High Frequency Piezo Composites Microfabricated Ultrasound Transducers for Intravascular Imaging

Transcription

1 High Frequency Piezo Composites Microfabricated Ultrasound Transducers for Intravascular Imaging Jian. R. Yuan 1, X. Jiang 2, l Pei-Jie Cao 1, 1 Boston Scientific Imaging, Fremont, CA Alain Sadaka 1, Rick Bautista 1, K. Snook 2, P. W. Rehrig 2, 2 TRS Technologies, Inc., State College, PA Abstract - High frequency ultrasound has been widely used for various applications in ophthalmology, dermatology, small animal studies and intravascular (IVUS) imaging as diagnosis and imaging tool. The research and development of high frequency transducer has become one of the most interesting areas in ultrasound technologies. This article briefly reviews the existing technologies for high frequency ultrasound transducer development, then, presents the state-of-the-art piezo composite microfabricated ultrasound transducer (PC-MUT) for the first time. Using this technology, PMN-PT single crystal 1-3 composite has been developed, of which the kerf width is as small as to ~ 4 m. High frequency (> 40MHz) transducer with advanced performance has been developed for the IVUS application. The bandwidth and sensitivity are almost doubled compared with the conventional ceramic transducer. The features of PC-MUT technology may lead to great imaging quality and more potential medical applications. Keywords: PC-MUT, piezoelectric composite, PMN-PT, IVUS, microfabrication, ultrasound transducer. I. INTRODUCTION High frequency transducers can provide images of subsurface structures with microscopic resolution. This new technology is being widely used in ophthalmic eye imaging, dermatology skin imaging, catheter-based intravascular imaging, intra-articular imaging, high-frequency flow imaging, and in-vivo imaging of mouse embryonic development [1]. It is well known that high frequency IVUS clinical procedures can really benefit from the use of broad-bandwidth and highsensitivity transducers, particularly, because they provide a means of extending the depth of field over existing devices with high resolution. To strive for a better acoustic performance of the transducers in the high frequency range, a number of different materials and technologies have been investigated for constructing high frequency ultrasound transducers and transducer arrays. PVDF (polyvinylidene) and its copolymers P(VDF-TrFE) (poly(vinylidene fluoride-trifluoroethylene)) have made substantial contribution as transducer materials for medical and biological ultrasound imaging [2]. Low acoustic impedance (4 ~5 MRayl) and low mechanical Q value make these materials good choices for high-frequency design with very nice bandwidth. Compared with the piezoceramic, both PVDF and P(VDF TrFE) films are of low dielectric permittivity <10, and their electromechanical coupling coefficient is also low around 0.2 ~ 0.3, that result the transducers in a low sensitivity and not suitable for the application, where a large transducer size is not allowed. PZT films overcome the disadvantage of low dielectric permittivity of polymer films, and have been actively investigated for the use of piezoelectric transducers and surface acoustic wave (SAW) devices. The ease of composition control, high purity, and low processing temperature, make solgel method very commonly used to fabricate PZT film [3], [4]. The basic principle of sol-gel process is that a metal organic solution contains the components of the oxide compound in an organic solvent. This solution was coated onto various substrates by spinning, dipping or spray coating. Because of the low acoustic impedance (~20 MRayl), reasonable piezoelectric performance (k t ~ ) and controllable fabrication thickness (1-100 µm), sol-gel thin PZT film is suitable for high frequency transducer fabrication [5] - [8]. Though, the PZT films may bring dielectric constant up to around 600 by using PZT powder, it also raises its acoustic impedance; Meanwhile, due to the porosity of the PZT materials, the electromechanical coupling coefficient is k t ~ 0.3, which is not high enough to provide a high acoustic performance. While piezo polymer films and PZT films technologies provide the transducers for the high frequency medical ultrasound imaging, there are many efforts on piezocomposite materials to strive high acoustic performance in both bandwidth and sensitivity [9], [10]. This is because the needs of the nature of high frequency ultrasound. The acoustic attenuation in the wave medium becomes very high in the high frequency range that may limit the use of transducers built with film materials, of which the acoustic performances are not advanced enough. Especially, in the IVUS application, where the acoustic medium, blood, has high attenuation to ultrasound. For 2-2 composite, besides the most common dice-fill technique [11], the stacking and bonding of thin piezoceramic plates has been shown to be an effective method for creating high-frequency 2-2 composites [12]. Tape-cast PZT is a viable alternative to this stack and bond technique for 2-2 composites production [13]. Interdigital pair bonding and more recently interdigital phase bonding have been determined to be /06/$ IEEE IEEE Ultrasonics Symposium

2 modified dice-fill methods for manufacturing high-volume fraction 2-2 or 1-3 composites in the high frequency range [14]. As for 1-3 composite, dice-fill technique is still being reported in the relatively low frequency range. Lithographic Galvanforming and Abforming (LIGA) process has been successful in the frequency range around 20 MHz with good acoustic performance [15]. Another technique is termed the lost mold technique. It may provide PZT composite with the thickness between m and aspect ratio of the order of A 26 MHz transducer was reported showing a good frequency response [16]. Recently, a new micromolding technique for fabricating high-frequency (>20 MHz) ultrasound transducers has been reported [17]. The technique combines sol gel processing with an epoxy-based, photo-resist Su-8 micromold to form miniature PZT structures. Compared to LIGA processing, it avoids the intermediate step of producing a nickel-plated mold. Instead, the PZT is formed directly using a photo-resist. The resulting structures can be fabricated with aspect ratios up to 3:1 and thicknesses up to 50 µm. The average k t achieved by this method is around 0.3. Evaluated all current available technologies, the energetic attempt, either to raise the composite working frequency, or improve the inferior piezoelectric properties when compared to composites made from bulk piezoceramics is extremely desired in order to have images with high resolution that would facilitate lumen boundary detection and tissue characterization. The major application of ultrasound technologies at Boston Scientific Imaging is high-frequency intravascular ultrasound (IVUS) imaging, i.e. imaging the arteries to evaluate the presence of these plaques. Currently, existing IVUS devices include a single element rotating catheter transducer that is encased in a sheath and is used to image diseased vessels and also to guide certain interventional procedures, i.e. stent deployment. Single element, rotating catheter transducers are also used in some intracardiac echocardiography (ICE) imaging devices. These devices operate at relatively lower frequencies in order to increase penetration depth so that other structures such as heart chambers, valves, etc., can be imaged. This paper presents the piezoelectric composite microfabricated ultrasound transducer (PC-MUT) technologies, which was the first time using photolithography based deep reactive ion etching (RIE) micromachining process for the high frequency (> 40MHz) 1-3 piezo composite development. Repeatable and producible processes were fully developed for fabricating advanced 1-3 PMN-PT composite with a post width of 13 m and a kerf width of 3-4 m (volume fraction around 60%). The composites characterization showed effective electromechanical coupling coefficients of The bandwidth and sensitivity of both ICE (10 MHz) and IVUS (40 MHz) transducers using PC-MUT technology are almost double of the conventional ceramic transducer. II. METHODS AND MATERIALS processes can be described as the following. PMN-PT single crystal wafers were lapped on both sides and polished on one side. The polished side was then coated with Ni as an electroplating seed layer. A SPR 220 positive photoresist was later coated onto the PMN-PT single crystal wafers, followed by a soft baking. After the UV exposure, the CD-26 developer was used for photoresist developing to form a photoresist mold. A Ni plating process was next used to form a thicker Ni layer through the photoresist mold. The PMN-PT wafer with the Ni hard mask was then put into RIE etching chamber for Cl2 based deep etching. The kerfs of etched PMN-PT single crystal post arrays were next filled with epoxy to form 1-3 piezoelectric composite structure. The wafer was then lapped on one side until PMN- PT posts were exposed. The wafer was then flipped over for the second side lapping until the final thickness (~ µm) was achieved. The 1-3 single crystal/epoxy composite was then formed with solid volume fraction of about , both sides of the composites were next coated with Cr and Au as electrodes. The composites were first poled under 10 KV/cm at room temperature for about a minute and then ready for characterizations. A detailed report on PC-MUT high frequency single crystal/epoxy composites will be presented by Dr. X. Jiang from TRS Technologies in this symposium separately [18]. B. Pz Flex modeling of profile angle effects: Figure 1 (a) is a SEM photo picture, which shows the final etched PMN-PT pillars are vertically uniform with straight side wall > 85 as described in Fig. 1 (b). PzFlex model. was created to analyze the effect on electrical impedance with respect to the profile angle. Fig. 2 is the model which defines a single pillar of PZT with a polymer kerf and a water load. Side boundary conditions are considered as symmetrical. The ceramic/polymer part of the model is defined using the Build program for skewed co-ordinate systems. More details of the program can be found from Fig. 1(a), PMN-PT pillars after etching A. 1-3 PMN-PT High Frequency Composite: We developed repeatable and controllable processes for high frequency 1-3 PMN-PT composite, although there were many technical challenges and difficulties. The major IEEE Ultrasonics Symposium

3 Besides prototypes, we have fabricated composite with thickness from µm. We also built three batches of 6 x 6 mm PMN-PT 40 MHz composite to study the material processes and performance stabilities and consistency. The average piezo performances are listed in the table PMN-PT high frequency composite shows very nice performance as expected same to the general composite, i.e. high coupling coefficient (> 0.7), high dielectric constant. Table 1, Average performance over 15 pieces 1-3 PMN-PT composites Property Value Figure 1(b), Scheme of profile angle k t 71 ± 4% tan δ ± ε T /ε ± 280 ε S /ε ± 40 Velocity (m/s) 3530 ± 160 Density (kg/m 3 ) 5560 ± 270 Kerf (µm) 5.3 top /2.4 bottom (3.85 average) Fig. 2: PzFlex model The electrical impedance and phase simulation results are shown in Fig. 3. The red and black curves correspond to the composite with etched profile angle of 85 o and 90 o respectively. The FEM modeling indicates that if the kerf width is narrow enough and the etching angle is constrained in a certain range, there is no significant effect from the profile angle. Fig. 3, PzFlex simulation on electrical impedance and phase of sloped and non-sloped PZT pillars. Volume Fraction 65 ± 5% Thickness (µm) 39.5 ± 1.5 C. Transducer fabrication: A 6 x 6 mm 2 composite wafer was first attached to a clean glass substrate. A mold was glued to the glass enclosing the composite material. Filled the mold with a special epoxy based backing material and degas the backing to remove any remained bubbles. After cured, remove the composite and backing stack from mold. A special process to control the thickness of matching layer material was developed, which may well control the thickness of matching layer deposit within 1 µm. Now, the whole acoustic stack of composite, backing and matching layer can be machined into required small size of IVUS transducers. The transducer, then, was built into an imaging core. The imaging core is the first step to build a full imaging catheter. It consists of a transducer housing, and a 1.5m long drive cable, within which a coax cable is used for electricity path. D. Pulse echo study: To compare the acoustic performance, 40 MHz transducer built with conventional ceramic material PZT-5H and 1-3 PMN-PT composite were tested under the same condition: the Panametric 5900PR was used as the pulser and receiver. The gain was set to 0dB. Energy was set to 1j. Pulse repeat frequency was 1kHz. Damp impedance was 50ohm. The band pass filter was from 1MHz to 150MHz. An acrylic flat surface IEEE Ultrasonics Symposium

4 was used as the target located at a distance that assures echo to arrive at 5 s from pulse trigger time. The ultrasound beam and target surface are adjusted to perpendicular to each other so that the echo amplitude was maximal. Room temperature distilled water was used for the experiments. As shown in Fig. 4, the controlled transducer has a sensitivity of 153 mv, its - 6dB fractional bandwidth is 43 % and the center frequency is 40.0 MHz. The pulse echo test for 1-3 PMN-PT composite transducer is shown in Fig. 5. It shows the composite transducer has a sensitivity of 191mv, 77 % -6dB bandwidth at center frequency of 40.9MHz. The spectrum has a smooth round shape. E. Phantom experiment set up: A Doppler flow phantom (ATS Laboratories Inc, Bridgeport, CT) was used in the experiment. The phantom has four artery mimicking flow channel with diameters from 2 to 8 mm. A stent was deployed inside the 2mm flow channel to mimic artery. The material for the flow channel is urethane rubber. The flow channel has a length about 20cm. A peristaltic pump with adjustable flow rate kept the fluid in circulation. The function of tank 1 was to get rid of bubble and maintain the temperature around 37 o C. The fluid in the circulation is Model 707 Doppler Test Fluid from ATS Laboratories. It is a reliable, stable, non-hazardous fluid formulated to mimic the acoustic and physical properties of human blood. A bifurcation was created to insert an imaging catheter and a thermometer to monitor the temperature. Tank1 Peristaltic pump Heat & stir Stent Thermometer Tube Tank2 Imaging Catheter Fig. 6, Artery phantom set up Fig. 4, Pulse echo of the controlled transducer Two imaging catheters are identical except the transducers mounted on the tip. A standard 40MHz PZT transducer was used for comparison. The imaging system is an ilab system manufactured by Boston Scientific. The pictures were taken when transducer was inside and outside the stent respectively.. Fig. 5, Pulse echo of the 1-3 PMN-PT composite transducer Fig. 7, Phantom image inside stent. (Left: controlled transducer; right: 1-3 composite transducer) IEEE Ultrasonics Symposium

5 When the transducer was inside stent, the phantom images are shown in Fig. 7. The left panel is from the controlled transducer while the right panel is from the composite transducer. By designing the desired pattern of electrode on the PC- MUT, this technology can be also used for the high frequency array development with advanced acoustic performance. REFERENCES Fig. 8, Phantom image outside stent. (Left: controlled transducer; right: 1-3 composite transducer) When the transducer was outside stent, the phantom images are shown in Fig. 8. The left panel is from the controlled transducer while the right panel is from the composite transducer. Two phenomena can be noticed from the phantom images. The speckle size in the right panel image of the composite is smaller than that of the left panel of conventional ceramic. This is due to the composite transducer has a broader bandwidth and short pulse length; The second is that the image in right panel is brighter than the left panel. It should be attributed to the higher sensitivity of the composite transducer. In the composite image, the outer surface of the phantom is pretty clear while it can be barely seen from image of ceramic transducer. More careful comparison will be given by animal study shortly. III. DISCUSSION AND CONCLUSIONS: A novel piezo composites microfabricated ultrasound transducers technology has been developed and presented. Either ceramic or single crystal materials can be used with this technology. The processes of this technology are repeatable and controllable. 40 MHz PMN-PT PC-MUT has been developed for IVUS application, which almost doubles acoustic performance both in bandwidth and sensitivity. Compared with all of current available techniques, PC-MUT PMN-PT transducer shows super performance and no needs of DC bias. Due to the nature of PMN-PT single crystal material, there would be no particle size issue. Hence, single crystal PC-MUT can be built at further high frequency range with advanced performance. [1] G. R. Lockwood, D. H. Turnbull, D. A. Christopher, and F. S. Foster, Beyond 30 MHz: Applications of high frequency ultrasonic imaging, IEEE Eng. Med. Biol., vol. 15, pp , [2] Foster, F. S., K. A. Harasiewicz, M. D. Sherar (2000). A history of medical and biological imaging with polyvinylidene fluoride (PVDF) transducers, IEEE Trans. Ultrason., Ferroelect., Freq. Contr. 47(6): [3] M. Sayer, M. Lukacs, and T. Olding, Emerging technologies for ferroelectric films and coatings, Integr. Ferroelect., vol. 17, no. 1, pp. 1-10, [4] M. Sayer, Piezoelectric films and coatings, Can. Ceram. Quart. J. Can. Cerm. Soc., vol. 65, pp , Aug [5] M. Sayer, M. Lukacs and F. S. Foster, Single elementand linear array PZT ultrasound biomicroscopy transducers, Ultrasonics Symposium, 1997 IEEE [6] Gentry, K.L.; Zara, J.M.; Sang-Don Bu; Chang-Beom Eom; Smith, S.W. Thick film sol gel PZT transducer using dip coating, Ultrasonics Symposium, 2000 IEEE Int. vol. 2, pp vol.2 [7] Lukacs, M., M. Sayer, F. S. Foster (2000). Single element high frequency (< 50 MHz) PZT sol gel composite ultrasound transducers. IEEE Trans. Ultrason., Ferroelect., Freq. Contr. 47(1): [8] Q. F. Zhou, K. K. Shung and Y. Huang, Fabrication of sol-gel modified piezoelectric thick films for high frequency ultrasonic applications, Ultrasonics Symposium, 2004 IEEE, vol. 3, pp [9] Wallace A. Smith, and Bertram A. Auld, Modeling 1-3 Composite Piezoelectrics: [9] Thickness-Mode Oscillations, IEEE Trans. Ultrason., Ferroelect., Freq. Contr. 38(1): [10] Moojoon Kim, Jungsoon Kim and Wenwu Cao, Electromechanical coupling coefficient of an ultrasonic array element, J. of Appl. Phys. 99, (2006) [11] J. Cannata, et al, Development of a high frequency (35 MHz) linear ultrasonic array using 2-2 composite elements Proceedings of 2004 IEEE Ultrasonics Symposium, pp [12] Ritter T. A., Shrout T. R., Tutwiler R., and Shung K. K., A 30-MHz piezo-composite ultrasound array for medical imaging applications, IEEE Trans. Ultrason., Ferroelect., Freq. Contr., vol. 49, no. 2, pp , [13] [13] W. Hackenberger, et al., 2-2 PZT-polymer composites for high frequency (>20 MHz) ultrasound transducers Proceedings of 2002 IEEE Ultrasonics Symposium, pp [14] R. Liu, K. A. Harasiewicz, and F. S. Foster, Interdigital pair bonding for high frequency ultrasonic composite transducers, IEEE Trans. UFFC, vol. 48, pp , [15] Y. Hirata et. al, Piezo Composite of Fine PZT Rods Realized by LIGA Process, Sensor and Materials, vol. 16, No. 4 (2004), pp [16] S. Cochran, A. Abrar, K. J. Kirk, Dou Zhang, T. W Button, Bo Su; and C. Meggs, Net-shape ceramic processing as a route to ultrafine scale 1-3 connectivity piezoelectric ceramic-polymer composite transducers, Ultrasonics Symposium, 2004 IEEE, vol. 3, pp Vol.3 [17] Guofeng Pang, Micheal Sayer and Geoffrey R. Lockwood, Fabrication of PZT Sol Gel Composite Ultrasonic Transducers Using Batch Fabrication Micromolding, IEEE Trans. Ultrason., Ferroelect., Freq. Contr. 53(9): September, 2006, [18] Xiaoning Jiang, Jian R. Yuan et. Al, Microfabrication of piezoelectric composite ultrasound transducer (PC-MUT), Ultrasonics Symposium, 2006 IEEE IEEE Ultrasonics Symposium