Piezoelectric materials lie at the heart of ultrasonic transducers. These materials convert electrical energy into mechanical form when generating an interrogating acoustic pulse and convert mechanical energy into an electric signal when detecting its echoes. This paper first surveys the piezoelectric materials in current use: piezoceramics, such as barium titanate, lead zirconate titanate, and modified lead titanate; piezopolymers, such as polyvinylidene difluoride and its copolymer with trifluroethylene; and piezocomposites, consisting of piezoceramic rods in a passive polymer matrix. Each material system has properties which commend them for use in the present single element transducers, annular arrays, sequenced linear arrays, and steered phased arrays. Looking to the future, new transducer possibilities are opening up due to recent piezoelectric material developments, such as, for example, synthesis techniques for fine-grained high-density piezoceramics, electrostrictive relaxor ferroelectric ceramics, novel piezoceramic forming methods, piezoceramic fiber synthesis, piezoceramic/metal multilayer structures, composite acoustoelectric materials, ferroelectric thin film growth and processing, and new piezopolymers. These innovations lead to fabrication of conventional transducers at high frequencies, fine-scale piezocomposites, 11/2-D and 2-D arrays, small intravascular transducers, as well as provide opportunities for new ultrasonic imaging techniques, using pitch-catch and non-resonant traveling wave transducers.

Piezoelectric polymers, such as PVDF and its copolymers, are finding increased use for ultrasound transducers in both medical and nondestructive testing applications. Because of their inherent properties of high compliance, low acoustic impedance, availability in large areas, and broadband acoustic performance, they are particularly useful in medical applications that require miniature transducers for high-frequency/high-resolution and low ultrasonic penetration, such as invasive medical imaging. The technology also provides great utility in nondestructive testing applications which require low-profile ultrasonic inspection of fiber-composite structures, especially for non-planar surfaces. This paper reviews some recent developments in piezoelectric polymer ultrasound transducer technology in both the medical and nondestructive testing (NDT) application areas. The presentation covers recent developments in invasive medical ultrasound transducers and transducer systems that employ the piezoelectric polymer technology for such applications as intraluminal imaging of the coronary arteries. A discussion of piezoelectric polymer ultrasound sensor arrays for NDT of fiber composite structures is also included.

Relaxor ferroelectric materials have been studied extensively in both theory and experiments for many years. Their applications to medical ultrasonic transducers have also been investigated. In this paper, we report an experimental study of a composite relaxor ferroelectric transducer and its nonlinear phenomenon at certain bias voltages. Novel applications of the relaxor transducer to diffraction-limited beam production, acoustic power measurement, low speckle medical imaging, and high resolution pulse-echo imaging are discussed.

This work describes the evaluation of various 1 - 3 connectivity transducer configurations, comprising a matrix of ceramic rods embedded in epoxy, for operation into air over the frequency range 100 kHz - 2 MHz. A dual strategy, involving simulation design and supported by experimental verification, is used to determine the main factors which influence through air operation of such structures. Specifically, finite element analysis is employed to determine the influence of ceramic rod shape and distribution, in conjunction with the characteristics of the epoxy filler materials, on transducer performance. A one dimensional linear systems model is then utilized for assessment of transducer behavior when configured as an actual probe assembly and connected to practical electrical and mechanical load environments. Some experimental examples, relevant to non-destructive evaluation, are presented, including through transmission scanning of carbon-fiber composite materials and remote detection of laser generated ultrasound.

The finite element method has been applied to the problem of predicting the eigenfrequencies, displacements, and stresses within an ultrasonic transducer. In the case of transducers suitable for ocean survey, fish detection, or air-ranging, these parameters are useful only as general predictors of transducer performance. In the current presentation, the finite element method has been extended to the prediction of parameters more directly useful to the designer. A finite element program has been written in Fortran to compile and run on a 33 MHz/386 PC. Eigenfrequencies, impedance, transmit and receive sensitivity, radiation pattern, displacement, and shear and principle stresses can be predicted for a transducer in its operating medium. Damping in acoustic isolation, backing, and matching layer materials are included in the model to provide an accurate and comprehensive design tool. Experimental results for different designs corroborating the predictions from the finite element models are presented. With several design iterations possible in an hour, new transducers have been designed in a day or so.

Diffraction-limited beams were first discovered by Durnin in 1987. These beams are pencil- like and have very large depth of field. Recently, we have discovered new families of diffraction-limited beams which contain some of the diffraction-limited beams known previously, such as, the plane wave and Durnin's Bessel beams, in addition to an infinite variety of new beams, such as X waves. In this paper, we generalize the new diffraction- limited beams to n-dimensional space, review the recent development of the diffraction-limited beams, and describe their applications to medical ultrasonic imaging, tissue characterization and nondestructive evaluation of materials. Advantages and disadvantages of these beams are discussed and their possible applications to other wave related fields are addressed.

The main requirement of pulsed-echo ultrasound applications such as cardiac imaging, 3-D imaging, or blood flow imaging can be identified as frame rate. Currently, frame rate is limited by the imaging depth range and the number of ultrasonic fires. For example, a 15 cm imaging range gives rise to 200 microsecond(s) lines or to a 2 s acquisition time for 100 planes of 100 lines in 3-D medical applications. The only way to increase frame rate is parallel beam formation in the receive mode. Simultaneous parallel beam forming allows us to divide the acquisition time by a factor proportional to the number of beam formed lines. In the technique developed by S. W. Smith, an increase frame rate of 16 is achieved. However, this technique is limited by a loss in lateral resolution due to the requirement for a wide illumination beam of the explored medium in the transmit mode. We propose an alternate illumination scheme that minimizes the loss in resolution in the transmit mode. In this technique, ultrasonic energy is transmitted simultaneously in several narrow beams. This technique works in pulsed mode and we have built the hardware needed for the simultaneous production of several beams. Each transducer is connected to a transmitter able to generate a sequence of excitation pulses. If n beams are to be transmitted, the excitation signal is the sum of n cylindrical wave fronts. For those, among elements where the n wave fronts are disjoint, the excitation signal is thus the succession of n pulses with specific time positions with respect to the system synchronization. For the others, the excitation signal is more complex since it consists in a (n - 1) level signal (0, 1, 2, ... n - 1 times the basic excitation signal). We show the performances of such a parallel transmit scheme based on beam plots as well as on tissue phantom images. This leads to an evaluation of the maximal number of beams compatible with current medical imaging quality standards. It is shown that a gain of 16 in the acquisition time can be achieved without any loss in lateral resolution.

A frequency scanning beam steering technique using an alternating polarity transducer array is investigated for presentation of B-mode ultrasound images for front viewing intracavitary applications. Arrays of alternating polarity PZT elements were computer simulated and fabricated to characterize their beam steering and resolution capabilities. A spherical lens was mounted in front of the array to improve focusing. A personal computer based system was used to test and generate sector images of wire targets using these arrays.

Recent advances in the field of medical ultrasound requires a dual frequency transducer capable of operation at two frequencies. This is usually achieved by having a broadband transducer which can be excited at two different frequencies. Such frequency characteristics will allow echo-amplitude imaging at the higher frequency where a high resolution is required and color flow imaging (CFI) at the lower frequency for improved sensitivity. In this paper a novel dual frequency transducer design, demonstrating two distinct resonance, is presented. This is achieved by varying the geometry of the piezoelectric ceramic in a manner which results in two distinct thickness mode resonance. The results showing the temporal characteristics of such transducer are presented. The parameters influencing the temporal response of this transducer are discussed. It is also shown that the two resonance frequencies are independent. Finally, the advantages of this new design over the conventional techniques are discussed.

Dynamic focusing is known to be the best way to improve image resolution, but it needs very complex and bulky hardware for an array transducer with a large number of elements, such as a sector phased array with 128 elements or a two-dimensional array with more elements. We describe an efficient architecture for ultrasonic dynamic focusing using both analog and digital processing techniques to minimize the hardware complexity. This new focusing hardware consists of a specially designed analog first-in-first-out (AFIFO) device, one for each element, a sampling clock generator (SCG), and a simple multi-input analog adder for a whole system. The SCG generates nonuniform sampling clocks to adjust the sampling instances of the input pulse echo of each AFIFO corresponding to the propagation times between the array element and each imaging point. The AFIFO stores and outputs analog values of these samples on a first-in first-out basis. Finally, the outputs of AFIFOs are summed together by the analog adder. We discuss the hardware reduction ratio of the proposed architecture compared with conventional ones. Preliminary experimental results also are presented.

A rapidly expanding area of medical treatment is using small invasive devices, e.g., balloon angioplasty catheters, to eliminate the need for conventional open surgery. The usual x-ray guidance requires patient and physician irradiation and the injection of contrast media, both undesirable. Ultrasound guidance, which would eliminate these hazards, has not been used because of the difficulty in determining with certainty the exact location of a particular point on the invasive device. By placing a transducer at such a point to act as a beacon, exact positioning by ultrasound imaging has been achieved. The required transducer's response must be almost omnidirectional, so that it detects the imaging system's beam independently of angle; the size of the transducer must be small, so that the device can penetrate into the body easily; finally, the cost of the transducer must be low, so that it may be thrown away after one use. We show how the transducer is designed to achieve the required angular response and size, and outline how the required transducers can be fabricated at low cost.

With the current clinical interest in invasive probes, small size has become paramount in many applications. When a mechanically steered probe is reduced in size, the exit window becomes tightly curved around the transducer. Angles of incidence upon that window increase, differences in sound speed at interfaces are more significant, and refraction increases. The incident angles may approach critical angles. How much does this hurt performance? A ray tracing technique of predicting field behavior is used to analyze the performance of annular array transesophageal probes, as an example. The properties of several different candidate polymers and fluids properties were determined at 37 degree(s)C. Probe performance was calculated when these comprised the fluid and exit window. With sound speed differences of 20%, degradation of resolution is significant, but subtle. Electronic focusing has to be chosen to produce optimum resolution.

The rapid increase in the use of ultrasound in both clinical and industrial applications requires more advanced and reliable imaging systems for calibrating and characterizing high performance ultrasonic transducers. The optically scanned hydrophone (OSH) is an alternative imaging system capable of quasi-real time imaging of broadband acoustic fields. The main application of the OSH is in the imaging and characterization of acoustic fields such as those emitted from clinical and therapeutic transducers. In this paper, the recent development of the OSH and its application to real time imaging of broadband acoustic fields are reported. Using improved fabrication techniques the optical sampling efficiency of the OSH has been considerably improved. This is achieved by adopting new assembly techniques and incorporating a novel differential electrode configuration. The improved optical sampling efficiency has provided a more competitive, versatile, and faster imaging system. The performance of the modified OSH is compared against the other types of hydrophone such as the spot poled and the needle types.

Two-dimensional (2-D) transducer arrays offer the potential for improving medical ultrasound imaging by producing symmetrically focused ultrasound beams which can be steered throughout a three-dimensional volume. Theoretical investigations of the beamforming properties of 2-D arrays have characterized the array parameters required to steer the beam up to 45 degree(s) off-axis. These investigations have also shown that the number of elements in a steered 2-D array can be dramatically reduced using a sparse set of elements, randomly distributed throughout the transducer aperture. The penalty paid for the use of a sparse array is the development of a `pedestal' sidelobe in the beam profile, the amplitude of which increases as the number of elements in the array decreases. The potential of 2-D arrays for medical imaging has been assessed by simulating images of spherical lesions embedded in a random scattering medium. Similar contrast characteristics over a range of cyst sizes are demonstrated for a dense 2-D array and a sparse array with 1/8th the number of elements, both operating at 5 MHz. A 32nd order sparse array was found to perform at a reduced level, producing unacceptable artifactual echoes within images of cysts. Experimental results are described which verify some of the theoretical predictions.

Ceramic-epoxy composite transducers, in the form of a matrix of piezoceramic rods embedded in an epoxy substrate, offer significant advantages for sonar array design in the frequency range 100 kHz - 2 MHz. Good matching to the water load, coupled with high sensitivity, low lateral crosstalk, and wideband performance are extremely attractive features for modern array design. This paper describes a simulation design strategy for both 1-D and 2-D composite array configurations, with specific emphasis placed on the latter structure. The development and subsequent evaluation of an interactive software design tool for the performance assessment of composite arrays, and in particular, their imaging potential for applications in underwater visualization systems, in which the array will be configured as part of a scanning system, is investigated. Firstly, finite element analysis (FEA) is used to evaluate the various factors which relate to the micro-structure of the composite material i.e., volume fraction, and the size and number of ceramic rods under the electrode. A linear systems approach is then adopted to investigate the macro-structure of the composite transducer, and considers the effects of material composition; ceramic-epoxy volume fraction and transducer backing impedance. Finally, the scattering responses from arbitrary target structures are considered, using practical array dimensions, as specified by the FEA and linear systems analysis. The model employed for target scattering is capable of simulating a wide range of composite configurations, and also of varying mechanical and electrical load conditions. In addition, it permits the analysis of realistic target scenarios, comprising an arbitrary arrangement of arcs, circles, and rectangles. A range of examples are presented, including both 1-D and 2-D transducer array configurations, and a selection of imaging results obtained via a 10 X 10 2-D array operating at 1.2 MHz. Good agreement between theory and experiment is obtained.

The problem of producing real-time, 3-D (volumetric) ultrasonic images can be divided into two main tasks: (1) acquiring 3-D echo data from a target volume in a sufficiently short time, and (2) reducing this 3-D data set into a suitable 2-D image. Both of these tasks can be accomplished simultaneously by a method, dubbed `Slit-O-Vision,' wherein a diverging lens is used to expand the (normally collimated) elevation beam pattern of a conventional linear or phased array transducer into a `fan beam.' The 2-D images produced in this manner make an object in the target volume appear three-dimensional, because the fan beam integrates the echoes at a given range across the entire `slice thickness' of the tomogram, thus projecting (collapsing) the volumetric echo data into a single 2-D projection image. This also allows the volumetric image to be updated 30 times per second. Slit-O-Vision may be useful in medical or sonar imaging applications.

Ultrasonic imaging arrays are severely restricted in design and imaging quality by artifacts associated with the geometry of the arrays. Linear and convex linear arrays are limited by the appearance of side lobes and grating lobes in the images, which obscure diagnostically useful information. Limited access to parts of the body useful for accurate diagnosis (e.g., intercostal imaging) in many cases forces the use of phased arrays or mechanical arrays in place of linear and curved linear arrays. One of the ways to avoid the limitations and utilize the advantage of the curved linear configuration is to use a concave design. This paper compares the results of simulations of the concave and convex designs, and the degree of reduction of grating lobes. The simulation results are verified with beam plots of each design. Concave arrays are fabricated and clinical results are shown for intercostal and subcostal abdomen imaging.

A set of sixteen rectilinear thermoelastic sources, equivalent to a phased array of ultrasonic transducers has been implemented on the surface of an isotropic solid from a multiple beam YAG laser. Introducing a time delay between each laser pulse, the elastic waves were focused in the sample. The ultrasonic beam is detected either by classical piezoelectric transducers or by a compact optical heterodyne probe which is sensitive to the normal displacement of the sample surface. Neglecting heat diffusion in the solid and considering the thermoelastic source as a surface center of expansion, the directivity patterns of laser generated longitudinal acoustic waves have been computed. Experiments performed on duraluminum samples in the thermoelastic regime are presented and compared with this analysis. It is shown that a high focusing and a very good sensitivity for longitudinal waves were achieved with this technique.

In this paper we describe and demonstrate a unique termination scheme whereby an array of coaxial wires is efficiently fabricated into a monolithic substrate resulting in a two-dimensional array structure forming a pad grid array connector. This structure has significant implications for high density cable interconnect applications in ultrasound transducer systems, circuit board and hybrid interconnect structures. A connector concept is proposed in which the interconnect is used as an in-line connector for a user-friendly transducer probe interconnect system. Future applications of the connector are described including use as an integrated substrate onto which both one and two-dimensional arrays can be fabricated or connected. Results including electrical testing are shown for a 12.7 mm (0.5') diameter, 100 position, 0.6 mm (.25') center-to-center rectilinear pad array connector utilizing 40 gauge 50 (Omega) coaxial wires.

Over the past eight years, 1 - 3 ceramic-polymer composites have come into widespread use as the piezoelectric material in many ultrasonic transducer arrays used for medical imaging. The success of these materials for this application is a result of three major advantages which the composites enjoy over piezoelectric ceramics alone. These are, increased thickness mode coupling, reduced acoustic impedance, and reduced lateral coupling. These three parameters can not all be optimized simultaneously. The engineering trade-offs in the composite design can be avoided if the polymer material is replaced with air. This paper discusses the nature of the trade-offs in the ceramic-polymer material and shows how the use of ceramic-air (air kerf) avoids these limitations. Requirements for good performance in ceramic-air composites are presented and experimental results for a transducer made from a ceramic-air composite are shown. The potential for improvements and the probable limitations for transducers made from air-ceramic composites are discussed.

Equivalent circuit descriptions suitable for describing the impulse response of transducer systems are considered, and the suitability of the Mason circuit is demonstrated. This description leads to consideration of multilayer structures that are applicable to medical ultrasound transducers. Of several interesting structures, one formed by folding thin plastic piezoelectric materials is shown to have the possibility of uncreased coupling coefficient and reduced resonant frequency compared to the basic film material.

The performance of a transducer possessing several piezoelectric layers is discussed. Techniques are presented for determining excitation functions so that a pre-defined transmission characteristic is obtained in an optimal manner. The performance of a multiple layer transducer in the reception mode is considered in detail. It is evident that a high degree of transmission and reception efficiency is attainable continuously from below the fundamental thickness mode resonance to above its third harmonic. This contrasts with conventional designs which possess a null at the second harmonic. Issues regarding the stability of the technique are addressed.

A Barker-coded multiple layer piezopolymer transducer design is described. The design is intended to improve the transmitting sensitivity of a PVDF transducer while retaining PVDF material's wide bandwidth properties. Computer simulations of the coded multilayered PVDF transducer were carried out and several prototypes of the transducers were built and tested. The results of comparisons between the measurements and theoretical predictions are presented. Advantages and disadvantages of the coded multilayer approach to bandwidth enhancement of ultrasound transducer are also discussed. Barker-coded transducers hold promise to emerge as a new class of ultrasound imaging transducers having a sensitivity on par with that offered by presently used PZT or composite transducers but with significantly wider bandwidth.

Efficient coupling of acoustic energy between different media requires fabrication of materials with precisely controlled acoustic properties. Current materials are formed by suspending particles in an epoxy matrix. Last year, we proposed a technique for making such materials using silicon micromachining. This paper describes the fabrication procedure and acoustic measurements of these materials and their applications. The acoustic properties of the composite materials described in this paper depend on the volume fraction of silicon, and the material used for the other phase of the composite. We investigated two different fillers, epoxy resin and RTV. Silicon structures with graded volume fractions were fabricated to provide broader bandwidth matching.

In this work we describe a method to measure the thickness of refractory lining of cigar shaped cars for the transportation of melted cast-iron, based on the pulse-echo technique at very low frequency. Following this approach two main problems arise related to the small dimensions of the transducers in respect to the ultrasonic wavelength: a poor directivity of the bulk wave and the presence of a strong surface wave. An improvement of the S/N ratio at the receiver input is obtained by using two unimodal transducers, one transmitter and one receiver, while the time discrimination between the bulk and the surface waves is obtained by taking advantage of the different wave velocities and by properly choosing the distance between the two transducers.

Polymer transducers with high center frequencies offer several potential advantages for ultrasonic imaging and tissue characterization of superficial tissue segments. The large bandwidths of these transducers permit resolution of small tissue structures and also provide detailed spectral data for characterizing stochastic tissue elements. We have integrated these transducers with digital systems and conducted initial examinations of the eye, skin, and in- vitro tissue specimens. Computed images have demonstrated superior resolution, and useful signal-to-noise ratios have been obtained for spectral bandwidths exceeding 35 MHz. Further investigations are required to develop compensating processing techniques for acoustic attenuation and frequency-dependent beam characteristics, which can be significant factors over these large bandwidths.

Detection of very small defects in solids needs the development of new focusing techniques. Focusing an ultrasonic wave on a defect of unknown shape through an interface of any geometrical shape is an important problem to be solved in acoustics, and is a challenge in nondestructive testing. In nondestructive testing, we have to detect metallurgical defects and the incident beam can be strongly distorted with the change of acoustical impedance created at the liquid-solid interfaces. The use of time reversal mirror (TRM) represents an original solution to this problem. They realize in real time a focusing process matched to the defect shape, to the propagation medium, and to the geometries of the mirror and of the medium. It is a self-adaptive technique which compensates for any geometrical distortions of the mirror structure as well as for distortions due to the propagation through interfaces between liquid and solid.

Conventional design for piezoelectric transducers tends to favor materials with high electromechanical coupling coefficient. However, when the converse piezoelectric effect is used in the design of transducers for maximum mechanical displacement, transducer performance measured by power conversion depends also on other factors such as the type of acoustic wave generated and material constants. An analysis of a piezoelectric transducer shows that maximum mechanical displacement is inversely proportional to the dynamic electromechanical coupling coefficient and the square of acoustic wave velocity. This implies that a slow shear wave in a material having low electromechanical coupling is preferred. Simulation results of different materials for transducer and substrate are presented.

Marine zooplankton form a significant part of the marine ecosystem since they are relatively low on the food chain and they exist in vast quantities. However, little is known of their behavior, how they feed, how they interact or their swimming patterns. To explore some of these issues a three-dimensional imaging sonar was developed to track the movements of these zooplankton in their native environment. The tracking problem requires a high frequency sonar with a fast frame update rate and reasonably high resolutions in three dimensions. It also requires a small array to minimize the proximity effects of the transducer package on the zooplankton behavior, to allow mounting of the transducer array on small remotely operated submersible vehicles and to reduce the cost of the sonar. This led to an array architecture which resolves the target volume of interest into a three-dimensional array of volumetric units that are digitized and stored in computer memory. This digitized array of numbers is then processed by the computer and the results displayed using a three-dimensional graphics imaging package to present a 3-D image depicting the back scatter strength of the target volume along with the location of any objects within that volume.