routine nondestructive testing within immersion applications.

Ultrasound transducer is a crucial component for several imaging modalities, where acoustic sensing is utilized as a part of hybrid or combination of pure optical and ultrasound imaging. However, conventional ultrasound transducers are opaque, and the optical components in the system require a different pathway to avoid any interference. The absence of coaxial integration limits the optical illumination area, adds complexity, and makes the system bulky. Moreover, the system performance is deteriorated due to the unoptimized illumination in hybrid imaging and the lack of correlation between the data acquired through the combined modalities. To overcome these drawbacks, different approaches to realize optically transparent ultrasound transducers have been explored that includes piezoelectric-based and microelectromechanical system-based capacitive and piezoelectric ultrasonic transducers. Unlike conventional opaque transducers, each component of transparent ultrasound transducers requires specific materials and a certain recipe which all together determine the overall performance of the transducer. In this article, we have performed a comparative analysis on the materials and the fabrication process used to make different types of transparent ultrasound transducer.

Electrostatic or capacitive ultrasonic transducers (EUTs or CUTs) were originally developed from the condenser transmitters and microphones with grooved and roughened backplates in the 1930s. Significant progress in theoretical and experimental research has been made since the introduction of micromachining techniques at the end of the 1980s. We have seen the advancement not only from the roughened ultrasonic transducers to the 2‐D array micromachined capacitive ultrasonic transducers (mCUTs), but also from the grooved ultrasonic transducers to the micromachined capacitive Lamb wave transducers (mCLWTs). Thus, the theoretical TDK model originally developed for grooved ultrasonic transducers [L.‐F. Ge, Proceedings 16th Intl. Conf. Acoustics/135th ASA Meeting, l08l–1082, 1998; and Sciences in China (A), (in English), 42, 1308–1315 (1999)] can be used for mCLWTs. It is most important for a reliable model to be able to predict the transducer’s natural frequencies, at which the lowest‐order antisymmetric Lamb wave (A0 wave) is excited. As an example using a typical transducer [M. H. Badi et al., IEEE Trans. Ultrason. Ferroelectr. Freq. Control 50(9), 1191–1203 (2003)], the fundamental frequency predicted using the TDK model is 2.18 MHz, which agrees well with the measured result of 2.1 MHz. [Work supported by NSFC (60374044).]

Precise tracking of the rapid and complex 3D movement of eukaryotic flagella is important for understanding their roles in cellular motility, sensory functions, and resource acquisition. Yet, achieving accurate 3D kinematic reconstruction of flagellar beating patterns, particularly highly non-planar ones, remains challenging. Here we present holoV3C, a method based on Digital Holographic Microscopy (DHM) that allows precise, label-free 3D tracking of highly non-planar eukaryotic flagella with high temporal resolution. This algorithm leverages phase anomaly detection to provide a combination of high temporal and axial resolution, with 0.25µm for beating mouse sperm flagella and down to 53nm for polystyrene particles, across large sampling volumes in a computationally efficient manner. Algorithmic validation is performed by tracking mouse sperm flagella over time, capturing approximately 600 points along a single flagellum to achieve high axial resolution. Furthermore, we apply holoV3C to reconstruct the highly non-planar beating dynamics of the 200-nm-diameter flagellum of the protist Reclinomonas americana with a temporal resolution of 200 frames per second. By enabling 3D tracking of non-planar eukaryotic flagella, holoV3C can yield important insights to advance our understanding of flagellar dynamics, opening new avenues in the study of microorganism motility and its ecological roles.

Disposable ultrasonic contact transducers have been constructed with inexpensive PVDF films for nondestructive testing (NDT) applications. This paper reports the temperature-dependent ultrasonic performance of commercial polyvinylidene fluoride (PVDF) films and PVDF sensors. PVDF film was evaluated for its material properties of interest for ultrasonic transducer performance including the relative dielectric constant /spl epsi//sub r/, dielectric loss tangent tan /spl delta//sub e/, electromechanical coupling constant k/sub t/, mechanical quality factor Q/sub m/, and acoustic impedance Z. Disposable PVDF transducers were then coupled to both low and high acoustic impedance test panels and their ultrasonic insertion loss/sensitivity and frequency bandwidth measured as a function of temperature in both the pulse-echo and pitch-catch modes. The results yielded center frequencies in the 3.5-16 MHz range, with -6-dB fractional bandwidths for some of the transducers exceeding 100%. The temperature dependence of the transducers' performance is reported for the -40 to 80/spl deg/C range and shows an approximate linear decrease in center frequency and increase in fractional bandwidth with increasing temperature. These measured results are compared with predictions from computer simulations based on Mason's model. These tests showed that the same disposable transducers can be used for both ultrasonic and acoustic emission sensing NDT applications. This paper summarizes the quantitative ultrasonic and vibration sensing performance for all of the disposable PVDF film transducers which were tested.

During the manufacture and the installation process, defects are easy to introduce in prefabricated cable accessories. The existing standards of cable accessories product require that there should be no defects such as bubbles and metal impurities in insulation, but no corresponding detection means are put forward. At present, the main defect detection methods are high voltage test and partial discharge test after installation. It is easy to misjudge because of the installation process, environmental noise and other reasons, and cannot locate the specific location of the defect. It is useful to find these defects through ultrasonic detection and ensure the safety of the cable accessories. In this paper, a set of ultrasonic automatic defect detection system for SR (silicone rubber) was developed. The relationships between the detection ability and the system parameters were investigated. The axial resolution and sensitivity of the ultrasonic wave in the SR Cable joint sample were measured using excitation signal with different pulse width. The relationships between the resolution (or the sensitivity) and pulse width were investigated. The attenuation coefficient and axial resolution were measured by ultrasonic probes with different central frequencies. The relationships between the attenuation coefficient (or the axial resolution) and the central frequency were also investigated. Finally, the negative pulse with pulse width of 200ns was selected as the excitation signal and 2.5MHz was selected as the central frequency of the probe. With the proper system parameters, the attenuation coefficient was 0.059dB/mm and the axial resolution was about 3mm in SR. Two different defect imaging methods were developed by analyzing the amplitude, phase and frequency spectrum of the ultrasonic echoes. The first imaging method, which applied to defects such as air gaps and impurities, constructed images using the amplitude of the flaw echo. The second imaging method, which applied to defects such as uneven distribution of material, constructed images using the average sound velocity. SR samples with different thicknesses and types of defect (air gaps, and metal impurities) were scanned by the system. The results showed that the detection system could accurately identify defects larger than 1 mm. The depth of defects could reach 50mm. The ultrasonic detection method could be applied to instruct the insulation diagnosis for SR Cable accessories.

Abstract— Compared to PZT transducers in medical applications, CMUTs reported on so far have broader fractional bandwidth (FBW) but lower transduction efficiency (TX and RX) [1]. Most fabricated CMUTs reported in the literature carried membranes of uniform thickness. Since there is a performance trade-off between transduction efficiency and FBW when designing CMUTs with uniform membrane thickness, there is limited room for performance improvement in these devices. However, wafer-bonding-based CMUT fabrication provides design flexibility by allowing fabrication of membranes with different thickness profiles. Herein, CMUTs featuring piston-shaped membranes are developed to improve device performance. According to our theoretical predictions, piston-shaped membranes should improve the CMUT performance in terms of output pressure, sensitivity, and broader fractional bandwidth. The large ratio of second resonant harmonic frequency to first resonant frequency improves FBW. Increased electrical field intensity in the CMUT cavity (due to the larger equivalent spring constant of this membrane, compared to classical membranes) improves TX and RX. The device performance also benefits from a flatter membrane shape that allows greater average membrane displacement and intra-cavity electrostatic pressure. CMUTs featuring piston-shaped membranes with different geometric shapes were designed and fabricated. In order to make a fair comparison, all designs have a similar first resonant frequency, and all devices are equal in size. Fabricating CMUTs featuring piston-shaped membranes is a more complex process than fabricating CMUTs with uniformly thick membranes. However, no yield-loss was observed when CMUTs featuring piston-shaped membranes with different geometric shapes were fabricated. The device characterization was carried out with both pitch-catch (PC) and pulse-echo (PE) immersion tests in oil. These devices achieved ~100% improvement in transduction performance (TX and RX) over CMUTs with uniform membrane thickness. For CMUTs with square and rectangular membranes, FBW increased from ~110% to ~150% and from ~140% to ~175%, respectively, over CMUTs with uniformly thick membranes. The new devices produced a maximum output pressure exceeding 1 MPa. Finally, performance optimization using the geometric shape was the same in both CMUTs featuring piston-shaped membranes and CMUTs with uniform membrane thickness.

This paper describes design, fabrication, and test of an integrated micromachined ultrasound transducer (MUT). This MUT can work as an emitter and a receiver of ultrasonic signals in air and is intended to be used in applications requiring both acoustic signal generation and sensing. Among these applications there can be detection or distance measurements of different objects where the nature either of these objects or of their surroundings does not allow the use of light-based methods. The device has been fabricated in a 0.8 mum CMOS process combined with deep reactive ion etching, integrating in the same chip a suspended membrane (plate) and the associated interface electronics. The plate is thermally actuated at its resonance frequency (40 kHz) during emission. During reception, a piezoresistive bridge placed on the membrane is used for monitoring its deflections. The main advantage of the design using an actuator and a sensor integrated in one multilayer structure is its simplicity. Comparing with capacitive transducers where two-electrode structure is used, considerations in terms of mechanical stability and membrane damping as well as the fabrication process are not critical. The integrated solution allows the use of an amplifier at the output of the piezoresistive Wheatstone bridge which results in the sensitivity of 35 mV/Pa; the maximum acoustic pressure generated by the transducer at 10 mm is 5 mPa

Optimization for a piezoelectric micromachined ultrasonic transducer on electromechanical coupling coefficient is investigated theoretically. The transducer is composed of (Au/Cr)/PZT/(Au/Cr)/Ti(Adhesive)/Si/SiO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> /Si multimorph, while titanium is used as adhesive layer, (Au/Cr) as the electrodes,PZT as exciting element, Si as substrate material. A theoretical model of a circular thin-film piezoelectric actuator on a clamped elastic plate is cited to predict the performance of piezoelectric micromachined ultrasonic transducer. The model is used to explore the effect of design and process parameters, such as Si thickness, PZT thickness and adhesive thickness, on the electromechanical coupling coefficient. The results show that an optimum ratio of PZT to Si thickness can be found to achieve the maximum coupling coefficient. The results can be used to form a set of design guidelines for the performance optimization of piezoelectric micromachined ultrasonic transducer.

Conventional bone conduction transducers, which are relatively large, suffer from poor performance at low frequencies. A new type of electro-dynamic transducer, the balanced electromagnetic separation transducer (BEST), was developed to improve the performance of the conventional transducers. By using a balanced suspension principle, the quadratic distortion forces, as well as the static forces between the vibrating parts, are principally counterbalanced. Both the distortion and the size of the transducer can therefore be considerably reduced. Moreover, the static and dynamic magnetic fluxes are separated, except in the air gap regions, giving a more efficient transducer. For example, in comparison with a conventional B71 transducer, a prototype of the BEST has: Lower total harmonic distortion (THD), by 20-25 dB, and improved sensitivity by 10-20 dB for 100 to 1000 Hz and by 2-10 dB for 1 to 10 kHz. From a clinical point of view, the BEST offers a chance to measure bone thresholds, at 250 and 500 Hz, which are reliable at hearing levels not possible before. For example, at 250 Hz the BEST has 23 dB higher sensitivity than the B71; the THD is improved from 61% (B71) to 3.3% (BEST) at 40 dB HL (ISO 389-3, 1994).

Piezoelectric materials have dominated the ultrasonic transducer technology. Recently, capacitive micromachined ultrasonic transducers (CMUTs) have emerged as an alternative technology offering advantages such as wide bandwidth, ease of fabricating large arrays, and potential for integration with electronics. The aim of this paper is to demonstrate the viability of CMUTs for ultrasound imaging. We present the first pulse-echo phased array B-scan sector images using a 128-element, one-dimensional (1-D) linear CMUT array. We fabricated 64- and 128-element 1-D CMUT arrays with 100% yield and uniform element response across the arrays. These arrays have been operated in immersion with no failure or degradation in performance over the time. For imaging experiments, we built a resolution test phantom roughly mimicking the attenuation properties of soft tissue. We used a PC-based experimental system, including custom-designed electronic circuits to acquire the complete set of 128 x 128 RF A-scans from all transmit-receive element combinations. We obtained the pulse-echo frequency response by analyzing the echo signals from wire targets. These echo signals presented an 80% fractional bandwidth around 3 MHz, including the effect of attenuation in the propagating medium. We reconstructed the B-scan images with a sector angle of 90 degrees and an image depth of 210 mm through offline processing by using RF beamforming and synthetic phased array approaches. The measured 6-dB lateral and axial resolutions at 135 mm depth were 0.0144 radians and 0.3 mm, respectively. The electronic noise floor of the image was more than 50 dB below the maximum mainlobe magnitude. We also performed preliminary investigations on the effects of crosstalk among array elements on the image quality. In the near field, some artifacts were observable extending out from the array to a depth of 2 cm. A tail also was observed in the point spread function (PSF) in the axial direction, indicating the existence of crosstalk. The relative amplitude of this tail with respect to the mainlobe was less than -20 dB.