Design, Fabrication, and Measurement of High-Sensitivity Piezoelectric Microelectromechanical Systems Accelerometers

JOURNAL OF MICROELECTROMECHANICAL SYSTEMS, VOL. 12, NO. 4, AUGUST 2003 433 Design, Fabrication, and Measurement of High-Sensitivity Piezoelectric Microelectromechanical Systems Accelerometers Li-Peng Wang, Member, IEEE, Richard A. Wolf, Jr., Yu Wang, Ken K. Deng, Lichun Zou, Robert J. Davis, Member, IEEE, and Susan Trolier-McKinstry, Senior Member, IEEE Abstract Microelectromechanical systems (MEMS) accelerometers based on piezoelectric lead zirconate titanate (PZT) thick films with trampoline or annular diaphragm structures were designed, fabricated by bulk micromachining, and tested. The designs provide good sensitivity along one axis, with low transverse sensitivity and good temperature stability. The thick PZT films (1.5 7 m) were deposited from an acetylacetonate modified sol-gel solution, using multiple spin coating, pyrolysis, and crystallization steps. The resulting films show good dielectric and piezoelectric properties, with P r values 20 C cm 2, r 800, tan 3%, and e 31 values 6 5 C m 2. The proof mass fabrication, as well as the accelerometer beam definition step, was accomplished via deep reactive ion etching (DRIE) of the Si substrate. Measured sensitivities range from 0.77 to 7.6 pc/g for resonant frequencies ranging from 35.3 to 3.7 khz. These accelerometers are being incorporated into packages including application specific integration circuit (ASIC) electronics and an RF telemetry system to facilitate wireless monitoring of industrial equipment. [981] Index Terms Accelerometer, piezoelectric, piezoelectric lead zirconate titanate (PZT). I. INTRODUCTION ACCELEROMETERS have been used in many fields, including for activation of automotive safety systems (airbags, electronic suspension), for machine and vibration monitoring, and in biomedical applications for activity monitoring. Micromachined accelerometers are widely used by the automotive industry, because of their low cost, small size, and broad frequency response [1]. Three sensing mechanisms, piezoresistive, capacitive, and piezoelectric are most commonly utilized for MEMS accelerometers; each one has limitations and advantages. Compared to piezoresistive and capacitive accelerometers, there have been fewer reports of micromachined piezoelectric accelerometers [1], [2]. ZnO and PZT films are the two primary materials reported for use in bulk- or surface-micromachined piezoelectric MEMS accelerometers. Manuscript received January 6, 2003; revised February 7, 2003. This work was suppported by the National Institute of Standards and Technology under WR-ATP-0001, The Pennsylvania State University Nanofabrication Facility, and Cornell Nanofabrication Facility for use of the DRIE tool. Subject Editor N. F. de Rooij. L.-P. Wang, R. A. Wolf, Jr., Y. Wang, R. J. Davis, and S. Trolier-McKinstry are with The Pennsylvania State University, University Park, PA 16802 USA (e-mail: li-peng.wang@intel.com). K. K. Deng and L. Zou are with the Wilcoxon Research, Gaithersburg, MD 20878 USA. Digital Object Identifier 10.1109/JMEMS.2003.811749 Since the electromechanical coupling coefficients and the piezoelectric constants of PZT are much higher than those of ZnO films, the charge sensitivities of piezoelectric MEMS accelerometers using ZnO films are relatively small [3] [7]. Therefore, this work concentrated on the use of PZT films. Several groups have previously reported on the use of PZT MEMS accelerometers. In 1996, Nemirovsky et al. [8] designed a PZT thin-film piezoelectric accelerometer with a calculated sensitivity of 320 mv/g, however, it has not been fabricated. In 1997, Kim et al. fabricated a surface-micromachined PZT accelerometer using cantilever beams as the sensing structure [9]. No dynamic frequency response measurement was reported. In addition, surface micromachining limits the thickness of microstructures; as a result, the sensitivity is limited. In 1999, bulk-micromachined accelerometers were fabricated and tested by Eichner et al. [10]. A seismic mass and two silicon beams were used as the sensing structure; an average sensitivity of 0.1 mv/g was measured and the resonant frequency was calculated at 13 khz. Beeby et al. [11], [12] fabricated a bulk micromachined accelerometer using PZT thick films, which were prepared by screen-printing processes. Their reported sensitivity of 16 pc/g is puzzling, however, as described elsewhere [13]. For inertial sensors, the minimum detectable signal is limited by Brownian noise [14], which is described in the following: where is the minimum detectable signal when the signal-to-noise ratio (SNR) is unity; is Boltzmann s constant; is the absolute temperature; is the resonant frequency of the sensing structure; is the effective sensor mass; and is the quality factor. If the minimum detectable signal is 30 g at 100 Hz with a SNR of 4, the minimum is 0.3 g, which is not small for MEMS devices. Therefore, bulk micromachining is preferred in this work, since it can produce a larger mass than surface micromachining. Previously reported bulk micromachined accelerometers were largely fabricated by wet processing of the silicon using crystal-orientation-dependent anisotropic etching (KOH, EDP, or TMAH). However, this requires corner compensation, accurate alignment with crystal orientation to obtain the designed structures and high-quality hard masking materials (such as silicon oxide and silicon nitride). In addition, the wet processes are difficult to control industrially and have low compatibility with IC 1057-7157/03$17.00 2003 IEEE

434 JOURNAL OF MICROELECTROMECHANICAL SYSTEMS, VOL. 12, NO. 4, AUGUST 2003 (a) (b) Fig. 1. Schematics of two MEMS accelerometer structures: (a) a trampoline-type accelerometer has a proof mass suspended at the junction of crossed beams and (b) half of a diaphragm accelerometer shows a proof mass suspended at the center of an annular membrane. processing. There are two previous reports on the development of PZT-based accelerometers fabricated using deep reactive ion etching: one optimized for high sensitivities at low frequencies ( 300 Hz) [15]; the other designed for broader bandwidth operation [13]. This paper is an amplification of the report in reference [13]. II. DEVICE DESIGN AND FEA SIMULATION Two sensing structures were investigated, a trampoline-style sensor, in which the proof mass is suspended at the juncture of crossed beams, and an annular diaphragm sensing structure with an annular membrane with a suspended proof mass at the center of the diaphragm (see Fig. 1). The piezoelectric film is deposited on a bottom electrode so that it can be poled through its thickness. For both designs, the stresses are of opposite signs close to the frame and close to the proof mass. Thus, the top electrodes were separated in order to pole the two sections in opposite directions; the areas are adjusted to be identical. When the outputs of the electrodes are connected in parallel and films are poled in the opposite directions, the pyroelectric output of the sensor is cancelled. The top electrodes are placed only over highly stressed areas in order to better match the charge amplifier ASIC. The overall sensor size was 6 mm by 6 mm. Compared with the beam-type sensing structures, the annular diaphragm design [13] has several advantages. First, it uses area more efficiently, so it has a smaller die size with the same electrode area. Second, the sensor has a higher resonant frequency and wider bandwidth because the structure is stiffer. Finally, the sensor is insensitive to transverse acceleration because of the symmetrical structure. From a fabrication standpoint, the design does not require etching through the silicon to release the diaphragm, therefore, the fabrication process is simplified and yield is enhanced. Consequently, most of the results are reported for the annular diaphragm design. Both structures were modeled by finite element analysis (FEA) using ANSYS software. The stress distributions are shown in Fig. 2 and resonant frequencies can be obtained. Combining the constitutive equations of the piezoelectricity [16], the charge sensitivity can be obtained by integrating the piezoelectric charge over the top electrode areas.

WANG et al.: HIGH-SENSITIVITY PIEZOELECTRIC MICROELECTROMECHANICAL SYSTEMS ACCELEROMETERS 435 (a) (b) Fig. 2. Stress distribution of (a) trampoline structure and (b) diaphragm structures. III. FABRICATION Both trampoline and annular diaphragm accelerometers were designed on the same photomask set; in addition, a test pattern was included as well. Five photomask levels were used here. The fabrication process flow is shown in Fig. 3 The starting substrate was a 4-in diameter, (100) n-type (resistivity -cm), double-side-polished silicon wafer. 0.8 m thermal oxide was grown on the wafers and the bottom Ti Pt electrode (200 Ti, 1500 Pt) was then sputtered (NOVA Electronic Materials Inc.). Thick PZT films were deposited using a chemical solution deposition approach. In brief, films were prepared from 2-methoxyethanol based solutions using lead acetate trihydrate, zirconium n-propoxide, and titanium isopropoxide as precursors. 0.7 to 0.8 Molar solutions were prepared and modified with 22.5 vol% acetylacetonate (Aldrich Chemical Company). The solution was batched to produce the morphotropic phase boundary composition,, with 20 mole% excess lead. The sol was deposited by spin-coating at 1500 rpm for 30 s on a photoresist spinner (Headway Research, Inc., Garland, TX). After deposition, each layer was subjected to a two-stage pyrolysis sequence. A 1-min 300 C heat treatment was immediately followed by one at 450 500 C. The layer was then crystallized to a phase-pure perovskite at 700 C for 30 s in an RTA furnace (A. G. Associates, San Jose, CA). A multilayering scheme was employed to achieve the desired film thickness m. The resulting PZT films had a nearly random polycrystalline X-ray diffraction pattern, relative permittivities in the range of 800 to 1000, a, C cm, and an value of 6 to C m. Fig. 4 shows the piezoelectric properties of the films as a function of thickness and composition. All devices were made using the 52/48 Zr/Ti ratio in order to maximize the piezoelectric response. Following the PZT deposition, the top electrode of Cr (80 ) and Au (800 ) was deposited via electron-gun and thermal evaporation, respectively. Photoresist was then coated on both sides of the wafer, front-to-back side alignment was done using a mechanical jig and the wafer was exposed in a Karl Suss MA6 contact aligner. After development, the top electrode was wet-etched, and the backside oxide was reactive ion etched in plasma (power: 200 W; pressure: 40 mtorr). After stripping the photoresist on both sides, the backside of the wafer was spin-coated with photoresist to protect the already patterned silicon oxide during the subsequent PZT etching step. After another photolithography step, a two-step wet etch process was utilized to pattern the thick PZT layer [17]. The bottom electrode was patterned by RIE using plasma (power: 400 W; pressure: 6 mtorr). A subsequent oxygen plasma ashing step was used to remove the photoresist. At this point, the frontside processing was finished. Thick resist photolithography was utilized to define the die frames, diaphragms, and proof masses of the trampoline structures. A 13 m thick AZ4620 photoresist, which can be obtained at a spin speed of 1500 rpm, was used. Before the DRIE steps, the frontside of the wafer was spin coated with Shipley 1813 photoresist to protect the films during the oxide etching between the two DRIE steps. The first DRIE step was used to define the thickness of the beams and diaphragms. After defining the beam and diaphragm thickness, BOE was used to remove the patterned oxide on the beam areas, after which a second DRIE step was used to etch through the wafers (visual endpoint detection). The sidewall of the proof mass had an angle of (see Fig. 5). After stripping the remaining front-side oxide, and stripping the resist, the dies were separated. An example of a completed annular accelerometer is shown in Fig. 6. IV. MEASUREMENT Dynamic frequency response characterization was performed to evaluate the performance of the piezoelectric MEMS accelerometers. First, a MEMS accelerometer was glued to a ceramic substrate which had screen-printed silver pads for wire bonding and soldering. Then the MEMS accelerometer was wire-bonded to the pads. Finally, a coaxial wire was soldered to the ceramic substrate for the signal output. The output of the MEMS accelerometer was connected to a charge amplifier, which has a 10 pf feedback capacitor with an amplification of 10 mv/pc. A comparison measurement was performed using a reference accelerometer (Wilcoxon Research 05 238), which

436 JOURNAL OF MICROELECTROMECHANICAL SYSTEMS, VOL. 12, NO. 4, AUGUST 2003 Fig. 3. Process flow of PZT MEMS accelerometers. was also connected to a charge amplifier (Wilcoxon Research CC701). The outputs of both accelerometers were connected to a two-channel dynamic signal analyzer. A swept-sine signal, generated by the analyzer and enhanced by a power amplifier, was used to drive an electromagnetic shaker (Wilcoxon Research AV50) to mechanically excite both accelerometers with the same magnitude. Therefore, the frequency response was obtained by a transfer function measurement. One of the measured frequency responses of the MEMS accelerometers is shown in Fig. 7. Through this measurement, the charge sensi-

WANG et al.: HIGH-SENSITIVITY PIEZOELECTRIC MICROELECTROMECHANICAL SYSTEMS ACCELEROMETERS 437 Fig. 4. Piezoelectric properties of PZT films as functions of film thickness and the Zr Ti ratio. Fig. 7. Typical frequency response of a MEMS accelerometer reported here; the resonant frequency is 17.4 khz and the subresonance sensitivity is 1.4 pc/g. The curve shows a typical frequency response of a MEMS prototype accelerometer from 500 Hz to 25 khz. For frequencies lower then 500 Hz, a flat frequency response range (60 to 500 Hz) can be obtained with appropriate electronics. Fig. 5. Etched profile of the proof mass (diaphragm type) has an angle of 83. Fig. 6. A diaphragm-type accelerometer compared to a penny after finishing the processing steps. tivity and resonant frequency of the MEMS accelerometers can be calculated and observed. V. RESULTS AND DISCUSSION Sensitivities range from 0.77 to 7.6 pc/g with resonant frequencies ranging from 35.3 to 3.7 khz were measured (see Table I). These high sensitivities and broad usable frequency ranges were obtained because of the good sensor design and high quality of the piezoelectric films; values of C m were obtained on the same wafer for which most accelerometer measurements were made [18]. The sensor-to-sensor frequency and sensitivity variations were caused largely by variations of the silicon diaphragm thickness. This is because of nonuniformity in the Si DRIE etching. The differences between measured and calculated (by finite element analysis) sensitivities were 2 to 30%. The discrepancy between the measured sensitivity and the FEA calculations are believed to be due to the large tolerance m during the mechanical front and backside alignment, residual stress in the membrane, and piezoelectric film variations. The transverse sensitivity was found to be 2% of the sensitivity along the principal axis. In the measurement, another important aspect of the sensor design was also observed. The ratio between the measured sensitivity and the FEA calculations is plotted as a function of the silicon thickness in Fig. 8. The ratio is nearly constant ( 60%) down to a silicon thickness of m, and then starts to increase with decreasing silicon thickness. One possible origin for this behavior would be that the piezoelectric constant starts to increase with decreasing silicon thickness below this point. If so, then the piezoelectric constant at a silicon thickness of 6.1 m is about two times larger than the average value for the heavily clamped films. This difference is much larger than the point-to-point variations in the piezoelectric coefficient of the film ( 5%). If this is the critical factor, then it is likely that the increased response results from removal of the clamping of the PZT films from the silicon substrate [19]. Similar increases in the piezoelectric response have been reported when the in-plane constraints on ferroelectric thin films are relieved by laterally subdividing them [20], [21]. Another possibility is that the structure itself deforms due to stress relief, leading to the part behaving more like a shell than a diaphragm. Consequently, the stresses may be amplified [22]. Finally, there is also a possibility that the mode shapes mechanically excited during the testing were not perfect. Further study to clarify the origin of this behavior is necessary.

438 JOURNAL OF MICROELECTROMECHANICAL SYSTEMS, VOL. 12, NO. 4, AUGUST 2003 TABLE I MEASURED RESULTS IN COMPARISON WITH THE SIMULATED RESULTS Note: 1. The thickness of the Si diaphragm was calculated based on the measured resonance frequency. The FEA sensitivity was then recalculated based on this Si thickness. The thickness of PZT film is 5.6m 6 0.09m for all devices. 2. The product of d *Young s modulus for the PZT film used in the calculation was 045 C/m. yield is enhanced. Finally, it was demonstrated that the behavior of the accelerometers depends on the thickness of the underlying Si passive layer. Therefore, uniform bulk silicon etch (oran SOI process) is necessary to have more controllable accelerometer responses for future potential mass production. REFERENCES Fig. 8. The ratio of the measured sensitivities to the theoretical ones plotted as a function of the silicon thickness; the PZT thickness is 5.6 m. VI. ACCELEROMETER INTEGRATION Wilcoxon Research has developed a low power ASIC for amplifying the piezoelectric signal, filtering and processing the signal and performing analog to digital signal conversion. The MEMS accelerometer connected to the ASIC is managed by a digital signal processor (DSP) which also controls a Bluetooth rf system for wireless transmission of the accelerometer signal. Battery power management is achieved by a combination of the ASIC and the DSP. The MEMS contributes to both the minimization of the size of the instrument and to the low power requirements of the circuitry thus enabling the integration of a low-profile, high sensitivity, wireless accelerometer. VII. CONCLUSION The piezoelectric MEMS accelerometers, combining a novel annular diaphragm design and high electromechanical coupling thick PZT films, demonstrate high sensitivities (0.1 to 7.6 pc/g) and broad usable frequency ranges (44.3 to 3.7 khz). The accelerometer design provides good sensitivity along one axis, with low transverse sensitivity and good temperature stability. Furthermore, the diaphragm design does not require etching through the silicon to release the diaphragm (unlike beam-type structures); therefore, the fabrication process is simplified and [1] N. Yazdi, F. Ayazi, and K. Najafi, Micromachined inertial sensors, Proc. IEEE, vol. 86, pp. 1640 1659, Aug. 1998. [2] C. Song, B. Ha, and S. Lee, Micromachined inertial sensors, in Proc. 1999 IEEE/RSJ, International Conference on Intelligent Robots and Systems, vol. 2, pp. 1049 1056. [3] P. L. Chen, R. S. Muller, R. D. Jolly, G. L. Halac, R. D. White, A. P. Andrews, T. C. Lim, and M. E. Motamedi, Integrated silicon microbeam PI-FET accelerometer, IEEE Trans. Electronic Devices, vol. ED-29, pp. 27 33, 1982. [4] P. L. Chen and R. S. Muller, Integrated silicon Pi-FET accelerometer with proof mass, Sens. Actuators, vol. 5, pp. 119 126, 1984. [5] P. Scheeper, J. O. Gullov, and L. M. Kofoed, A piezoelectric triaxial accelerometer, J. Micromech. Microeng., vol. 6, pp. 131 133, 1996. [6] R. de Reus, J. O. Gullov, and P. R. Scheeper, Fabrication and characterization of a piezoelectric accelerometer, J. Micromech. Microeng., vol. 9, pp. 123 126, 1999. [7] D. L. DeVoe and A. P. Pisano, Surface micromachined piezoelectric accelerometers (PiXL s), J. Microelectromech. Syst., vol. 10, pp. 180 186, June 2001. [8] Y. Nemirosky, A. Nemirovsky, P. Maralt, and N. Setter, Design of a novel thin-film piezoelectric accelerometer, Sens. Actuators, vol. A56, pp. 239 249, 1996. [9] J. H. Kim, L. Wang, S. M. Zurn, L. Li, Y. S. Yoon, and D. L. Polla, Fabrication process of PZT piezoelectric cantilever unimorphs using surface micromachining, Integrated Ferroelectrics, vol. 15, pp. 325 332, 1997. [10] D. Eichner, M. Giousouf, and W. von Munch, Measurements on micromachined silicon accelerometers with piezoelectric sensor action, Sens. Actuators, vol. A76, pp. 247 252, 1999. [11] S. P. Beeby, N. Ross, and N. M. White, Thick film PZT/micromachined silicon accelerometer, IEEE Electron. Lett., vol. 35, no. 23, pp. 2060 2062, 1999. [12], Design and fabrication of a micromachined silicon accelerometer with thick-film PZT sensors, J. Micromech. Microeng., vol. 10, pp. 322 328, 2000. [13] L.-P. Wang, K. Deng, L. Zou, R. Wolf, R. J. Davis, and S. Trolier- McKinstry, Microelectromechanical Systems (MEMS) accelerometers with a novel sensing structure using piezoelectric Lead Zirconate Titanate (PZT) thick films, IEEE Electron Device Lett., vol. 23, no. 4, pp. 182 184, 2002. [14] T. B. Gabrielson, Fundamental noise limits for miniature acoustic and vibration sensors, J. Vibration Acoustics, vol. 117, pp. 405 410, Oct. 1995.

WANG et al.: HIGH-SENSITIVITY PIEZOELECTRIC MICROELECTROMECHANICAL SYSTEMS ACCELEROMETERS 439 [15] K. Kunz, P. Enoksson, and G. Stemme, Highly sensitive triaxial silicon accelerometer with integrated PZT thin film detectors, Sensors and Actuators, vol. A92, pp. 156 160, 2001. [16] IEEE Standard on Piezoelectricity, ANSI/IEEE 176-1987. [17] L.-P. Wang, R. Wolf, Q. Zhou, S. Trolier-McKinstry, and R. J. Davis, Wet-etch patterning of lead zirconate titanate (PZT) thick films for microelectromechanical systems (MEMS) applications, in Mat. Res. Soc. Symp. Proc., vol. 657, 2001, p. EE5.39. [18] J. F. Shepard, P. J. Moses, and S. Trolier-McKinstry, The wafer flexure technique for the determination of the transverse piezoelectric coefficient (d ) of PZT thin films, Sens. Actuators, vol. A71, pp. 133 138, 1998. [19] K. Lefki and G. J. M. Dormans, Measurement of piezoelectric coefficients of ferroelectric thin films, J. Appl. Phys., vol. 76, no. 3, p. 1764, 1994. [20] S. Buhlmann, B. Dwir, J. Baborowski, and P. Muralt, Size effect in mesoscopic epitaxial ferroelectric structures: Increase of the piezoelectric response with decreasing feature size, Appl. Phys. Lett., vol. 80, no. 17, pp. 3195 3197, 2002. [21] A. L. Roytburd, S. P. Alpay, V. Nagarajan, C. S. Ganpule, S. Aggrawal, E. D. Williams, and R. Ramesh, Measurement of internal stresses via the polarization in epitaxial ferroelectric films, Phys. Rev. Lett., vol. 85, no. 1, pp. 190 193, 2000. [22] P. Muralt, A. Kholkin, M. Kohli, and T. Maeder, Piezoelectric actuation of PZT thin-film diaphragms at static and resonant conditions, Sens. Actuators, vol. A53, pp. 398 404, 1996. Li-Peng Wang (M 01) received the B.S. degree in mechanical engineering from Chung-Yuan Christian University, Taiwan, in 1992. He received the M.S. degree in mechanical engineering in 1997 and the Ph.D. degree in engineering science in 2001, both at Pennsylvania State University, University Park. His Ph.D. research was focused on MEMS sensors based on piezoelectric lead zirconate titanate (PZT) thin films. In 2001, he joined the Wireless Microsystems Department, Intel Corporation and is responsible for material development, design, and process integration for RF MEMS applications. Richard A. Wolf, Jr. received the B.S. and M.S. degrees in materials science and engineering from the Pennsylvania State University, University Park, in 1999 and 2001, respectively. His thesis research focused on characterizing the temperature dependence of the piezoelectric response of lead zirconate titanate films for MEMS-based sensors and actuators. Ken K. Deng received the B.S. degree in mechanical engineering from Beijing University of Posts and Telecommunications (BUPT), China, in 1985 and M.S. degree in mechanical engineering from the University of Maryland at College Park (UMCP) in 1995. Currently, he is pursuing the Ph.D. degree at UMCP and his doctoral work focuses on the development of a novel MEMS environment-sensing suite. In 1996, he joined Wilcoxon Research, Inc., working in the field of piezoelectric transducers. He has been a leading MEMS designer in two MEMS projects entitled a development of a test and measurement grade MEMS piezoelectric accelerometers and developing a monolithic MEMS environmental sensing suite respectively. He has two pending patents in the field of bulk and MEMS accelerometer designs. Lichun Zou received the B.S. and M.S. degrees in physics from Northeast University, Shenyang, China, in 1982 and 1987, respectively. She has served as a faculty member at Northeast University, Boston, MA, for seven years. In 1990, she joined the Applied Solid State Group at Department of Physics, Queen s University, Canada, where her research focused on the development, design and microfabrication of integrated devices structure and MEMS for ferroelectric, piezoelectric and medical ultrasonic applications. She is currently working on the development novel MEMS based and relaxor single crystal based piezoelectric accelerometers, and characterization of new piezoelectric material for sensor application at Wilcoxon Research, Inc. Robert J. Davis (M 91) received the B.S. degree in engineering science from The Pennsylvania State University, University Park, in 1983 and the Ph.D. degree in applied physics from Cornell University, Ithaca, NY, in 1990. At the time of this research project, he was at the Pennsylvania State University serving as Associate Director of the Electronic Materials and Processing Research Laboratory, a Research Associate in the Applied Research Laboratory, and a member of the Graduate Faculty in Engineering Science and Mechanics. His research group there worked on several MEMS fabrication projects. He is the author or coauthor of over 20 journal articles in the areas of MEMS, dry etching and damage of semiconductors, and optoelectronic devices. He is currently the lithography group manager in Process Engineering at TriQuint Semiconductor Texas. His past affiliations include the IBM Thomas J. Watson Research Center, at which he was part of the Exploratory Devices and Technologies department; the Columbia University Department of Electrical Engineering; and the Max-Planck-Institute for Solid State Research in Stuttgart, Germany. Dr. Davis is a member of the American Physical Society and the American Vacuum Society. Yu Wang received the B.S. and M.S. degree in solidstate electronics from Huazhong University of Science and Technology (HUST), China, in 1985 and 1988, respectively. From 1989 to 1995, he was Research Associate in HUST, focusing on the plasma-related thin film process. He entered Pennsylvania State University, University Park, in 1995 and received Ph.D. degree in 2000. His dissertation focused on novel MEMS shear beam strain gauge and PZT accelerometer. Currently, he is Member of Technical Staff in Maxim Integrated Products, Inc., Sunnyvale, CA, and is interested in design and process of analog and mixed signal IC. Susan Trolier-McKinstry (M 92 SM 02) received B.S., M.S., and Ph.D. degrees in ceramic science from The Pennsylvania State University, University Park. In 1992, she joined the faculty of the Materials Science and Engineering Department of The Pennsylvania State University, where she is currently a Professor and Director of the W. M. Keck Smart Materials Integration Laboratory. She has held visiting appointments at the Army Research Laboratory and the Ecole Polytechnique Federale de Lausanne. Her research interests include thin films for dielectric and piezoelectric applications, MEMS, spectroscopic ellipsometry, and templated grain growth of electroceramics.