Meetings & Events

Publishing Alliance

2009 MRS Fall Meeting & Exhibit

November 30 - December 4, 2009 | Boston
Meeting Chairs: Kristi Anseth, Li-Chyong Chen, Peter Gumbsch, Ji-Cheng Zhao

Symposium DD : Microelectromechanical Systems--Materials and Devices III

2009-11-30 Show All Abstracts

Symposium Organizers

Joerg Bagdahn Fraunhofer Center for Silicon Photovoltaics

Norman Sheppard MicroCHIPS, Inc.

Srikar Vengallatore McGill University

Kevin Turner University of Wisconsin-Madison

DD1: Resonators and High Frequency Devices

Session Chairs

Joerg Bagdahn

Monday PM, November 30, 2009

Fairfax (Sheraton)

2:30 PM - **DD1.1

Silicon micromechanical resonators for precision sensing

Ashwin Seshia ¹
¹Engineering, University of Cambridge, Cambridge United Kingdom

Show Abstract

Micromachined silicon resonators can be adapted as highly sensitive monitors of their physical and chemical environment. Non-dissipative physical processes resulting in structural and material perturbations in the resonator can be monitored by measuring induced resonant frequency shifts. Dissipative processes coupling to resonator dynamics can be monitored through changes in the quality factor. Accurate models for resonator dynamics, their interaction with the environment and integration into more complex sensor and signal processing systems are essential to realize applications in precision sensing. A number of underlying research threads are discussed here. First, a description of quality factor enhancement in micromachined silicon resonators through mode shape selection and resonator design is presented. Micromachined silicon resonators with quality factors of over 1 million at room temperature under moderate vacuum levels and associated frequency-quality factor products in excess of 10^13, close to the limits defined by intrinsic thermal loss, have been demonstrated. The quality factor of these resonators are increased by limiting vibrational energy loss to the substrate and substantially reducing thermoelastic dissipation through mode shape selection. Second, the role of electromechanical transducers and electrical interfaces in resonator design will be presented. Electrical transduction methods are described that utilize non-linear electrostatic actuators and readout mechanisms that leverage the piezoresistive properties of silicon to allow enhanced all-electrical probing of micromechanical vibratory response. Additionally, by incorporating actuators internal to the micromechanical structure, further enhancement in coupling the mechanical and electrical domains is attained, particularly at high frequencies. Thirdly, by constructing coupled resonator arrays, parametric sensitivities orders of magnitude higher than that for a single degree of freedom resonator are obtainable. This increase in parametric sensitivity can be accompanied by an intrinsic rejection of common mode environmental drift by leveraging the principle of vibration localization. The underlying research will be highlighted by application case studies. Oscillators incorporating CMOS-compatible high Q silicon square-extensional and wine-glass mode micromachined resonators exhibit short-term frequency stabilities of better than 1 part per billion. These oscillators can be viewed as building blocks for sensors and related systems. Two particular resonant sensor case studies are discussed viz. (1) thin film monitors demonstrating sensitivities of 136 Hz/nm and a mass flux resolved noise floor of under 17 pg/cm^2 and (2) tuning fork strain sensors with a resolution of close to 10 pε in a 1 Hz bandwidth. These sensors form essential components of systems currently under development for environmental and infrastructure monitoring.

3:00 PM - DD1.2

MEMS Microresonators for High Temperature Sensor Applications.

Dharanipal Doppalapudi ¹, Jeremie LeClair ¹, Richard Mlcak ¹, Harry Tuller ^{1 2}
¹, Boston MicroSystems Inc., Woburn, Massachusetts, United States, ²Department of Materials Science and Engineering, MIT, Cambridge, Massachusetts, United States

Show Abstract

3:15 PM - DD1.3

A Study of Dissipation in Nanocrystalline Diamond MHz Frequency MEMS and NEMS.

Matthias Imboden ¹, Pritiraj Mohanty ¹
¹Physics, Boston University, Boston, Massachusetts, United States

Show Abstract

Enabled by recent technological advances, it is now possible to create a variety of nanomechanical oscillators with resonances at microwave frequencies. Such nano- and micro-electromechanical systems (NEMS and MEMS) have become established devices as both commercial products as well as tools for basic research. Diamond, well known for its extraordinary mechanical properties, such as being the hardest known natural material, is an ideal material for NEMS and MEMS applications. Using CVD growth processes, diamond thinfilms can be grown controllably to engineer novel NEMS devices. Here, MEMS and NEMS performance is discussed in doubly clamped structures with emphasis on dissipation at both room temperature and sub Kelvin temperatures.We present frequency and dissipation scaling laws for doubly clamped diamond resonators [1]. The device lengths range from 10 to 19 microns corresponding to frequency and quality-factor ranges of 17 to 66 MHz and 600–2400, respectively. The authors find that the resonance frequency scales as 1/L² confirming the validity of the thin-beam approximation. The dominant dissipation comes from two sources: for the shorter beams, clamping loss is the dominant dissipation mechanism, while for the longer beams, surface losses provide the most significant source of dissipation.We also present Kelvin to millikelvin-temperature measurements of dissipation and frequency shift in megahertz-range resonators fabricated from ultra-nanocrystalline diamond [2]. Frequency shift δf/f0 and dissipation Q-1 demonstrate temperature dependence in the millikelvin range similar to that predicted by the glass model of tunneling two level systems. The logarithmic temperature dependence δf/f0 is in good agreement with such models, which include phonon relaxation and phonon resonant absorption. Dissipation shows a weak power law, Q^-1 ~ T^1/3, followed by saturation at low temperature. A comparison of both the scaled frequency shift and dissipation in equivalent micromechanical structures made of single-crystal silicon and gallium arsenide indicates universality in the dynamical response.[1] M. Imboden, A, Gaidarzhy, P. Mohanty, J. Rankin, and B.W. Sheldon, Scaling of dissipation in megahertz-range micromechanical diamond oscillators, Appl. Phys. Lett., 90, 173502 (2007)[2] M. Imboden, P. Mohanty, Evidence of universality in the dynamical response of micromechanical diamond resonators at millikelvin temperatures, Phys. Rev. B 79, 125424 (2009)

3:30 PM - DD1.4

Thermo-mechanical Stability of Ultrananocrystalline Diamond Resonators.

Vivekananda Adiga ¹, Suresh Sampath ², Arindom Datta ², John Carlisle ³, Robert Carpick ^{4 1}
¹Materials Science and Engineering, University of Pennsylvania, Philadelphia, Pennsylvania, United States, ², Innovative Micro Technology, Santa Barbara, California, United States, ³, Advanced Diamond Technologies, Romeoville, Illinois, United States, ⁴Mechanical Engineering, University of Pennsylvania, Philadelphia, Pennsylvania, United States

Show Abstract

Tetragonal sp³-bonded diamond has the highest known atomic density. The nature of the bond and its high density enable diamond to have superior physical properties such as the highest Young’s modulus and acoustic velocity of all materials, and excellent tribological properties. These make diamond promising for high frequency micro/nanomechanical devices. Recently, conformal thin diamond films have been grown at CMOS-compatible temperatures in the form known as ultrananocrystalline diamond (UNCD). We have measured the Young’s modulus (E), Poisson’s ratio and the quality factors (Q) for microfabricated overhanging ledges and fixed-free beams composed of UNCD films grown at lower temperatures.¹ The overhanging ledges exhibited periodic undulations due to residual stress. This was used to determine a biaxial modulus of 838 ± 2 GPa. Resonant excitation and ring down measurements of the cantilevers were conducted under ultra high vacuum (UHV) conditions on a customized atomic force microscope to determine E and Q. At room temperature we found E = 790 ± 30 GPa, which is ~20 % lower than the theoretically predicted value of polycrystalline diamond, an effect attributable to the high density of grain boundaries in UNCD. From these measurements, Poisson’s ratio for UNCD is estimated for the first time to be 0.057 ± 0.038. We discuss how adjusting for overhang is critical for these measurements. We also measured the temperature dependence of E and Q in these cantilever beams from 30 K to 300 K. Mechanical stiffness of these cantilevers followed Watchman’s empirical relation whereby E increased linearly with the reduction in temperature until 150 K where it then saturates. This is the first such measurement for UNCD and strongly suggests that the nanostructure plays a significant role in modifying the thermo-mechanical response of the material. The Q varied from 5000 to 16000 and showed a moderate increase as the cantilevers were cooled below room temperature. The results suggest that defects in the grain boundaries significantly contribute to the observed dissipation in UNCD resonators as opposed to thermoelastic dissipation and dissipation due to coupling between thermal and acoustic phonons. Influence of the overhanging ledges at the cantilever bases on temperature dependent resonant frequency shifts and quality factors will also be discussed.1. “Mechanical stiffness and dissipation in ultrananocrystalline diamond resonators” Adiga V.P., A. V. Sumant, S. Suresh, C. Gudeman, O. Auciello, J. A. Carlisle, R.W. Carpick, Vol. 79, pp 245403.1-8, Physical Review B, 2009

3:45 PM - *

Break

4:15 PM - **DD1.5

ZnO Thin Film Surface Acoustic Wave based Lab-on-a-Chip.

Jack Luo ¹, Yongqing Fu ², Xiaoye Du ³, Sunglyul Maeng ⁴, Daesik Lee ⁴, Andrew Flewitt ³, William Milne ³
¹Centre for Material Research & Innovation, University of Bolton, Bolton United Kingdom, ²Department of Mechanical Engineering, School of Engineering and Physical Sciences,, Heriot Watt University,, Edinburgh United Kingdom, ³Department of Engineering, University of Cambridge, Cambridge United Kingdom, ⁴, Electronics and Telecommunications Research institute, Daejeon Korea (the Republic of)

Show Abstract

Lab-on-a-chip (LOC) is one of the most important microsystems with promising applications in microanalysis, drug development, diagnosis of illness and diseases etc. LOC typically consists of two main components: microfluidics and sensors. Integration of microfluidics and sensors on a single chip can greatly enhance the efficiency of biochemical reactions and the sensitivity of detection, increase the reaction/detection speed, and reduce the potential cross-contamination, fabrication time and cost etc. However, generally the mechanisms used for microfluidics and sensors are different, making the integration of the two main components complicated and increasing the cost of the systems. We have developed a new type of micropump and mixer using surface acoustic wave (SAW) as an actuation mechanism using low-cost ZnO piezoelectric films on Si-substrates, with a great potential for microfluidic applications. The SAW-based micropump and micromixer are simple in structure and fabrication, are low cost and can act as active pumping and mixing devices without moving-parts, hence are reliable and effective. SAW devices were fabricated on nanocrystalline ZnO thin films deposited on Si substrates using sputtering. When an AC signal at the intrinsic resonant frequency is applied to the interdigited transducer, acoustic waves are generated through the piezoelectric effect and travel on the surface. Coupling of acoustic waves into a liquid induces acoustic streaming and motion of a droplet if it is on a hydrophobic surface. A streaming velocity up to ~5cm/s was obtained. When the surface energy of the SAW device is reduced by using a self-assembly monolayer, the acoustic wave can be effectively used to pump droplets up to a speed of ~1cm/s. It was also found that a higher order mode wave, the Sezawa wave is more effective in streaming and transportation of microdroplets. A novel SAW device on a ZnO island was also fabricated which can be used to pump and mix liquids remotely, thus avoiding direct contact of the very-reactive ZnO film with biochemical solutions. The SAW devices have also been utilized for biosensing. The well defined resonant frequency of the SAW devices decreases once additional mass is attached to the surface, and the sensitivity of the acoustic devices is proportional to the operating frequency. The high frequency Sezawa wave of the ZnO SAW devices has been utilized to detect reactions between pro-state antigens and its antibodies. The shift of the resonant frequency of the SAW device was found to increase as the concentration of the pro-state antigens increases, verifying its potential for early stage pro-state cancer detection. These results clearly demonstrated the feasibility of using a single actuation mechanism for both microfluidics and sensing, and hence show its suitability for the LOC. This greatly simplifies the fabrication and operation of the microsystems, and enhances the sensitivity and performance of the microsystems.

4:45 PM - DD1.6

A New Approach for Accurate Measurement of Internal Friction in Thin Films: Applications to Microresonators used in MEMS.

Sairam Prabhakar ¹, Guruprasad Sosale ¹, Srikar Vengallatore ¹
¹Mechanical Engineering, McGill University, Montreal, Quebec, Canada

Show Abstract

Accurate measurement of internal friction is essential for fundamental studies of the mechanical behavior of thin films, and for the design of high-Q microresonators employed in MEMS for applications in sensing and communications. Typically, these measurements are performed by depositing the thin film on a relatively thick substrate, and then measuring the structural damping of the composite film/substrate system in flexural or torsional vibration under reduced pressure. However, the measured damping is due to losses from several sources, including anchor losses at the supports, thermoelastic damping in the composite beam, and internal friction in the thin film. Identifying the magnitude of internal friction from the measured structural damping has been a long-standing problem in this field. Here, we propose a solution to this problem by taking advantage of recent developments in our understanding of thermoelastic damping (TED). This mechanism provides an absolute, thermodynamically-mandated, lower-bound on structural damping in flexural mode resonators. Furthermore, microscale resonators fabricated using single-crystal silicon can be designed to operate at the thermoelastic limit, which ensures that all other losses (including viscous air damping and clamping losses) are demonstrably negligible in these monolithic devices. Now, if these resonators are coated with a thin film, then the measured damping contains only two components: internal friction in the thin film, and thermoelastic damping in the film/substrate composite. The latter is calculated using an exact theory that does not require any free parameters [1], and then subtracted from the measured damping to yield an accurate estimate of the internal friction in the film.As an illustration of this approach, we present data for the internal friction in Al films with thickness ranging from 75 to 500 nm. First, a set of single-crystal silicon beams with thickness ranging from 100 to 150 μm, and fundamental natural frequency ranging from 0.1 to 1.0 kHz, were fabricated. Measurements at room temperature and low pressure showed that these devices operate at the thermoelastic limit with 10^-6 < Q^-1 < 5x10^-5. Thin films of Al were then sputter deposited on these silicon beams, and the internal friction extracted from the measured structural damping. The effects of thickness and frequency on internal friction in these films, and implications for the selection of metallic coatings in the design of layered microresonators, will be discussed.[1] S. Prabhakar and S. Vengallatore, “Thermoelastic damping in bilayered micromechanical beam resonators,” J. Micromechanics Microengineering, vol. 17, pp. 532-538 (2007)

5:00 PM - DD1.7

Micromachined Active Piezoelectric Structures for Applications above 600 °C.

Jan Sauerwald ¹, Denny Richter ¹, Erik Ansorge ², Bertram Schmidt ², Holger Fritze ¹
¹Department of Physics, Metallurgy and Materials Science, Technische Universitaet Clausthal, Goslar Germany, ²Institute of Micro and Sensor Systems, Otto-von-Guericke-Universitaet Magdeburg, Magdeburg Germany

Show Abstract

The operation of active devices in harsh environments requires appropriate materials and adapted operation principles. For example, high-temperature transducers based on single crystalline piezoelectric langasite (La3Ga5SiO14) meet both requirements. Bulk acoustic wave (BAW) langasite resonators coated with gas sensitive films could be operated up to at least 1000 °C and are demonstrated to be selective in-situ gas sensors. The materials properties and the application limits of langasite are well known. Further, large size and high quality wafers are available. Those capabilities motivate the creation of micromachined active langasite components. The specific objective of this work is to prepare thin membranes for BAW gas sensors and field emission diodes.The fundamental concept includes monolithic structures in order to minimize thermal stress. Therefore, conductive structures are prepared by doping. Heavy Sr doping increases the conductivity of langasite by about four orders of magnitude. Locally doped areas serve as monolithic electrodes as demonstrated by the excitation of resonators up to about 800 °C. In order to control the preparation process the chemical diffusion coefficients of potential dopants like Pr, Sr and Nb are determined.Initially, the thermal stability of miniaturized structures is tested since short diffusion distances and large surface to volume ratios might result in additional degradation processes in comparison to bulky structures. Test structures like tips and edges are prepared and annealed for 48 h at temperatures up to 1300 °C. SEM photographs do not indicate any rounding or other types of degradation. Therefore, langasite is expected to show dimensional long-term stability at typical operation temperatures up to 1000 °C. BAW langasite resonators are prepared by wet chemical etching in form of thin membranes. Resonance frequencies of up to 60 MHz and, thereby, increased mass sensitivities are achieved. Nearly isotropic etching results in slightly concave surfaces leading to moderate resonator quality factors (Q factor). Convex surfaces prepared by multi-step processes improve the energy trapping of the resonators. Biconvex membranes show a clear improvement of the Q factor in comparison to nominally planar resonators. At 700 °C, 16 MHz biconvex and planar resonators show Q factors of 820 and 400, respectively. In addition, arrays of membranes are machined. Further, field emission diodes are prepared and demonstrated to be operational. The radius of the tips is estimated to be as low as 40 nm. Those diodes and other active electronic elements are intended to process e.g. sensor signals already close to the sensor element.Finally, application examples demonstrate the capabilities of langasite membranes. For example, simultaneous determination of mechanical and electrical properties of thin sensor films by arrays of resonant sensors enables detection of CO in hydrogen containing atmospheres.

5:15 PM - DD1.8

Fatigue Properties of Silicon Thin Films in Harsh Environments.

Michael Budnitzki ¹, Pierre-Olivier Theillet ¹, Eva Baumert ¹, Olivier Pierron ¹
¹Mechanical Engineering, Georgia Institute of Technology, Atlanta, Georgia, United States

Show Abstract

MEMS devices have promising applications in harsh environments. For devices that cannot be hermetically packaged, the reliability of the structural materials, such as silicon (Si) thin films, must be assessed in these harsh environments. This study investigated the fatigue degradation properties of mono- and poly-crystalline silicon thin films as a function of temperature (from 30 to 80 °C), relative humidity (from 30 to 90% RH), testing frequency (static test, 4 vs. 40 kHz), and nanometer-scale surface coating. 2-micron-thick polycrystalline Si notched cantilever beam structures subjected to fully reversed sinusoidal loading at resonance (stress amplitudes between ~1.45 and 1.6GPa, ~40kHz) were used to investigate fatigue degradation in a high-temperature (80°C), high-humidity (90%RH) environment. Comparison to a reference set of S-N data obtained at moderate conditions (30°C and 50%RH) reveals a strong tendency for faster degradation in harsh environment, with damage rates exceeding the reference by two orders of magnitude. Transmission electron microscopy (TEM) on vertical through-thickness slices shows highly localized oxide thickening (up to 50nm) after cycling. Protection from the environment by means of ~20nm alumina, deposited on the specimen’s surface using atomic layer deposition (ALD) technique, results in a drastically different frequency evolution behavior at 30°C and 50%RH. No oxide thickening was observed in TEM for coated run-out specimens. A model is proposed to explain the different degradation behavior of the ALD-alumina coated samples. Thickened oxides after cycling appear consistent with the reaction-layer fatigue mechanism, even though the critical processes such as room-temperature, stress-assisted oxidation remain elusive.The low-cycle and high-cycle fatigue properties of 10-micron-thick, single-crystal Si films were also investigated using kHz-frequency resonating structures. The influence of resonant frequency (4 vs. 40 kHz) and environment (from 30°C/50%RH to 80°C/90%RH) on the resulting S-N curves and fatigue rates was monitored. Fatigue rates were assessed based on the resonant frequency decrease rate (0.01Hz resolution in resonant frequency measurement). The fatigue rates are most sensitive to stress, with 4 orders of magnitude decrease from 3.2 to 1.5-2 GPa. The fatigue rates are also much more sensitive to relative humidity than temperature or partial pressure of water, indicating the effective environmental parameter is the adsorbed water layer. Last but not least, unambiguous frequency effects are reported. The underlying mechanisms for the fatigue behavior of Si films are discussed in light of these experimental data.

DD2: Poster Session: Material Characterization and Devices

Session Chairs

Joerg Bagdahn

Kevin Turner

Tuesday AM, December 01, 2009

Exhibit Hall D (Hynes)

9:00 PM - DD2.1

Piezoelectric Nanogenerator by Using p-type ZnO Nanowire Arrays.

Jinhui Song ¹, Ming-Pei Lu ², Ming-Yen Lu ^{1 3}, Min-Teng Chen ³, Yifan Gao ¹, Lih-Juann Chen ³, Zhong Lin Wang ¹
¹Material Science and Engineering, Georgia Institute of Technology, Atlanta, Georgia, United States, ², National Nano Device Laboratories, Hsinchu Taiwan, ³Department of Materials Science and Engineering, National Tsing Hua University, Hsinchu Taiwan

Show Abstract

Harvesting energy from environment is a potential approach for building self-powered nanodevices/nanosystems. Piezoelectric nanogenerators using nanowires (NWs) are a technology for converting mechanical energy into electricity. Utilizing the semiconductor and piezoelectric properties possessed by ZnO NWs, an energy generator has been demonstrated for converting mechanical energy into electricity [1,2]. Under the straining created by an external force, a piezoelectric potential is created in the NW as a result of elastic deformation, which drives the flow of charge carriers through an external load. A Schottky barrier formed between the electrode and the NW serves as a “gate” that controls the flowing direction of the charge carriers. This is the principle of the nanogenerator. All of the existing literatures on nanogenerator are established using n-type ZnO NWs, while little is known about the characteristics of p-type ZnO NWs for energy harvesting. In this paper, we demonstrate the nanogenerators which could convert mechanical energy to electricity by using the p-type ZnO NWs [3]. The p-type ZnO NWs produce positive output voltage pulses when scanned by a conductive AFM in contact mode. The output voltage pulse is generated when the tip contacts the stretched side (positive piezoelectric potential side) of the NW. In contrast, the n-type ZnO NW produces negative output voltage when scanned by the AFM tip; and the output voltage pulse is generated when the tip contacts the compressed side (negative potential side) of the NW. In reference to theoretical simulation, these experimentally observed phenomena have been systematically explained based on the mechanism proposed for nanogenerator.[1] Wang, Z. L.; Song, J. H. Science 2006, 312, 242[2] Wang, X. D.; Song, J. H.; Liu, J.; Wang, Z. L. Science 2007, 316, 102.[3] Lu, M.P.; Song, J.H.; Lu, M.Y.; Chen, M.T.; Gao, Y.F.; Chen, L.J.; Wang, Z.L. Nano Letters., 9 2009 3, 1223.[4] Research supported by DARPA, DOE, NIH.

9:00 PM - DD2.10

The Effect of Ion Delivery on Polypyrrole Strain and Strain Rate under Elevated Temperature.

Yenmei Keng ¹, Priam Pillai ¹, Ian Hunter ¹
¹Mechanical Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States

Show Abstract

Conducting polymers can act as actuators when an electrochemical stimulus causes the materials to undergo volumetric changes. Ion flux into the polymer causes volumetric expansion and ion outflow causes contraction. Polypyrrole is an attractive actuator material due to its ability to generate up to 30 MPa active stress and 10-26% maximum strain with voltage supply lower than 2 V. The polymer’s mechanical performance depends upon the solvent used and the dominating ion species. In this study, we used 1-butyl-3-methylimidazolium hexafluorophosphate (BMIM-PF6) to characterize the effect of temperature increase on ion flow and how it contributes to strain and maximum strain rate of polypyrrole. In this solvent, the cation BMIM+ diffuses in and out of the polymer under applied voltage to cause strain changes. For approximately each increment of 10oC from 27-83oC, isotonic tests were done with +/-0.8 V square pauses, using a custom built device that is capable of performing temperature controlled dynamic mechanical analyses and electrochemistry simultaneously. Results showed that, independent of voltage polarity, from 27-83oC the charge per polymer volume increased by 146% and the strain increased in a similar way from 1.2% to 5.4%. Therefore, the increase in strain resulted from the charge delivered to the polymer in higher quantities at higher temperature. Moreover, both the maximum charge and strain rates were higher at positive voltage than at negative voltage throughout the same temperature range. Positive voltage caused the maximum strain rate to increase exponentially from 0.28%/s to 1.8%/s, while negative voltage caused it to increase more linearly from 0.23%/s to 0.88%/s. The results suggest that BMIM+ ions get expelled faster than being attracted in to the polymer, perhaps due to the ions preferentially remaining in the bulk solution. As the temperature increased, the ionic mobility increased and as a result, BMIM+ ions get populated back into the solvent even faster.

9:00 PM - DD2.11

Effect of Humidity on Dielectric Charging Process in Capacitive RF MEMS Switches Based on Kelvin Force Microscopy Surface Potential Measurements.

U. Zaghloul ^{1 2}, G. Papaioannou ^{3 1}, F. Coccetti ¹, P. Pons ¹, R. Plana ^{1 2}
¹MINC, LAAS/CNRS, Toulouse France, ²UPS, University of Toulouse, Toulouse France, ³Solid State Physics Section, University of Athens , Athens Greece

Show Abstract

Reliability issues especially the dielectric charging have prevented the commercialization of RF-MEMS Capacitive switches. In spite of the extensive study performed on this topic, a comprehensive understanding of the dielectric charging process is still missing. Moreover, a very few work has been performed in order to study the effect of the humidity on the dielectric charging process. On the other side, Kelvin Force Microscopy (KFM) has recently proven to be a very efficient method for assessing the dielectric charging. Yet, all of the previous KFM measurements had been performed in ambient air and hence the effect of humidity has not been considered.In this work we present for the first time the effect of humidity and hence the surface charge on the charging process in PECVD silicon nitride films based on KFM. The measurements have been performed under different relative humidity levels; in ambient air (30% - 40% RH) and under a N2 flow (RH 10%). The effect of the dielectric film thickness and the substrate nature on the charging process have been investigated through depositing SiN films with different thicknesses ranges from 100nm to 500nm over bare silicon substrates and over evaporated Au layers.For both measurements performed in ambient air and under the N2 flow, the surface potential distribution, hence the injected charge is found to decay exponentially with time following the stretched exponential law. The decay time constant is found to be much higher in case of N2 flow (1.70E+07 sec) than in the case of ambient air measurements (558sec). Besides, it is found that the surface potential distribution becomes more confined as the relative humidity decreases. For example, the Full Width at Half Maximum (FWHM) at t=0 is measured to be 1.1µm in case of 40% RH measurements comparing to 0.72µm in case of 10% RH. Charge injection time has been controlled to range from 0.1 ms to 60 sec. In air, the FWHM is found to be always larger than the FWHM measured in N2 flow for different injection times. Moreover, the FWHM increase linearly with increasing the charge injection time for measurements performed under different relative humidity levels. However, the rate of increase in the FWHM is higher for ambient air measurements. Charge lateral diffusion has been observed in case of ambient air where there it has not been observed for the N2 flow measurements. Obviously, the more hydrophobic SiN material in case of N2 flow prevents water condensation at the surface and thereby inhibits lateral charge migration due to the electrical conductivity of a possible water film. The surface potential distribution is found to be more confined in thinner than in thicker films and the relaxation time is found to be larger in the thinner dielectric films, independently of the substrate nature. Finally, the decay time constant is found to be smaller in case of dielectric films deposited over Au layers comparing to films deposited over bare silicon substrates.

9:00 PM - DD2.12

Improved Characterization of ECD and PVD MEMS Processes by X-ray Methods.

Emmanuel Nolot ¹, Agathe Andre ¹, Assunta Vigliante ², Nikolai Vyshatko ², John Pointet ¹
¹, CEA-LETI-MINATEC, Grenoble France, ², BRUKER AXS GmbH, Karlrsruhe Germany

Show Abstract

9:00 PM - DD2.13

Functionalized Nanocrystalline Diamond for Resonator Gas Sensors.

Frank Mendoza ^{1 2}, Nelson Sepulveda ^{1 3}, Brad Weiner ^{1 4}, Gerardo Morell ^{1 2}
¹Institute for Functional Nanomaterials, University of Puerto Rico, San Juan, Puerto Rico, United States, ²Physics, University of Puerto Rico, San Juan, Puerto Rico, United States, ³Electrical Engineering , University of Puerto Rico, Mayaguez, Puerto Rico, United States, ⁴Chemistry, University of Puerto Rico, San Juan, Puerto Rico, United States

Show Abstract

9:00 PM - DD2.14

Mechanical Characterization of Atomic Layer Deposited (ALD) Alumina for Applications in Corrosive Environments.

Kuang-Shun Ou ¹, I-Kuan Lin ², Ping-Hsin Wu ¹, Kuo-Shen Chen ¹
¹Mechanical Engineering, National Cheng-Kung University, Tainan Taiwan, ²Mechanical Engineering, Boston University, Boston, Massachusetts, United States

Show Abstract

Reliability and longevity of microelectronics and MEMS devices for corrosive environment applications, such as devices for oceanic signal transductions and those for oil well sensing, are traditionally major concerns of device performance evaluation. The metallic circuits are, in general, vulnerable under the attack of corrosive agents such as salty water or toxic gases. As a result, a barrier layer is required for enhancing the anti-corrosion capability of the devices. Recently, due to their nature of low temperature deposition, atomic layer deposited (ALD) dielectric films have become potential barrier materials for the above applications. Thin ALD alumina was fabricated for characterizing its capability as a barrier material for corrosive environments in this study. The specimens were fabricated (via Savannah S100, Cambridge NanoTech Inc.) by using trimethyl aluminum (TMA) and water vapor under a pressure of 0.93 Torr at 100, 200, and 300°C. The substrates were 4” silicon wafer (525μm) and Aluminum foils (380μm). The deposition rate was approximately 0.9 nm per cycle and a final thickness of 200nm was achieved after 2280 cycles. In our preliminary study, under a 90°C salty water environment, the measured resistance of aluminum wire without ALD alumina protection increased from 0.7Ω to 18MΩ in just two hours. On the other hand, under the protection of 200 nm ALD alumina, the resistance was 0.8Ω and remained essentially unchanged. However, it was also observed that after a four-day soaking, certain portions of the ALD layer were peeled off. It was believed that the salty water diffused from the micro defects existed in the ALD alumina and oxidized the aluminum wires underneath. The volume expansion and stress generated during this corrosion process caused fracture and delamination of the ALD alumina films. As a result, environmental effects on the fracture behavior of ALD alumina must be characterized. Vickers micro indentations tests were conducted on ALD alumina coated on both silicon and aluminum substrates with a load of 490~1000 mN. For fresh ALD alumina, it was interesting to find that cracks generated on the alumina/silicon samples. On the other hand, the alumina/aluminum samples exhibited no visible cracks. However, for the samples under the corrosive environment, the situation was quite different. A program is currently underway to perform detail fracture toughness variations with respect to the environmental variables. Meanwhile, the fatigue behavior of ALD films is also being characterized since the cyclic thermal stress generated by the temperature produced underneath the device during operation could possibly induce subcritical crack propagation of the protective ALD layers, which could be a serious concern for the long term reliability of related electronics and MEMS.

9:00 PM - DD2.17

Laser Induced Spatial Non-Uniformities in the Shape Memory Response of NiTi Thin Films.

Andrew Birnbaum ¹, Gen Satoh ², Y. Lawrence Yao ²
¹Materials Science and Technology, Naval Research Lab, Washington, District of Columbia, United States, ²Mechanical Engineering Dept., Columbia University, New York, New York, United States

Show Abstract

9:00 PM - DD2.18

Thin Film Bulk Acoustic Wave Resonators for Continuous Monitoring in the Physical, Chemical and Biological Realms.

Greg Ashley ¹, Paul Kirby ²
¹CMRI, Bolton Univeristy, Bolton United Kingdom, ², Cranfield University, Cranfield United Kingdom

Show Abstract

Thin Film Bulk Acoustic Wave Resonator (FBAR) is a miniaturized version of Quartz Crystal Microbalance (QCM) with dimensions a thousand times smaller, and is operated in the UHF/Microwave frequency range (300MHz-<10GHz).FBAR devices were fabricated on silicon substrate which was back etched to form an unsupported metal-piezoelectric-metal stack. The typical dimension of the FBAR is 100μm x 100μm, and the thickness of the ZnO or AlN piezoelectric films deposited by sputtering ranged 0.5μm to 3μm depending on the desired frequency of operation. A 50Ω impedance matched coplanar waveguide Ground-Signal-Ground configuration was used for the FBAR.FBAR characteristic resonance responses tend to be noisy with a low Q factor, typically 30 to 300. In order to extract the exact resonant frequency and to monitor its change with time, the FBAR resonance curves were processed in real-time with a second order polynomial fitting, making the output signal smooth and stable enough for sensing. FBARs have been used for various sensing applications with main results summarized.The FBAR is extremely sensitive changes in environmental temperature. Temperature fluctuations were measured with FBAR and even relatively low frequency devices (300MHz) gave responses ~50 KHz per 1oC with resolution of 10Hz. The responses were almost instant and were in proportional agreement with that measured by thermocouples.Also, FBAR was found to be responsive to incident light. Light fluctuations were measured with FBAR using an external light detector to correlate the incident brightness. From dark (0% saturation) to normal room florescent lighting (40% saturation) Sub GHz FBAR was shown to decrease output frequency by 2KHz.FBAR was adapted to act as a partially selective chemical sensor by the addition of the hygroscopic polymer Poly vinyl pyrrolidone (PVP). Typically the measurement ranges were <%20RH to %90RH, FBAR responded within <1 second, and with almost perfect proportional correlation to independent humidity meters measuring the same gas flow. PVP coated Sub GHz FBAR using second harmonic @~700MHz gave 4MHz change from %RH20 to %RH60.FBAR was used to detect antibody-antigen reaction. Direct comparison of FBAR performance against Quartz (QCM) for immunosensor applications was performed. 10MHz QCM units gave 1HZ frequency change per 1ng of additional antibody-antigen reaction, whereas a typical sub GHz FBAR gave 1KHz/ng, showing three orders of magnitudes greater response.The experimental results have shown it is responsive to changes in the physical realms and to addition of various coatings, demonstrated the potential as a universal sensing platform with unprecedented sensitivity.

Admin Notes

Previous Session Type (Withdrawn)

Date Withdrawn

For Change Sheet

No Show

ScholarOne Abstracts® (patent #7,257,767 and #7,263,655). © ScholarOne, Inc., 2009. All Rights Reserved.
ScholarOne Abstracts and ScholarOne are registered trademarks of ScholarOne, Inc.
Terms and Conditions of Use

9:00 PM - DD2.19

Suppression and Modeling of Inelastic Deformation in Multilayer Microcantilevers with Nanoscale Coating.

I-Kuan Lin ¹, Xin Zhang ¹, Yanhang Zhang ^{1 2}
¹Mechanical Engineering, Boston University, Boston, Massachusetts, United States, ²Biomedical Engineering, Boston University , Boston, Massachusetts, United States

Show Abstract

Multilayer microcantilevers present in micro-/nano- electromechanical system (MEMS/NEMS) applications serve passive and active structural roles. The application and commercialization of MEMS devices suffer from reliability problems. For example, the undesired inelastic deformation of MEMS infrared (IR) detectors, designed to sense thermal images, seen during long term operation. Appropriate thin film coatings, such as atomic layer deposition (ALD), have been demonstrated to be promising solutions for these reliability problems in MEMS devices. However, the micro/nano- mechanics within and/or between thin films and coatings are not fully understood, especially when temperature, time, microstructural evolution and material nonlinearities play significant roles in thermomechanical response. The overall goal of this work is to suppress and understand of the inelastic deformation and microstructural evolution in multilayer microcantilevers with nanoscale coatings, in order to better predict, design, characterize, and manufacture successful MEMS structures and devices for the next generation.In this work, the Au/SiN_x microcantilevers were held at 100 °C, 150 °C and 200 °C for approximately 60 hours. The curvature of Au/SiN_x microcantilevers decreased by 2%, 15% and 21% of the initial curvature, respectively, due to the combined effect of creep and stress relaxation. Finite Element Method (FEM) with Power-law creep was used to describe the inelastic deformation behavior of the Au/SiN_x microcantilevers during isothermal holding. To increase the reliability of Au/SiN_x microcantilevers, the 10 nm ALD Al₂O₃ layer was coated to prevent creep behavior during 150 °C isothermal holding. The time-dependent inelastic deformation can be suppressed by ALD from 15 % to 4%. Nonlinear deformation mechanisms associated with various microstructural changes in the multilayer microcantilevers were explored in this work. The microstructure evolutions due to isothermal holding in Au/SiN_x were studied using an atomic force microscope (AFM) and scanning electron microscope(SEM). The grain growth present in both the vertical and lateral directions induced the inelastic deformation during isothermal holding. Normally, in the Au layer of uncoated microcantilevers, the smaller grains coalesce to produce larger grains, increasing the surface roughness. On the other hand, the grain size of Au in coated microcantilevers does not change because the ALD layer constrains the grain growth. This microstructural evolution could be used as evidence for the suppression of creep behavior in coated microcantilevers during isothermal holding. Finally, the constraint forces of the ALD layer and the stress evolution of microcantilevers were systemically studied at different ALD thicknesses and isothermal holding temperatures using FEM.

9:00 PM - DD2.2

Modeling of a Circularly Interdigitated Piezoelectric Microactuator.

Oliver Myers ¹, M. Anjanappa ², Carl Freidhoff ³
¹Mechanical Engineering, Mississippi State University, Mississippi State, Mississippi, United States, ²Mechanical Engineering, University of Maryland Baltimore County, Baltimore, Maryland, United States, ³, Northrop Grumman Corporation, Baltimore, Maryland, United States

Show Abstract

9:00 PM - DD2.20

The Tunability in Mechanical Properties and Fracture Toughness of Sputtered Silicon Oxynitride Thin Films for MEMS-based Infrared Detectors.

I-Kuan Lin ¹, Ping-Hsin Wu ², Kuang-Shun Ou ², Kuo-Shen Chen ², Xin Zhang ¹
¹Mechanical Engineering, Boston University, Boston, Massachusetts, United States, ²Mechanical Engineering , National Cheng Kung University, Tainan Taiwan

Show Abstract

The silicon oxynitride (SiON) films, which have unique tunability in their absorbance peaks, could become ideal candidates for IR absorbers in MEMS-based IR detectors, instead of Si₃N₄ films, in order to detect targets which have specific IR radiation. Also, the undesired residual stress induced deformation after fabrication can be minimized by adjusting the oxygen and nitrogen content in the films. Although extensive studies have been conducted on sputtered SiON films in microelectronics and optical applications, only a few attempts have been made to study their mechanical properties. Therefore, the mechanical properties of SiON films associate with different compositions of oxygen and nitrogen contents are studied in this work by using an indentation technique.SiON films with a thickness of 800 nm were deposited at room temperature using Discovery 18 RF magnetron sputtering system (Denton Inc.). The SiON samples with different compositional variations of nitrogen/oxide ratios were deposited by putting both silicon oxide and silicon nitride targets into the sputtering chamber, and adjusting the RF powers applied to them. The sputtering pressure was maintained at 4 mTorr during the deposition with an argon flow rate of 25 sccm. Subsequently, the energy dispersive X-ray (EDX) spectrometer (Oxford Instruments) was employed to quantitatively analyze the compositions of the sputtered SiON films.The nanoindentation (Hysitron Inc.) testing with Berkovich tip revealed that the Young’s modulus and hardness significantly increased with increasing nitrogen content in SiON film composition. In addition, the residual stress and fracture toughness of SiON films were also studied by using a MA-1450 Profilometer (KLA Tencor) with the Stoney formula and a microindenter (Akashi MVK-H11) with Vickers tip, respectively. The residual stress of the varied SiON film compositions increased from -635 MPa to 569 MPa with increasing nitrogen content and the zero stress state of SiON film was obtained by further adjusting nitrogen/oxide ratios in the films. Finally, the fracture toughness of the SiON films, ranging from 0.7 MPa√m to 0.9 MPa√m , was found to be enhanced by increasing the oxygen content in the films. These scientific insights from this study will not only benefit the rearch on IR detectors, but also contribute to MEMS, optical devices and IC structure fabrication in regards to maintaining the structural integrity and improving the fabrication performance.

9:00 PM - DD2.21

Electrical Characterization of Aluminum Nitride (AlN) Metal-Insulator-Metal (MIM) Mesa Structures.

Jonathan Thornton ², Sarah Caprio ³, Sridhar Kuchibhatla ¹, Richard Farrell ¹, Dimitris Korakakis ^{1 4}
²Department of Mechanical and Aerospace Engineering, West Virginia University, Morgantown, West Virginia, United States, ³Deparment of Chemical Engineering, West Virginia University, Morgantown, West Virginia, United States, ¹Lane Department of Computer Science and Electrical Engineering, West Virginia University, Morgantown, West Virginia, United States, ⁴National Energy Technology Laboratory, West Virginia University, Morgantown, West Virginia, United States

Show Abstract

9:00 PM - DD2.22

Mechanical and Optical Properties of Reactively Sputtered Silicon-rich Silicon Nitride Films.

Xiaoning Wang ¹, Selcuk Yerci ², I-Kuan Lin ¹, Luca Dal Negro ², Xin Zhang ¹
¹Mechanical Engineering, Boston University, Boston, Massachusetts, United States, ²Electrical and Computer Engineering, Boston University, Boston, Massachusetts, United States

Show Abstract

9:00 PM - DD2.23

Effect of Controllable Surface Roughness on the Durability of MEMS Contacts.

John Kelley ¹, Michael Jespersen ¹, Steven Patton ¹, Richard Vaia ¹, Andrey Voevodin ¹
¹Materials and Manufacturing Directorate, Air Force Research Laboratory, WPAFB, Ohio, United States

Show Abstract

Microelectromechanical systems (MEMS) have a wide range of uses in the aerospace, defense, automotive, and wireless communications industries, but widespread commercialization of next generation MEMS devices has been hindered by reliability problems. Structural design changes necessary for improving the switching speed and electrical performance while reducing the size of MEMS are expected to exacerbate these durability problems, especially failure due to adhesion and dielectric charging. This is primarily due to incorporation of extremely smooth surfaces that will increase the strong adhesion forces between contact surfaces. Consequently, wider application of MEMS technology requires development of surface modification technologies to mitigate these effects. In our previous work, noble metal nanoparticle liquids (NPLs) were employed as surface lubricants for MEMS contacts, significantly improving junction durability in terms of number of contact cycles until failure, in both high- and low-current regimes. This improvement was thought to be a consequence of introducing multiple nanoscale surface asperities and the ability of the NPL to reconfigure itself on the contact surface. In a subsequent study, bio-derived gold/palladium hybrid nanoparticles were cast onto MEMS contact surfaces. These materials showed no liquid behavior at room temperature but were just as effective as the NPLs in improving MEMS contact durability. This suggests the improvement in contact performance is in large part due to the nanoscale surface roughness introduced by the adsorbed nanoparticles. In order to more closely examine the impact of controllable surface roughness on MEMS contact performance, gold contact surfaces were roughened by physical etch, chemical etch, or mechanical methods. Contact durabilities were examined as a function of peak-to-valley heights and radius of curvature of the surface features. The results of these studies and their implications for next generation MEMS technology will be discussed.

9:00 PM - DD2.24

Impact of Time and Temperature on the Restoring Force in RF MEMS Capacitive Switches.

Mark McLean ¹, Walter Brown ¹, Rick Vinci ¹, John Papapolymerou ², Yuan Li ²
¹, Lehigh University, Bethlehem, Pennsylvania, United States, ², Georgia Institute of Technology, Atlanta, Georgia, United States

Show Abstract

9:00 PM - DD2.26

Adhesion Enhancement and Structure Stabilization of Piezoelectric MEMS Devices by Oxygen-controlled Conductive Oxide Electrodes.

Seung-Hyun Kim ¹, Chang Young Koo ², Miso Kim ³, Wenyan Jiang ¹, Scott Hoffmann ¹, Brian Wardle ³, Angus Kingon ¹
¹Division of Engineering, Brown University, Providence, Rhode Island, United States, ²R&D Center, INOSTEK Inc., Ansan, Gyeonggi, Korea (the Republic of), ³Dept. of Aeronautics and Astronautics, MIT, Cambridge, Massachusetts, United States

Show Abstract

Recently, MEMS (Micro electro mechanical system) process has been applied for making electrical and mechanical devices such as bulk acoustic resonators (FBAR) or inkjet printer heads for next generation printers using piezoelectric films. Piezoelectric thin films have been key components of high precision sensors, actuators and ultrasonic probes which play an important role in the maintenance of many critical industrial processes. Among ferroelectric family, Pb(Zr,Ti)O3 films are considered as most promising candidates for MEMS applications since they can produce high mechanical strain under applied electric field.Generally the bulk micromachining and surface micromachining processes are used to make the 3-dimensional structure type devices. These MEMS processes are done by stacking structure type, which has top and bottom electrodes. However, there have been serious adhesion problems between piezoelectric thin films and top electrodes during or after the patterning and the etching process because of different stress level and different lattice parameters or chemical attack during the removal process of etch residues. These phenomena have been serious obstacles to realize the commercial MEMS devices. Presently, adhesion issues of bottom electrodes are better understood mechanisms and can be essentially overcome by using Ti, Ta, or Ir adhesion layer. On the other hand, poor adhesion property and its solution between top electrodes and piezoelectric thin films are significantly less understood and ill-defined. At present, several mechanisms and explanation have been proposed for this failure mode, however, it is difficult to distinguish between the possible degradation process and phenomena from standard MEMS process.To solve these drawbacks, we systematically investigated the influence of conductive oxide top electrodes using the sputtering deposition system. It is found that oxygen controlled conductive top electrodes on the PZT thin films can achieve excellent adhesion properties and induce excellent ferroelectric and piezoelectric properties for piezoelectric MEMS devices.

9:00 PM - DD2.27

Tribological Properties of Phenyl-terminated Thiols in Adhesive and Non-adhesive Contacts.

Yutao Yang ¹, Andrew Jamison ², T. Randall Lee ², Marina Ruths ¹
¹Chemistry, University of Massachusetts Lowell, Lowell, Massachusetts, United States, ²Chemistry, University of Houston, Houston, Texas, United States

Show Abstract

9:00 PM - DD2.28

Characterization of Aluminum Nitride (AlN) Thin Films through Micro Infrared Spectroscopy.

Sridhar Kuchibhatla ¹, Lee Rodak ¹, Dimitris Korakakis ^{1 2}
¹Lane Department of Computer Science and Electrical ENgineering, West Virginia University, Morgantown, West Virginia, United States, ², National Energy Technology Laboratory, Morgantown, West Virginia, United States

Show Abstract

9:00 PM - DD2.29

A New Approach to Tailoring the Electrical Properties of Silicon at Room Temperature.

Simon Ruffell ¹, Kallista Sears ¹, Andrew Knights ², Jodie Bradby ¹, Jim Williams ¹
¹, Australian National University, Canberra, Australian Capital Territory, Australia, ²Department of Engineering Physics, McMaster University, Hamilton, Ontario, Canada

Show Abstract

9:00 PM - DD2.3

Nanoindentation Characterization and Finite Element Modeling of The Viscoelastic Behavior for PDMS Micropillars.

Ping Du ¹, I-Kuan Lin ¹, Hongbing Lu ², Xin Zhang ¹
¹Mechanical Engineering, Boston University, Boston, Massachusetts, United States, ²Mechanical and Energy Engineering, University of North Texas, Denton, Texas, United States

Show Abstract

Polydimethylsiloxane (PDMS) is an important base material widely used in bio-MEMS devices to exploit the intrinsic cellular mechanical forces in bio-microsensors and bio-microactuators. The accuracy of these devices relies on choosing an appropriate material constitutive model to convert the measured structural deformations into corresponding reaction forces. Previous studies predict the contractile forces of these cells based on the deflection of PDMS micropillars using simple beam bending theory of linear elastic material. However, PDMS is an inherently viscoelastic material and its elastic modulus changes with loading rates and elapsed time. This paper reports a more in-depth viscoelastic characterization on the bending behavior of the PDMS micropillars by both the nanoindentation technique and the finite element modeling (FEM).Two forms of PDMS samples, film and micropillars are fabricated by replica molding method. The film samples are molded in 3 mm thick and then cut into 50x50 mm² squares. While the micropillar samples are patterned with 20x20 μm² holes in a 40 μm thick SU-8 photoresist mold by the soft lithography technique. The PDMS are formed by mixing a PDMS prepolymer (Sylgard 184, Dow Corning) with a curing agent at a volume ratio of 10:1, and cured at 65 °C for 90 min.The viscoelastic constitutive law of PDMS is determined by the nanoindentation technique on the film samples. Nanoindentation is an effective technique for measurements of the mechanical properties of both time-independent and time-dependent materials in micro- or nano-scales. In nanoindentation test, an ideal step loading history cannot be achieved but only approximated by applying a ramp loading with very short rise time. In this work, we apply a constant-rate displacement loading history using a diamond spherical indenter with a tip radius of 50 μm. The 2500 nm indentation depth is performed with three different loading rates, 25 nm/s, 50 nm/s and 100 nm/s. Using viscoelastic contact mechanics analysis, Young’s relaxation modulus is extracted from the resulting load-displacement data and expressed in a generalized Maxwell model. The results show a reasonably good agreement with a Young’s modulus of 0.5 ~ 0.8 MPa after 100 sec indentation. Subsequently, a viscoelastic FEM model of PDMS is built to simulate the micropillar bending. The micropillar samples are bent by a lateral force using a wedge indenter, and the experimental results are compared with the FEM simulation to validate the viscoelastic model. The verified model can be further applied in the more complicated structures representative of actual micropillars under various loading conditions, such as the force and displacement relationship of PDMS micropillar-based sensors and actuators.

9:00 PM - DD2.30

Nonlinear Piezoresistance of Polysilicon Films.

Joao Gaspar ¹, Marek Schmidt ¹, Oliver Paul ¹
¹Department of Microsystems Engineering (IMTEK), University of Freiburg, Freiburg, BW, Germany

Show Abstract

Thin-film polysilicon is used in a wide range of electronic applications such as MOS gates, emitters and passive devices. In addition, it has proven to be a suitable gauge material with higher gauge factors than metal film sensors, thus constituting the backbone of piezoresistive-based MEMS sensors. The influence of processing conditions and operating temperature on linear piezoresistive coefficients of doped poly-Si has been thoroughly characterized. The same however does not apply to higher order effects, which cannot be neglected for stresses above 200-300 MPa.This paper reports on the piezoresistive characterization of poly-Si layers. In addition to the reliable extraction of mechanical properties, our extended high-throughput method enables the acquisition of the linear and nonlinear longitudinal piezoresistive coefficients, π_l,1 and π_l,2, respectively, with statistical significance. Contrasting with previous studies where linear terms only were obtained, or π_l,2 coefficients were acquired for strains up to only 0.2 %, the data presented here are extracted for strains up to the fracture value, ca. 1 %, leading to more reliable values of higher order piezoresistive parameters.Multiple test structures are processed and sequentially measured on the same micromachined c-Si substrate. Each structure is composed of an inner movable frame connected to a fixed frame through parallel springs and supports a bridging poly-Si specimen uniaxially deformed when the frames are displaced with respect to each other. The test is conducted at temperatures between 25°C (RT) and 150°C, during which the force response, elongation and electrical resistance of the microtensile specimen are recorded. These data are converted to stress σ vs. strain ε and relative resistivity change Δρ/ρ vs. σ curves, from which Young’s modulus E, fracture strength μ and π_l,1 and π_l,2 in the model Δρ/ρ = π_l,1 σ + π_l,2 σ² are extracted. Poly-Si films characterized include LPCVD layers implanted by various phosphorous doses and in-situ doped layers deposited between 580 and 625°C, and further annealed at 1050°C for up to 10 hours.To illustrate some results, in-situ 580°C n-doped films (6.4×10^–4 Ωcm at RT) have E = 153.4 GPa, µ = 1.41 GPa and coefficients π_l,1 = –1.36×10^–10 Pa^–1 and π_l,2 = 3.27×10^–20 Pa^–2. As the RT resistivity of the implanted layers decreases from 10^–1 to 3×10^–3 Ωcm, the π_l,1 and π_l,2 coefficients vary from –0.11 to –1.08×10^–10 Pa^–1 and from 0.67 to 5.90×10^–20 Pa^–2, respectively. An increase of –π_l,1 with doping, contrary to bulk behavior, has been also observed by others and is an indication of the contribution of grain boundaries to piezoresistance. A decrease in π_l,2 of about 12.5% occurs for the most conductive layers as the temperature is ramped from RT to 150°C.The detailed dependences of both mechanical and piezoresistive (linear and nonlinear) parameters of poly-Si on processing, annealing and operating temperature will be presented and discussed.

9:00 PM - DD2.31

Characterization of the Elastic Properties of SU-8 Microposts and Their Adhesion Strength to Silicon Substrates.

Joao Gaspar ¹, Chiemi Ishiyama ², Masato Sone ², Yakichi Higo ², Oliver Paul ¹
¹Department of Microsystems Engineering (IMTEK), University of Freiburg, Freiburg, BW, Germany, ²Precision & Intelligence Laboratory, Tokyo Institute of Technology, Yokohama Japan

Show Abstract

SU-8 is an epoxy-type, near-UV negative photoresist that can be spun to a wide range of thicknesses, from below 1 µm to above 2 mm, and still be processed by standard lithography to achieve structures with aspect ratios up to 20. Because of these attributes, along with its excellent chemical and thermal resistances and biocompatibility features, SU-8 is used as a structural material in many microfluidics and microelectromechanical applications. As such, mechanical properties like elastic constant and adhesion strength to a specific substrate strongly influence the functionality and reliability of SU-8 microparts, motivating the work presented here.This paper reports on the mechanical characterization of transversally loaded SU-8 microposts and the extraction of both Young’s modulus E and adhesion strength σ_max to silicon from measured load-displacement curves. Multiple microposts are processed and sequentially measured on the same crystalline silicon substrate using a customized high-throughput setup. The testing apparatus is fully automated thus enabling the fast acquisition of reliable data with statistical relevance, as never reported before for this type of test. The SU-8 posts consist of cylinders with a radius of 62.5 μm and heights ranging from 50 to 150 µm. The structures are evenly spaced by 525 µm resulting in a layout density of about 360 posts per 10×10 mm². The substrate is placed on a motorized x-y-θ table that aligns and steps the position of each post with respect to a loading 45°-inclined micro-jig, which in turn is mounted on an actuation head comprising x-y-z stages, piezoelectric actuator and load-cell for transversal loading of the top part of the cylinder and simultaneous force measurement with a resolution of 1 mN. Displacements, which can be resolved down to 10 nm, are monitored using a laser-deflection detector.Upon contact of the tip, the force response of the structure is linear with displacement, from which a spring constant and, in turn, elastic modulus E are extracted. Spring constants of 19.0, 16.6 and 9.3 mN/µm are obtained for 50, 100 and 150 µm high posts, respectively, with standard deviations below 8%, averaged from 25 structures of each geometry. This expected scaling validates the mechanical model used, from which E = 3.09±0.26 GPa is determined, within the 2-4 GPa range specified by the manufacturer. The posts are loaded until brittle delamination occurs at their base and the failure force is used to evaluate the maximum stress, and thus adhesion strength, at the micropost-substrate interface. Expectedly, the disrupting force decreases with increasing height. A mean adhesion strength of 80.9 MPa and predictability Weibull modulus of 8.9 are obtained.Current work involves the acquisition of larger data sets and evaluation of the influence of process parameters such as exposure dose, post exposure and heat curing in the elastic constant and adhesion of microposts with broader range of aspect ratios.

9:00 PM - DD2.32

Characteristics of the QVGA IRFPAs using Sputtered Silicon Antimony Thin Film as a Detection Layer.

Hojun Ryu ¹, Dong-Keun Lee ², Seong Mok Cho ¹, Sang-Hoon Cheon ¹, Woo Seok Yang ¹, Chang Auck Choi ¹
¹, ETRI, Daejeon Korea (the Republic of), ²Dept. of Electronics Enginnering, Dankook University, Cheonan Korea (the Republic of)

Show Abstract

The microbolometers have been noticed for its compact size, low-cost, and the compatibility with semiconductor process. The microbolometer generates the voltaic signal by changing its resistance for responding the infrared radiation. Hence the microbolometer should have the appropriate electrical and thermal properties when it has been fabricated. An IR detecting layer in the membrane with heat capacity C which converts the incident infrared radiation to heat, and which is attached to a heat sink at temperature Ts via thermal conductance G. Furthermore the detecting layer should have a high temperature coefficient of resistance (TCR), low thermal conductance and low noise. PECVD deposited amorphous silicon has been known a good IR detecting material for its thermal property and process compatibility. But the 1/f noise is a big problem to obtain high detectivity. According to the previous work, the sputtered silicon showed low 1/f noise than PECVD deposited silicon film. In this study we fabricated microbolometers using silicon antimony thin film as a IR detection layer. Obtaining a QVGA IRFPA (Infrared Focal Plane Array), readout IC(ROIC) wafers have been processed firstly. In the ROIC, 320x8 array used for CDS (Correlated Double Sampling) and another 320x8 blind pixel array for skimming. To integrate the ROIC and IRFPA structures, we especially designed IRPFA’s anchor area where the electrical paths exist. PECVD deposited Si3N4 layers have been used for insulation and protection layer. For thermal isolation and maximizing IR of 8 um to 14 um wavelength, the 2 um gap between aluminum refletion layer and membrane has formed by sacrificial polyimide ashing. The size of unit pixel is 50 um x 50 um having 2 um legs, which has more than 77 % fill factor. To achieve high NETD (Noise Equivalent Temperature Difference), 1/f noise factor should be minimized. Therefore we measured the 1/f noise characteristics with dynamic signal analyzer (SR 780). The relation of current spectral density is given by below equation. S_(if^2 )=(KI_b^2)/f^α Where Ib is bias current and K is 1/f noise parameter. In order to know the effect of fabrication method for the film, we compared the 1/f noise characteristics of sputtered silicon antimony and PECVD deposited amorphous silicon film. The sputtered silicon antimony film has low noise level compared to the PECVD deposited film. The 1/f noise parameter K which is proportional to the total noise level is 0.4 x 10-10 to 224.9 x 10-10 with deposition condition. In case of using developed film as a bolometric material, we calculated the detectivity of microbolometer. At the similar TCR, the total noise of PECVD deposited silicon has shown much larger than that of sputtered silicon antimony film. Therefore the performance of microbolometer which is adopted sputtered silicon antimony film as a bolometric material would be better than using the case of PECVD silicon film.

9:00 PM - DD2.33

Thickness-Dependant Electrical Characteristics of Nitrogen-Doped Polycrystalline 3C-SiC Thin Film Deposited by LPCVD.

Man I Lei ¹, Te-Hao Lee ¹, Mehran Mehregany ²
¹Department of Materials Science and Engineering, Case Western Reserve University, Cleveland, Ohio, United States, ²Department of Electrical Engineering and Computer Science, Case Western Reserve University, Cleveland, Ohio, United States

Show Abstract

Silicon carbide (SiC) is an excellent material for realization of micro-electro-mechanical systems (MEMS) suitable for use in harsh environments, e.g., high temperatures and corrosives. In this context, polycrystalline 3C-SiC (poly-SiC) is used as a key structural film in fabrication of MEMS devices. Here, we study the effect of film thickness on the electrical resistivity of nitrogen-doped poly-SiC thin films deposited by dual precursors of dichlorosaline and acetylene, using ammonia as the doping source gas, in a low pressure chemical vapor deposition (LPCVD) reactor. We have consistently observed that the resistivity of our nitrogen-doped poly-SiC thin films decreases substantially, from 3.8x10^-2 Ω-cm for very thin films to 5.5x10^-3 Ω-cm for 1 μm films, above which thickness the resistivity stabilizes. To understand this effect, van der Pauw resistivity measurements have been performed on a series of SiC films ranging in thickness from 100 nm to 1 μm and deposited on 4-inch (100) Si wafers, electrically isolated with a 500 nm thermally-grown silicon dioxide layer. Secondary ion mass spectrometry analyses show nitrogen concentration remains constant at about 9x10²⁰ atoms/cm³ for the aforementioned range of film thicknesses. In addition, Hall measurements are performed on the same van der Pauw structures to obtain the active carrier concentration contributing to electrical conduction. The results reveal that only 45% to 55% of the dopants are electrically active in 100 nm films. The percentage of active dopants rises to 80% when film thickness increases to 700 nm. The extracted electron mobility also shows an increasing general trend with film thickness. From the studies of surface roughness and microstructure obtained from atomic force microscopy and cross-sectional transmission electron microscopy, small grains are formed at the initial stage of deposition, which grow into larger columnar grains as film thickness increases. The presence of large density of grain boundaries and limited grain growth near the SiC/oxide layer interface contributes to the increased electrical resistivity as a result of increased trapped mobile carriers and reduced carrier mobility. Temperature coefficient of resistance of each film thickness are measured and compared; thinner SiC shows a larger resistance dependency on temperature.

9:00 PM - DD2.34

Identification of Optimized Ti-Ni-X Shape Memory Alloy Compositions for Thin Film Actuator Applications.

Robert Zarnetta ^{2 1}, Alan Savan ¹, Michael Ehmann ¹, Alfred Ludwig ^{1 2}
²Research Department “Integrity of Small-Scale Systems and High-Temperature Materials (IS3/HTM), Ruhr-University Bochum, Bochum Germany, ¹Institute for Materials, Ruhr-University Bochum, Bochum Germany

Show Abstract

Thin film shape memory alloys (SMA) are of high interest for miniaturized actuators due to their high energy density and tailorable transformation temperatures. For a successful application of SMAs in micro-actuators several requirements have to be fulfilled simultaneously: high actuation effect (switchable stress value, Δσ), high transformation temperatures (here AF), low thermal hysteresis width (ΔT) and a small temperature interval of transformation (ΔI). In literature most of the time these requirements were optimized individually and only for very restricted sets of alloys. Here we present our combinatorial approach towards fabrication and characterization of shape memory alloys thin film actuators and introduce a figure of merit (FOM) in order to determine the compositions with optimal shape memory properties for micro-actuator applications.Ti-Ni-Cu shape memory thin films within a broad composition range, i.e. Ti64 xNi27+xCu9 (x= 0 – 22 at. %) and Ti49-xNi37Cu14+x (x= 0 – 9.5 at. %) were investigated by a high- throughput cantilever deflection method. The transformation temperatures and the thin film stress changes were determined. A clear dependency of the actuator response on the thin film composition was found. The optimal shape memory alloy composition with improved functional properties, i.e. high stress change, high transformation temperatures, low thermal hysteresis width and small temperature interval of transformation, were identified using a figure of merit defined as (Δσ*Af)/(ΔT* ΔI). Ti50Ni41Cu9 and Ti45Ni46Cu9 exhibit the best shape memory properties for compositions showing a B2-B19 and a B2-R phase transformation, respectively. Furthermore, data for binary and other ternary shape memory alloy systems, e.g. Ti-Ni-Pd, -Hf, -Ag will be analyzed with respect to their FOM and compared to the Ti-Ni-Cu thin film actuators. Acknowledgments: Financial support from the Research Department IS3/HTM at the Ruhr-University Bochum and from the DFG within “SFB 459” is acknowledged.

9:00 PM - DD2.6

Characteristics of Output Voltage and Current of Integrated Piezoelectric Nanogenerators.

Rusen Yang ¹, Yong Qin ¹, Cheng Li ¹, Liming Dai ², Zhong Lin Wang ¹
¹School of Materials Science and Engineering, Georgia Institute of Technology, Atlanta, Georgia, United States, ²Department of Chemical and Materials Engineering, University of Dayton, Dayton, Ohio, United States

Show Abstract

9:00 PM - DD2.7

Polymeric Azobenzene Wire Actuators.

Jianxia Zhang ¹, John Wiley ¹
¹Chemistry/Advanced Materials Research Institute, University of New Orleans, New Orleans, Louisiana, United States

Show Abstract

9:00 PM - DD2.8

Material Sensitive MEMS Resonator Optimization.

Ambarish Roy ¹, Bradley Barber ², Kanti Prasad ¹
¹Electrical and Computer Engineering, University of Massachusetts Lowell, Lowell, Massachusetts, United States, ², Skyworks Solutions, Woburn, Massachusetts, United States

Show Abstract

Wireless applications are rapidly increasing which results in an increase in the allocations of the frequency spectrum. Newer technologies have therefore pushed their carrier frequencies higher and higher, which complements the demand for smaller and more portable devices. For emerging wireless applications, Bulk Acoustic Wave (BAW) resonators are promising following the demand of smaller size, low power consumption and higher operating frequencies. They consist of a thin piezoelectric film between electrodes where through electromechanical coupling energy is stored as both mechanical and electromagnetic energy. Being slower than electromagnetic waves, the wavelength of acoustic waves is short and therefore helps in miniaturization of these devices. To decouple the BAW resonator structure from the substrate, Solidly Mounted Resonators (SMR) is one of the types of structure used. Analogous to the transmission line impedance transformer, the SMR consists of an acoustic mirror which is made up of layers of alternate low and high impedances of thicknesses roughly equal to the quarter wavelength of the desired center frequency.Designed data comprises of various parameters of the materials that are used in the stack up of the SMR, for e.g. Speed of acoustic wave in the material, Density of material, its Dielectric constant, Piezo constant, and Conductivity etc. The initial data of these parameters are taken from the text book values and are then evolved as the design grows. These parameters generally vary from one material manufacturer to another and by the processing methods used, with which they are deposited, processed etc. at the fabrication facility. Considering the spectrum where the central frequency is located, the fabricated data after measurements has the tendency to follow the modeled or designed data. But if a wide spectrum is investigated the measured data appears not to have matched entirely with the designed data. This proves that there is a constant need of adjustments to these parameters as in when required.A research to evaluate these parameters by unique extraction techniques based on electrical measurements is proposed. We present a material sensitive model of SMR taking into account the speeds and densities of different materials. The technique used to extract important information about material properties involves comparing the modeled data to the measured data after fabrication. The sensitivity of the impedance peaks in the response of the material sensitive SMR to the variation of each material’s parameters is very high, resulting in a better accuracy for the optimization. Different optimization processes are tested for quicker simulation. The models designed are uniquely simulated for the different optimization schemes to generate solutions with least error. Various designing complications and complexity reduction techniques are discussed.

9:00 PM - DD2.9

Fabrication,Transport and Raman Studies of Pulsed Laser Deposited Al/Ga Doped PBCO Thin Films.

Hom Kandel ^{1 2}, Tar Pin Chen ², Hye Won Seo ², Milko Iliev ⁴, Fumiya Watanabe ³, Zhongrui Li ³, Lixi Yuan ², Dever Norman ^{1 2}
¹Applied Science Department, University of Arkansas at Little rock, Little rock, Arkansas, United States, ²Physics Department, University of Arkansas at Little rock, Little rock, Arkansas, United States, ⁴Texas Center for Superconductivity and Physics Department, University of Houston, University of Houston, Houston, Texas, United States, ³Nanotechnology Center, University of Arkansas at Little Rock, Little Rock, Arkansas, United States

Show Abstract

2009-12-01 Show All Abstracts

Symposium Organizers

Joerg Bagdahn Fraunhofer Center for Silicon Photovoltaics

Norman Sheppard MicroCHIPS, Inc.

Srikar Vengallatore McGill University

Kevin Turner University of Wisconsin-Madison

DD3: Devices and Materials

Session Chairs

Srikar Vengallatore

Tuesday AM, December 01, 2009

Fairfax (Sheraton)

9:30 AM - **DD3.1

CMOS Nanowire FET Biosensors.

Mark Reed ¹
¹, Yale, New Haven, Connecticut, United States

Show Abstract

10:00 AM - DD3.2

In-situ Measurement of Actuation in Thin Films of Conducting Polymers.

Lauren Montemayor ¹, Priam Pillai ^{2 3}, Ian Hunter ^{2 3}
¹Aeronautics and Astronautics, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States, ²Mechanical Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States, ³, Institute of Soldier Nanotechnologies, Cambridge, Massachusetts, United States

Show Abstract

10:15 AM - DD3.3

Microfabricated Large-area Freestanding Nanostructured Solid-Oxide Fuel Cell Membranes for Portable and Clean Energy.

Bo-Kuai Lai ¹, Alex Johnson ¹, Shriram Ramanathan ¹
¹, Harvard School of Engineering and Applied Science, Cambridge, Massachusetts, United States

Show Abstract

Solid oxide fuel cells (SOFCs) are electrochemical devices that combine hydrogen or hydrocarbon with oxygen to produce electricity at elevated temperatures. Their high electric conversion efficiency, high power density, and clean emissions are attractive for stationary and automobile applications. By taking advantage of well-established microfabrication and thin film deposition techniques, low-footprint SOFCs for portable applications have received increasing attention in recent years. The heart of μSOFCs is a microfabricated membrane consisting sub-μm cathode, electrolyte, and anode layers, with each possessing distinctly different physical-chemical properties. To maintain structural integrity and functionality of such thin heterostructures, microstructures of each layer and microfabrication procedures need to be controlled precisely. In particular, stress-induced failures in these μm-scale or sub-μm membranes need to be avoided. The failure takes place in both compressive and tensile stress regions when stresses in membranes exceed critical values. In this presentation, we will discuss our recent results on designing large area nanostructured oxide fuel cell membranes for portable power. Synthesis, stress design, and thermal protocols to enable robust fuel cell devices will be presented. Moreover, microfabrication procedures for membranes that minimize possible Si-poisoning and formation of reaction layers at interfaces were introduced. By precisely controlling the processing parameters, we were able to demonstrate crack-free oxide cathodes with electrical performance compatible to their macrocrystalline counterparts and large-area self-supported μSOFC membranes. Power sources on silicon platforms with active area ranging from hundreds of microns to millmeters square and with fuel cell device thickness below 250 nm will be presented. MEMS routes for 3-D integration of such power sources will be highlighted. These devices can produce power down to 200oC range and open up opportunities for portable clean energy sources.

10:30 AM - DD3.4

Phase Transformation Behavior of Patterned Lines of NiTi Shape Memory Thin Films.

Xu Huang ¹, Ainissa Ramirez ^{1 2}
¹Mechanical Engineering, Yale University, New Haven, Connecticut, United States, ²Materials Science and Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States

Show Abstract

10:45 AM - DD3.5

Thick Films of the TiO2 – TeO2 System for Integrated Functional Electronic Devices.

Paula Vilarinho ¹, Su Xinming ¹, Aiying Wu ¹
¹Dep. of Ceramics and Glass Engineering, University of Aveiro, Aveiro Portugal

Show Abstract

Tellurium glasses are of technical interest due to their high refractive index, good optical transmission in the IR region and semiconducting properties, in addition to the low melting points and absence of hygroscopic properties. Comparatively, polycrystalline tellurium based oxides have been much less exploited for electronic applications [1-2]. However, because most of the tellurium-based oxide materials can be synthesised and sintered at temperatures below 900οC, they become potential candidates for integrated passive electronic devices. Indeed, in integrated passive electronic devices the ceramic layers need to be co-fired with the inner metal electrodes and the manufacturing temperatures are determined by the melting point of the conducting metals. Concomitantly, due to a current demand for reduced volume of microelectronic devices, i.e. high density, small size and lightweight integrated components and modules, fabricated at low costs, thin and/or thick films are being considered to replace the in use bulk materials. The fabrication of tellurium based thick films has not yet been reported.In this study thick films of TiO2- TeO2 system were prepared by electrophoretic deposition and the dielectric properties characterized. The influence of the processing parameters on the dielectric properties of single-phase TiTe3O8 thick films was considered. 53 μm thick TiTe3O8 films sintered at 700οC for 5 h exhibit εr = 60 and tanδ = 0.006. TCε of these TiTe3O8 films was calculated to be +78 ppm/οC between 35 and 200οC. TeO2 and TiO2, which display a negative (-119 ppm/οC) and a positive (450 ppm/οC) temperature coefficient of resonant frequency, respectively were used to tune the temperature coefficient of resonant frequency of TiTe3O8 thick films. The comparison with the behaviour of equivalent ceramics is established.ReferencesM. Udovic , M. Valant, D. Suvorov, Journal of the European Ceramic Society, (21) 1735–1738 (2001).[2] Do-Kyun Kwon, M. T. Lanagan, T. R. Shrout, Journal of the American Ceramic Society, 88 (12) 3419–3422 (2005).

11:00 AM - *

Break

11:30 AM - **DD3.6

Piezoresistive Force Sensing Optimization and Applications.

Beth Pruitt ¹
¹mechanical engineering, Stanford, Stanford, California, United States

Show Abstract

12:00 PM - **DD3.7

Materials Issues in MEMS Thermomechanical Actuators.

S. Spearing ¹, Prasanna Srinivasan ¹
¹, University of Southampton, Southampton United Kingdom

Show Abstract

12:30 PM - DD3.8

Processes for Integration of Piezoelectric/Diamond Hybrid Films and Fabrication of Device Structures for High Performance MEMS/NEMS.

Orlando Auciello ^{1 2}, David Czaplewski ², Anirudha Sumant ², Bernd Kabius ³, Jon Hiller ³, Derrick Mancini ²
¹Materials Science Division, Argonne National Laboratory, Argonne, Illinois, United States, ²Center for Nanoscale Materials, Argonne National Laboratory, Argonne, Illinois, United States, ³Center for Electron Microscopy, Argonne National Laboratory, Argonne, Illinois, United States

Show Abstract

A new generation of low power microelectromechanical and nanoelectromechanical system (MEMS/NEMS) devices will require new materials and the integration of dissimilar materials, and new micro and nanofabrication processing techniques to achieve high device performance. Most devices for MEMS/NEMS are currently based on silicon because of the available surface micromachining technology. However, the poor mechanical and tribological properties of Si are not suitable for many high-performance devices, such as resonators and switches. A novel ultra-nanocrystalline diamond (UNCD) material developed in thin film form at Argonne exhibits exceptional mechanical and tribological properties that make UNCD a suitable material for a new generation of high-performance MEMS/NEMS. Piezoelectric-based MEMS/NEMS attracts much attention due to their high sensitivity and low electrical noise in sensing applications and high-force output in actuation applications. Pb(ZrxTi1-x)O3 (PZT) thin films exhibit excellent piezoelectric properties to enable low voltage (≤ 3 V) piezoelectrically actuated devices. The integration of PZT and UNCD is challenging, due to the PZT/UNCD interface and the need to grow PZT at high temperature in oxygen in the presence of a carbon-based material such as diamond. In addition, lithography and RIE process need to be developed to fabricate hybrid PZT-UNCD devices. We will present work done in our laboratory to achieve integration of PZT and UNCD layers via robust TiAl oxygen diffusion barriers and to develop lithography and RIE process to fabricate hybrid PZT/UNCD structures. Specifically, we will report work focused on developing RIE chemistry for etching dissimilar materials such as oxide PZT, diamond UNCD, Pt electrode and TiAL diffusion barrier layers. We will present data and movies showing low voltage (1-3 V) actuation of hybrid PZT/UNCD piezo-actuated resonator and nanoswitches tested up to 1 billion cycles in the range 10 KHz to 1 MHz. This work was supported by the US Department of Energy, BES-Materials Sciences, under Contract DE-AC02-06CH11357

12:45 PM - DD3.9

Control of the Crystallographic Orientation of PZT Thin Films on Silicon Substrates by a Single Process Step.

Minh Nguyen ¹, Ruud Steenwelle ¹, Matthijn Dekkers ¹, Dave HA Blank ¹, Guus Rijnders ¹
¹, University of Twente, Enschede Netherlands

Show Abstract

DD4: 3D Integration and Assembly

Session Chairs

Norm Sheppard

Tuesday PM, December 01, 2009

Fairfax (Sheraton)

2:30 PM - **DD4.1

MEMS Assembly Using AlGe Eutectic and Reactive Multilayer Wafer Bonding.

Matthias Petzold ¹, Bianca Boettge ¹, Joerg Bagdahn ², Maik Wiemer ³, Arnold Schneider ⁴, Holger Rank ⁴, Roland Mueller-Fiedler ⁴
¹, Fraunhofer Institute for Mechanics of Materials, Halle Germany, ², Fraunhofer Center for Silicon Photovoltaics, Halle Germany, ³, Fraunhofer Research Institution for Electronic Nano Systems , Chemnitz Germany, ⁴Microsystem Technologies (CR/ARY) , Robert Bosch GmbH, Stuttgart Germany

Show Abstract

The contribution addresses current research results using the eutectic Al/Ge system as well as new possibilities based on reactive multilayer systems regarding wafer bonding for MEMS assembly. Several advantages of Al/Ge eutectic bonding are highlighted in comparison to the currently applied seal glass bonding techniques. These include a potentially higher bonding strength providing the opportunity to increase the level in device shrink and in integration density. Moreover, also semiconductor process compatibility, reliability considerations, and the possibility to form electrically conducting interconnects contribute to attract attention for the Al/Ge bonding system. On the other hand, the formation of the eutectic bond zone depends on the bonding temperature conditions, the thermal budget and the metal film thickness. These factors determine the nucleation kinetics in the bonding zone with respect to the amount of liquid phase formed, and thus affect both the bond interface homogeneity and technological process stability. Results of microstructure investigations including SEM, TEM and x-ray analysis as well as mechanical characterization based on shear, micro chevron and nanoindentation tests will be presented to characterize the microstructure of Al/Ge bonds and to assess the achieved bonding quality.An additional alternative to eutectic wafer bonding technologies is given by current approaches based on reactive multilayer systems. These technologies utilize the exothermic reaction occurring in PVD-formed multilayer stacks such as Ti/Al nanofilms. While the reaction provides the required heat for bonding, additional solder layers allow the formation of a stable intermediate bonding layer between metallized Si wafers. In this study, particular respect was given to characterize the multilayer microstructure and reaction kinetics, as well as to demonstrate first potential MEMS application examples. In addition, advantages and limitations of reactive multilayer bonding compared to eutectic bonding and the required further developments will be discussed.

3:00 PM - DD4.2

Aluminum Thermo Compression Bonding Characterization.

Erkan Cakmak ¹, Viorel Dragoi ², Eric Pabo ¹, Thorsten Matthias ¹
¹Technology, EV Group Inc., Tempe, Arizona, United States, ²Technology Development & IP, EV Group, St. Florian/Inn Austria

Show Abstract

Different methods of wafer-level bonding are nowadays widely used in the MEMS industry. The application, use, requirements, and wafer bonding process conditions differ significantly with the application type. One method that has been applied in the manufacturing of MEMS devices is metal thermo-compression bonding. The application includes but is not limited to sealing frames for hermetically sealed devices, electric contacts in 3D technology devices, mechanical support between two wafers. Physical, electrical, and thermodynamic properties of the given material plays a crucial role in the feasibility of the bond process and must be considered already at the design stage of an application.Aluminum Thermo Compression Bond (TCB) method is an interesting process as it may be considered as an alternative to Au or Cu.One major difference between the metals listed above is surface oxidation. Gold is a noble metal and does not react with oxygen to form oxides. Copper on the other hand is relatively easily oxidized under ambient conditions. However the main advantage to the copper oxides is that it can be removed fairly easily with a wet chemical approach or the copper-oxygen reaction is reversed by exposure to reducing gases such as forming gas (H2/Ar) at elevated temperatures. Aluminum will readily form a passivation layer when exposed to oxygen. This passivation layer is very stable. The TCB of identical materials is a solid state reaction and relies on interdiffusion of atoms/molecules. The bond is created by the application of high force to the wafer stack under high temperatures for extended periods of times. By increasing the temperature the material hardness decreases which therefore reduces the amount of pressure necessary to bring the two surfaces into immediate contact.The interface atoms diffusion is greatly inhibited by metal oxides on the initial surfaces brought into contact. Metal surface oxides act as a diffusion barrier and are highly undesirable. In this work the Aluminum TCB is investigated. The effects and significance of various bond process parameters and surface treatment methods are reported.The bonding process variables investigated were bond temperature, bond force and bond time. Bonded wafer samples were investigated with Scanning Acoustic Microscopy (SAM), Scanning Electron Microscopy (SEM), and four point bending test. Experiments were completed on 150 mm Si substrates with 5000 Å sputtered Aluminum (1% Si) on both wafer surfaces. A positive correlation between bond strength and bond temperature was determined. A bond temperature of 500 °C and greater is necessary to obtain bond strengths of 8-10 J/m2. It was found with increasing bond temperature the bond quality in terms of percent bonded area increased. Gaseous pretreatment of the Al surface did not affect the bond quality. Bond time within the limits chosen in this experiment did not affect the bond quality either.

3:15 PM - DD4.3

Polycrystalline Silicon-Germanium Electrode Contact Technology Improvement for MEMS Applications.

Gert Claes ^{1 2}, Simone Severi ¹, Stefaan Decoutere ¹, Jean-Pierre Celis ², Ann Witvrouw ¹
¹CTI, IMEC, Leuven Belgium, ²MTM, K.U.Leuven, Leuven Belgium

Show Abstract

Introduction
Poly-crystalline silicon-germanium (poly-SiGe) enables a monolithic CMOS-MEMS integration due to its low, CMOS-compatible deposition temperature (≤450°C) [1]. A low resistive contact between poly-SiGe and CMOS top metal layer is needed for a successful monolithic integration with low interconnect parasitic resistance. Eyoum et al. [2] proposed a Ni-silicide process to achieve low contact resistances. In this work an improved contact quality is achieved without the use of complex and thus costly silicide processing. The use of a simple, standard Ti-TiN interlayer is proposed.
To connect the MEMS bottom electrode to the CMOS top metal, a via through the CMOS passivation is defined by lithography, followed by a RIE etch and resist removal step. This resist strip includes an oxygen plasma to remove polymers. During oxygen plasma exposure, the contact interface can oxidize, resulting in a non-ohmic behavior of the contact.
In this work the contact resistance between boron doped SiGe films and AlCu(0.5%)-TiN films was investigated using different contact cleaning treatments, SiGe recipes and optional interfacial layers.
Results and discussion
In a first test a 4 μm thick poly-SiGe layer consisting of 4x1 μm stack with amorphous PECVD seedlayer (as reported in [2]) was tested with and without HF dip after contact etch and strip. The contact resistance was studied using cross bridge Kelvin resistor structures with contact sizes ranging from 2x2μm² to 10x10μm². All cases resulted in non-ohmic contacts. The use of a H₂-plasma before the SiGe deposition clearly results in a more clean contact interface, but the contact is still non-ohmic. The removal of the amorphous seedlayer from the stack did not result in any improvement.
Subsequently thin Ti-TiN (5/10nm) and Ta-TaN (5/10nm) layers were tested as optional interfacial layers. Also different B₂H₆ flows were used during the 400 nm SiGe CVD deposition.
Almost all of these test conditions resulted in ohmic contacts.
It can be seen that the soft sputter etch improves the contact interface. Lowering the B₂H₆ flow from 60 to 20 sccm also results in a lower resistance. A higher B₂H₆ flow results in smaller grained SiGe and thus an increased number of grain boundaries. The obtained contact resistivity is comparable with the one achieved by the Ni-silicide processing for all contact sizes (6.2±0.4 x 10^-7 Ωcm² for a 2x2μm² contact) for SiGe deposited at 450°C.
Conclusion
The electrical resistivity achieved in this work is better than previously reported data by Ni-silicide processing. The use of a simple Ti-TiN interlayer ends up in a good CMOS-MEMS contact interface without the use of a complex Ni-silicide process.
REFERENCES:
[1] A. Witvrouw et al., Mat. Res. Soc. Symp. Proc. Vol. 782, pp 25-36, 2004
[2] M. Eyoum et al., J. of Electrochem. Soc. 151 (3) J21-J25, 2004
[3] G. Claes et al., Mat. Res. Soc. Symp. Proc. Vol. 1075-J05-02 (2008)

3:30 PM - DD4.4

Three-dimensional Vertical Self-assembly of Heterogeneous Microstructures using Railed Microfluidics.

Su Eun Chung ¹, Yoonseok Jung ¹, Sunghoon Kwon ¹
¹, Seoul National University, Seoul Korea (the Republic of)

Show Abstract

We demonstrate three-dimensional (3D) vertical self-assembly of two-dimensionally patterned microstructures using optofluidic maskless lithography (OFML) and railed microfluidics. With conventional MEMS technology, heterogeneous materials’ patterning in vertical direction is processed parallel to the substrate; first by patterning a single material in xy axis and then by repeating the patterning with deposited different materials layer by layer in z direction. Thus, it generally requires complex integration steps including sacrificial layer, production cost and time for vertical patterning such as void area inside the final device structure. These difficulties in vertical heterogeneous patterning are solved with our method. 3D railed assembly method enables easy, fast and diverse vertical patterning and 3D assembly of microstructures in microfluidic channels. The key principle of 3D railed assembly is based on a rotation of 2D patterned microstructures from xy axis to yz axis and repeating the rotation process to assemble various 2D patterned structures in x direction. Since every horizontally patterned structure is rotated to vertical direction in 90 degrees, 3D railed assembly technology shows diverse vertical patterning and vertical assembly in contrast to the conventional fabrication method which deposits flat patterned layers. The microfluidic channel is divided into two areas: fabrication region (80 micrometers height) for 2D photopatterning of microstructures using OFML [1], and assembly region (300 micrometers height) for rotation in 90 degrees of fabricated 2D structure and assembly of rotated structures using railed microfluidics [2]. The channel has common side-guiding rails in both edge sides to move fabricated structures along the rail from fabricated region to assembly region. First, 2D photopatterning is done in fabrication area using OFML, a dynamic synthesis of photocurable resin flowing in microfluidic channels via projection of UV patterns controlled by digital micromirror device [1]. Then, fabricated 2D structure is fluidically transferred and erected in 90 degrees due to an anchor at the bottom of the channel guided by side rails in the assembly area. Therefore, individually fabricated 2D microstructures with different patterns are assembled vertically due to the fluidic force within the deep assembly region guided by side rails. On the basis of 3D vertical self-assembly method, we finally demonstrated heterogeneous self-assembly of 2D patterned microstructures composed of different materials, such as superparamagnetic materials, silicon devices and living cells, with different patterns. Our method is fast and simple way to assemble various materials in 3D with low manufacturing cost.[1] S. E. Chung et al, Applied Physics Letters, 91, 041106, 2007.[2] S. E. Chung et al, Nature Materials, 7, 581-587, 2008.

3:45 PM - *

Break

4:15 PM - DD4.5

Electrowetting-Controlled Liquid Mirrors Based on Self-Assembled Janus Tiles.

Michael Bucaro ¹, Paul Kolodner ², J. Ashley Taylor ³, Alex Sidorenko ⁴, Joanna Aizenberg ¹, Tom Krupenkin ³
¹School of Engineering and Applied Sciences, Harvard University, Cambridge, Massachusetts, United States, ², Bell Laboratories, Murray Hill, New Jersey, United States, ³Mechanical Engineering, University of Wisconsin-Madison, Madison, Wisconsin, United States, ⁴Chemistry & Biochemistry, University of the Sciences in Philadelphia, Philadelphia, Pennsylvania, United States

Show Abstract

4:30 PM - DD4.6

Self-Sealed and Self-Limited Nanochannels by E-Beam Lithography and Atomic Layer Deposition.

Sung-Wook Nam ¹, Min-Hyun Lee ¹, Seung-Hyun Lee ¹, Do-Joong Lee ¹, Stephen Rossnagel ², Ki-Bum Kim ¹
¹Department of Materials Science and Engineering, Seoul National University, Seoul Korea (the Republic of), ², IBM. T. J. Watson Research Center, Yorktown Heights, New York, United States

Show Abstract

The electrical manipulation of ions and bio-molecules is a promising direction for bio-health applications of electrofluidics. Surface charges on a nanochannel wall allow a screening effect of ions which eventually serve as majority carriers in nanochannels. In order to utilize surface charges in electrofluidic system, reliable fabrication of nanopores or nanochannels is essential with molecular level feature size, namely, sub-10 nm.(1) Here, we report on the fabrication of self-sealed and self-limited nanochannels by atomic layer deposition (ALD). We started with trench structures which have ~60 nm width and 60~80 nm depth on an a-Si (20 nm)/SiO2 (300 nm) substrate fabricated by e-beam lithography and reactive ion etching (RIE) processes. A two-step RIE was employed to sequentially remove a-Si and SiO2 through SF6 and CHF3 based etch-chemistries, respectively. Then, the sample was dipped in dilute HF (DHF) to selectively and controllably remove SiO2 underneath a-Si. This isotropic wet etching process led undercut-shaped trench structures known as pipes. The pipe has 80~100 nm diameter with a narrow trench-opening such as ~60 nm. Finally, thick (~100 nm) ALD film deposition self-sealed the pipe structures, forming a closed tube. During the ALD process, conformal films leaving a center, longitudinal void which will becomes the nanochannel. The nanochannel diameter is only determined by DHF wet-etching time regardless lithographically-defined dimension, which offers “self-limiting” process. By controlling DHF wet-etching times, we successfully generated sub-10 nm open nanochannels using dielectric ALD film, such as TiO2 or Al2O3. In order to build electrofluidic transistor structures, core(dielectric)/ shell(metal) nanochannels were fabricated through masked ALD film depositions. Patterned ALD films of metal(Ru) and dielectric(Al2O3) on pipe structures enabled us to build core/shell (Al2O3/Ru) nanochannels, where the Ru film serves as a gate-electrode to control an ionic current. The full integration scheme of nanochannel electrofluidic transistors will be presented.(1) S. W. Nam, M. J. Rooks, K. B. Kim, and S. M. Rossnagel, Nano Lett. 9, 2044 (2009)

4:45 PM - DD4.7

Layer-by-Layer Assembly in Nanofluidic Devices.

Jonathan DeRocher ¹, Pan Mao ², Jun Young Kim ³, Jongyoon Han ^{4 5}, Michael Rubner ³, Robert Cohen ¹
¹Chemical Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States, ²Mechanical Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States, ³Materials Science and Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States, ⁴Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States, ⁵Biological Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States

Show Abstract

DD5: Poster Session: Devices and Fabrication

Session Chairs

Norm Sheppard

Srikar Vengallatore

Wednesday AM, December 02, 2009

Exhibit Hall D (Hynes)

9:00 PM - DD5.10

Fabrication of Microstructures with High Aspect Ratios by Anisotropic Electrochemical Etching of Al.

Tatsuro Fukushima ¹, Kazuyuki Nishio ¹, Hideki Masuda ¹
¹Applied Chemistry, Tokyo Metropolitan Univ., Tokyo Japan

Show Abstract

9:00 PM - DD5.11

Breast Cancer Detection Through Breath Analysis by Porous SnO2 Based Gas Sensors.

Kevin Bechtold ¹, Sean Chislett ¹, Chunlei Wang ²
¹Biomedical Engineering, Florida International University, Miami, Florida, United States, ²Mechanical and Materials Engineering, Florida International University, Miami, Florida, United States

Show Abstract

9:00 PM - DD5.12

A Novel Impedance Sensing Technique for Monitoring Spreading of Somatic Stem Cells.

Yiling Qiu ¹, Xin Zhang ¹
¹Mechanical Engineering, Boston University, Boston, Massachusetts, United States

Show Abstract

There is growing interest to cure myocardium damage with cardiac side population (CSP) cells, a candidate for somatic stem cells. Real-time and sensitive approaches are demanded for the study on cell-substrate interaction of stem cells due to its vital importance to proliferation and differentiation. In this work, we developed an impedance-based assay for automatically monitoring spreading of CSP cells cultured on a microfabricated impedance sensor. The electrode coverage of CSP cells was detected through our impedance-sensing technique and correlated to the cell spread. An impedance sensor was fabricated on glass slides pre-coated with indium tin oxide (ITO) to enable the electrical sensing as well as the experimental validation via optical observation. A precision LCR meter was used to measure the impedance at 2 kHz and 50 mV of excitation. Cell culture on the working electrode will influence on the testing current due to the dielectric of cell membrane. We can describe impedance components influenced by the cell culture with an equivalent circuit model in which the parallel resistance (Rc) is primarily determined by cell coverage on the electrode. Therefore, the Rc values would be separated from the impedance measurement and correlated with the electrode coverage area measured optically. Since Rc was used to describe resistance to the current flow that traveled directly from the working electrode to culture medium, it was expected that Rc values would be inversely proportional to the uncovered electrode area. The results of time-lapse photography and 1/Rc values presented a good linear correlation. The change of uncovered electrode area was primarily due to the significant spreading of CSP cells To get a quantitative measure for the dynamics of cell spread, the time necessary for half-maximum reduction of impedance (tZ50) was determined, and the results was 2.9 ± 0.3 h. In general, this work demonstrates the monitoring of CSP cell spread using our impedance-sensing assay. Further development of this assay is towards the high-throughput study of cell biology with efficient and inexpensive electrical sensing techniques.

9:00 PM - DD5.13

Electrodeposition of Fe-rich, FeNiCo Microcomponents for Recurve Micro-Bimorphs.

Elizabeth Podlaha ¹, Hana Kim ¹, Robert Simmons ², Michael Murphy ²
¹Chemical Engineering, Northeastern University, Boston, Massachusetts, United States, ²Mechanical Engineering, Louisiana State University, Baton Rouge, Louisiana, United States

Show Abstract

Recurve micro-bimorph elements, consisting of bi-layers of materials with low and high coefficients of thermal expansion (CTE) are of interest as microsensors and microactuations for the direct thermo-mechanical conversion of a thermal input into a deflection or force. Fabricating these elements require a lithography step followed by electrodeposition. One approach to fabricating these elements is a LIGA or UV-LIGA (a German acronym for Lithography, Electrodeposition, Molding) process, with a lithography step to create a pattern, electrodeposition of a metal to fill the pattern, and subsequent planarization of the metal layer. The sequence is repeated for each of the two materials in the bimorph. While most electrodeposited metals and alloys exhibit high CTE, few have low ones. Our work addresses the development of a low CTE, electrodeposited Super Invar-like alloy, having a composition near or at bulk conditions: Fe-rich, 64 wt % Fe, 31 wt % Ni, and 5 wt % Co. Two types of substrates were examined, i. an array of micro-posts, as a test sample, and ii. bimorph structure. Different pulse plating parameters were applied and the deposit composition distribution along the length of the microstructures was examined. While pulse plating (-60 mA cm-2, 2 s / 0 mA cm-2, 27 s) was critical in achieving a uniform composition distribution in 100 micron deep features, the processing time was considerably longer. Using a pulse scheme with a non-zero step increased the processing time but introduced more cracks into the deposit. The microcracks were characterized by SEM. The CTE of the micro-post arrays was also determined and exhibited negative values in the axial direction, unlike bulk equivalents, i.e. the micro-posts shrink with temperature. Structures with micron size cracks exhibited large, negative CTE values during the first heat cycle of the CTE test and then exhibited less dimensional change with a subsequent heat cycle. Although cracks are undesirable for the development of the bimorph, the CTE measurement can be used to evaluate the presence of cracks, and can be used as a guide to select the best plating conditions. A micro-bimorph structure was subsequently fabricated with crack-free electrodeposits.

9:00 PM - DD5.14

Package-level Multilayered Piezoelectric Micropumps for Lab-on-Chip.

Wenli Zhang ¹, Richard Eitel ¹
¹Chemical and Materials Engineering Department, University of Kentucky, Lexington, Kentucky, United States

Show Abstract

Efficient compact fluid control and transport remains a key challenge for the practical application of microfluidic systems in chemical analysis and synthesis, biomedical devices, biological sensing, and environmental monitoring. Micropumps with piezoelectric actuation offer an alternative to motor driven pumps and have the potential to be integrated into lab-on-chip microfluidic devices at the chip or package level. Current piezoelectric micropumps are fabricated using MEMS processing techniques often incorporating additional silicon/glass hybrid bonding methods. The piezoelectric actuator is mounted on a silicon or glass membrane in a similar manner-adhesive bonding which is critical for the reliability and yield of this type of actuator and not compatible with integrated fabrication.The major goal of current research is to design, fabricate and test a novel package-level multilayered piezoelectric micropump that can be easily processed and integrated into microfluidic devices and systems. The micropump mainly consists of three layers: the top and bottom layers with piezoelectric actuators on both sides of the pump chamber and an intermediate layer with microliquid conduits (microfluidic channels, liquid rectification elements, and pump chambers). This novel design would double the stroke volume without adding to the complexity of the micropump fabrication which exhibits the unique capability of the multilayer processing approach. All layers are fabricated using homogeneous piezoelectric thick films and then laminated and sintered to obtain the package-leveled micropump. A high-strain piezoelectric thick film tape material that can be sintered at low temperature (~900°C) with silver conductors has been developed using traditional solid state ceramics processing. The micropump performance is analyzed and evaluated by both numerical simulations and experimental results.The successful design, fabrication and optimization of this type of micropump will provide key functionalities required for future highly integrated microfluidic systems. The future work will optimize of processing and material properties of active layers to enable the integration of piezoelectric micropumps with commercial Low Temperature Cofired Ceramic (LTCC) materials and processes.

9:00 PM - DD5.15

Liquid Encapsulation by Bonding-in-Liquid Technique.

Yoshiyuki Okayama ¹, Takeshi Ninomiya ¹, Yasuyuki Matsumoto ¹, Norihisa Miki ¹
¹Mechanical Engineering, Keio University, Kanagawa Japan

Show Abstract

We propose and demonstrate Bonding-in-Liquid Technique (BiLT) for encapsulation of liquid in MEMS devices. Liquid encapsulation enables innovative MEMS devices by exploiting the virtues of liquid, such as hydraulic displacement amplification, deformable scanning mirrors, and dye-sensitized photovoltaic devices with liquid electrolytes. Interfusion of air bubbles, heterogeneity of the liquid quantity, and leakage of the encapsulated liquid must be averted not to deteriorate device performances. Several liquid encapsulation processes have been proposed, such as parylene deposition and polymer thermal bonding. However, they involve vacuum and/or thermal processes and cannot be applied to volatile liquids. In BiLT proposed herein, two structural layers are passively aligned and brought into contact in liquid, when the encapsulation cavities are uniformly filled up with the liquid without air bubbles. A UV-curable resin is used as an adhesive that does not require heat or vacuum environment but UV to bond the two layers. The detailed processes of BiLT are (a) UV-curable adhesive resin is coated onto the bonding surface. The layer contains not only encapsulating cavities but also concave-shaped structures for the following passive alignment. (b) The other layer with convex-shaped structures is brought into contact in liquid, when the two layers are passively aligned by matching concave and convex structures. (c) UV light is irradiated and the two layers are permanently bonded while the contact is maintained with a bonding jig. We successfully achieved encapsulation of DI water and glycerin in PDMS and silicon structural layers. No air bubbles were trapped inside the chamber and no liquid remained in the bonding interface. While conventional aligners are not applicable to BiLT, we experimentally evaluated the accuracy of the passive alignment process in liquid. We manufactured a PDMS layer with cylinders and a silicon layer with cavities. The height and depth of the cylinders and cavities were designed such that the PDMS cylinders elastically deformed when the two layers were brought into contact. The elastic averaging enabled the passive alignment of the two layers. We investigated the bonding accuracy with respect to the number of pairs of concave/convex structures and the height of PDMS cylinders while in contact. The bonding accuracy improved as the number of pairs increased while the height of PDMS cylinders did not show a strong correlation. The alignment accuracy of 6µm in BiLT was achieved with 12 pairs of the concave/convex structures. We manufactured a hydraulic displacement amplification mechanism by BiLT. Glycerin was encapsulated in silicon cavities by largely-deformable-PDMS thin membranes. The displacement at the input was successfully amplified at the output associated with the ratio of the cross-sectional areas and 10-folds of displacement amplification was successfully achieved.

9:00 PM - DD5.16

Deposition Profile Control of Carbon Films on Patterned Substrates using a Hydrogen-assited Plasma CVD Method.

Takuya Nomura ¹, Kazunori Koga ^{1 4}, Masaharu Shiratani ^{1 4}, Yuichi Setsuhara ^{2 4}, Makoto Sekine ^{3 4}, Masaru Hori ^{3 4}
¹Department of Electronics, Kyushu University, Fukuoka Japan, ⁴, JST CREST, Tokyo Japan, ²JWRI, Osaka University, Ibaraki, Osaka, Japan, ³Dept. of Electrical Engineering and Computer Science, Nagoya University, Nagoya, Aichi, Japan

Show Abstract

DLC is a term used to describe a variety of carbon based materials with attractive properties, such as a high mechanical hardness, chemical inertness, optical transparency and wide band gap [1,2]. DLC films have widespread applications as protective coatings in several areas such as car parts, biomedical coatings and micro-electromechanical systems (MEMS). Deposition profile of the DLC and a-C:H films in trenches is one of the concerns to realize coatings on patterned substrates. We have succeeded in controlling deposition profile of Cu in trenches, and have realized sub-conformal, conformal and anisotropic deposition profiles using a H-assisted plasma CVD method [3,4]. We are applying the method to plasma CVD carbon films. Here, we report dependence of deposition profile of plasma CVD carbon films in trenches on bias voltage, substrate temperature, and aspect ratio. Experiments were performed using a capacitively-coupled plasma CVD reactor. The excitation frequency of the main discharge was 28 MHz and the supplied power Pm below 45W. Toluene diluted with H₂ was supplied at a total flow rate of 90 sccm. The pressure was 13 Pa. First, we have studied effects of substrate bias voltage on deposition rate at the bottom of the trenches. Deposition does not take place without the bias voltage, whereas it does with the bias voltage from -10 to -50 V. These results indicate that ion induced surface reactions are the key to realizing the deposition and a kinetic energy of incident ions above 30 eV is needed to induce the reactions. Next, we have studied aspect ratio dependence of deposition rates at the bottom and sidewall of trenches. The deposition rates tend to increases from 7.39 nm/min to 16.7 nm/min with decreasing the aspect ratio from 4.2 to 1.1, because the incident radical flux per surface area in a trench increases with decreasing the aspect ratio. Finally, we have studied dependence of deposition rate on the substrate temperature. The deposition increases from 16.7 nm/min to 38.1 nm/min with decreasing the temperature from 220^oC to 100^oC probably because the etching rate of carbon films by H decreases with decreasing the temperature. At the substrate temperature above 250^oC, no deposition takes place at the sidewall of trenches. Based on the results, we have succeeded in realizing three deposition profiles of carbon films in trenches: subconformal, conformal, and anisotropic. [1] J. Robertson, Adv. Phys. 35, 317 (1986).[2] J. Robertson, Surf. Coatings Technol. 50, 185 (1992).[3] K. Takenaka, M. Kita, T. Kinoshita, K. Koga, M. Shiratani, and Y. Watanabe, J. Vac. Sci. Technol. A22, 1903 (2004).[4] K. Takenaka, M. Shiratani, M. Takeshita, M. Kita, K. Koga, and Y. Watanabe, Pure Appl. Chem. 77, 391 (2005).

9:00 PM - DD5.17

Fabrication of Multilayered Metal Oxide Hybrid Films by Layer-By-Layer Deposition of Inorganic Nanosheets.

Min-Sun Song ¹, Myong A Woo ¹, Tae Woo Kim ¹, Seong-Ju Hwang ¹
¹Department of Chemistry and Nano Sciences, Center for Intelligent Nano-Bio Materials (CINBM), Ewha Womans University, Seoul Korea (the Republic of)

Show Abstract

9:00 PM - DD5.18

Cholesterol Sensor Based on 3-Dimensional Porous Carbon Microelectrodes Fabricated by C-MEMS Technique.

Varun Penmatsa ¹, Jung-Hoon Yang ¹, Chunlei Wang ¹
¹, Florida International University, Miami, Florida, United States

Show Abstract

9:00 PM - DD5.19

Formation of Nanoparticle-Containing Multilayers in Nanochannels via Layer-by-Layer Assembly.

Jun Young Kim ¹, Jonathan DeRocher ², Pan Mao ³, Michael Rubner ¹, Robert Cohen ², Jongyoon Han ⁴
¹Department of Materials Science and Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States, ²Department of Chemical Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States, ³Department of Mechanical Engneering, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States, ⁴Department of Electrical Engineering and Computer Science and Department of Biological Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States

Show Abstract

9:00 PM - DD5.2

Characterization and Modeling of Electroporation Parameters of a Micro-Electroporation Biochip.

Aeraj Haque ^{1 4}, Mahvash Zuberi ^{1 3 4}, Ruben Diaz-Rivera ^{5 2}, David Porterfield ^{1 2 4}
¹Agricultural and Biological Engineering, Purdue University, West Lafayette, Indiana, United States, ⁴Bindley Bioscience Center, Purdue University, West Lafayette, Indiana, United States, ³Center for Paralysis Research, Purdue University, West Lafayette, Indiana, United States, ⁵Mechanical Engineering, University of Puerto Rico at Mayagüez, Mayaguez, Puerto Rico, United States, ²Weldon School of Biomedical Engineering, Purdue University, West Lafayette, Indiana, United States

Show Abstract

Advancements in micro and nano fabrication have lead to the development of planar micro-pore electroporation technology. This technology is the focus of considerable research because of the greater control it provides through single cell manipulation and independence from cell size. In this work we for the first time report on the electrical characterization of a planar micro-pore single cell electroporation biochip in order to better understand this phenomenon. The microchip was fabricated on a 1µm thick Silicon Nitride (Si₂N₃) membrane deposited through a chemical vapor deposition process on a silicon substrate. 2-3 µm diameter micropores were fabricated on this membrane through a reactive ion etching process. The microchip is integrated with a custom manufactured polycarbonate packaging to yield a compact device for studying single cell electroporation on a planar biochip. Electrical characterization of electroporation pulses was performed around this pore using the 2 dimensional Vibrating Probe (VP) technology. Spatial current density measurements were determined to follow a 4th order polynomial profile in the radial direction and a three parameter hyperbolic decay profile in the z-plane. A multiphysics simulated model was developed on COMSOL FEMLAB which correlates with actual measurements on the micropore. VP measurements and the model reveal a very high current density close to 9600000 uA/cm² concentrated at the micropore which rapidly degrades even at 20 microns away from the pore site. This electric field concentration achieved in micro-electroporation is the key behind targeted electroporation, minimum collateral damage and high cell viability. Preliminary current density measurements around electroporated HepG2 cells show that membrane electrophysiological recovery is a transient process spanning a time period in minutes to hours. This work not only showcases the first experimental characterization of a micro-electroporation biochip but also demonstrates the application of the 2-D VP technology as an effective MEMS characterization tool. Work is under way to apply the vibrating probe technique as a cell viability indicator as a measure to attain optimized single cell electroporation.

9:00 PM - DD5.20

Control and Operation Schemes for Micro-Thermal Conductivity Detectors in Gas Chromatography.

Bradley Kaanta ¹, Hua Chen ², Xin Zhang ¹
¹, Boston University, Boston, Massachusetts, United States, ², Schlumberger Doll Research, Cambridge, Massachusetts, United States

Show Abstract

As the use of sensor networks has expanded, the demand for robust detectors able to operate in a variety of environments has grown. This presentation explores the fabrication and testing of a micro thermal conductivity detector (uTCD), which can be integrated into a micro-gas chromatography (GC) system, operating in constant temperature and constant voltage modes. The sensitivity of a single uTCD can be increased by a factor of 10 simply by using a different operational method. This microfabricated device is composed of a resistive heating element suspended in a micro-channel, which creates excellent thermal isolation and a high heat transfer coefficient between the element and the fluid. The resistive element is composed of thin film nickel, which exhibits a high coefficient of thermal resistance, allowing for accurate detection and control of device temperature. Gas chromatography systems use a detector to identify and quantify gases in a sample. Such systems can be utilized in the real-time testing of gases inside industrial equipment, improving efficiency and reducing pollution, or placed underground, to reduce the environmental impact and expense of resource exploration. A thermal conductivity detector is an excellent choice for miniaturization because, as the dimensions of a GC system are decreased, the amount of sample being processed is reduced by a power function with volume. For any detector which is mass sensitive, such as a flame ionization detector or a mass spectrometer, the output signal is reduced directly with sample volume. In contrast, a uTCD is sensitive only to the concentrations of substances within a mixture. Therefore, volume reduction has no effect on the output. TCDs measure the different physical properties of a fluid as it flows over an electrically heated resistive element; the fluid’s properties can be determined by using one of two different operating modes. In constant voltage mode the output of the system is related to the temperature change of the resistive element. As a less thermally conductive fluid is exposed to the active device, less heat is transported away, increasing the filament temperature. In constant temperature mode, the filament is maintained at a fixed operating temperature, using feedback. The change in power required to maintain this fixed temperature is measured. Physical testing and theoretical models have shown that constant temperature uTCD operation is on the order of 10 times more sensitive, as well as having a broader linear operational region than in a constant voltage control mode. However, constant voltage control allows for easier elimination of common mode electrical noise, which aids in lowering the device’s detection limit. These results concur with previously published results for macro-scale devices. The control method utilized could be selected depending on the operating requirements of a fielded system.

9:00 PM - DD5.21

Defining Fine Pitch through-via Structures on Flexible Dielectric Substrates using a Laser Ablation Technique.

Pavan Samudrala ², Aparna Prabhakar ¹, Robert Trzcinski ¹
², L J Gonzers Associates, Cranford, New Jersey, United States, ¹, IBM, Yorktown Heights, New York, United States

Show Abstract

9:00 PM - DD5.24

Fabrication of Microfluidic Hybridisation Applicator in Thermalplastic Elastomeric Materials for DNA Detection on BioFET Array.

Kebin Li ¹, Matthias Geissler ¹, Liviu Clime ¹, Emmanuel Roy ¹, Teodor Veres ¹
¹, Industrial Materials Institute, Boucherville, Quebec, Canada

Show Abstract

Medical diagnostics is an area of great interest that has recently fuelled an increased interest for the development of disposable rapid diagnostics devices for point-of-care applications. Most of the microfluidic devices are fabricated using either `hard lithography` in semiconductors or glass or `soft lithography` in polymers such as PDMS [1]. It is highly demanded on fabrication technologies and materials that can provide rapid and low-cost fabrication, simples procedures for bonding as well as stable surface treatments for biomolecules capture and immobilization in order to commercialize the lab-on-chip systems (LOC). On the other hand, the use of DNA microarrays[2,3] based on multiplexed fluorescence measurements has revolutionized the practice of life sciences research, providing quantitative information on complex biological systems and many applications such as gene expression analysis or disease diagnostics. The integration of the microarray technologies with microfluidics systems can offer the possibility to dramatically reduce the amounts of costly solutions, DNA probes, proteins or other molecules to be used but also provide the prospect for their integration into systems for electrical based detection technologies. Thermoplastic elastomer (TPE) has been recently demonstrated to be biocompatible and surface stable materials that allow to rapidly fabricate complex microfluidic devices at relatively low cost because of its remarkable advantages in the soft fabrication and microfluidic behavior[3,4,5]. Herein we will report on the fabrication of the TPE based microfluidic devices by using the hot-embossing technique, the fabricated microfludic devices will allow a fast hybridization of DNA probes immobilized on surface of BioFET arrays. To validate the performance of the devices, DNA hybridization performance testing were carried out on thermal plastic substrates. Three types of DNA probes (P1 and P2 were labelled with Cy5 and P3 was labelled with Cy3) were firstly spotting on Zeonor substrate with line width of 50um. Hybridization was done under the microfluidic environment with shuttling functionality, the target DNA is complementary to P1 only the concentration of the target DNA used is in 10nM, it shows that the uniform hybridization process could take place within 5 minutes under the microfluidic environment. 1. Y.N. Xia et al., Ann. Rev. Mater. Sci. 1998, 28 :1800 2. A. Marshall et al. Nat. BioTechnol., 1998, 16, 27-31.3. M. Dufva, Biomol. Eng., 2005, 22, 173–184.4. E.Roy, M. Geissler, J. C. Galas, G. A. Diaz-Quijada and T.Veres ( Submitted Lab on Chip 209)5. M. Geissler, E. Roy, G.A. Diaz-Quijada, J. C. Galas, and T. Veres (Accepted , ACS Applied Materials & Interfaces). 6 Maxime D. Guillemette et al., Integr. Biol., 2009, 1, 196 – 204.

9:00 PM - DD5.25

Characterization of Direct Functionalization on 3D Pyrolyzed Carbon for DNA Detection.

Jung-hoon Yang ^{1 2}, Penmatsa Varun ¹, Hiroshi Kawarada ², Chunlei Wang ¹
¹Department of Mechanical and Materials Engineering, Florida International University, Miami, Florida, United States, ²Consolidated Research Institute for Advanced Science and Medical Care, Waseda University, Tokyo Japan

Show Abstract

9:00 PM - DD5.26

Simulation of Microfluidics and Quasi-statics of an Integrated BioMEMS Device.

Divya Padmaraj ^{1 2}, Wanda Zagozdzon-Wosik ¹, Rohit Pande ¹, John Miller ³, Lei-Ming Xie ², Piotr Grabiec ⁴
¹Electrical and Computer Engineering Department, University of Houston, Houston, Texas, United States, ²Texas Center for Superconductivity, University of Houston, Houston, Texas, United States, ³Physics Department, University of Houston, Houston, Texas, United States, ⁴, Institute of electron technology, Warsaw Poland

Show Abstract

A BioMEMS chip was designed and fabricated for monitoring mitochondrial parameters to study its functionality. The system was first simulated using COMSOL Multiphysics 3.5a and then fabricated as an SU-8 based microfluidic. In order to characterize the mitochondrial membrane parameters using fluorescent probing, 3 flow cytometers were designed: a 30 µm and a 15 µm channel width cytometer, and a 20 µm channel width cytometer with electrodes. Two focusing stages were used to prevent analyte diffusion. An impedance cytometer with vertical side-wall electrodes was designed to perform dielectric characterization to study the mitochondrial membrane potential. In bioimpedance measurements of cells' suspensions, electrical double layer of electrodes plays a critical role in producing artifacts observed at low frequencies. Therefore, for parallel dielectric spectroscopy studies of mitochondria, a 4 electrode array (600 µm x 50 µm) was designed with meshed pickup electrodes in which the two inner electrodes consisted of a mesh with square holes 7.5 µm by 7.5 µm in size. Due to the decrease in contact area, polarization effects seen at low frequencies were reduced using the mesh design.Microfluidic simulations showed analyte focusing, together with the influence of the second focusing inlets in avoiding diffusion. Simulations of the vertical sidewall electrodes showed an improvement in field uniformity, as the electrode height matched the channel height. In the case of the meshed electrode, the presence of dielectric in the mesh holes and the mesh design influences the electric field and charge distribution, introducing non-uniformity at a micro-scale. Due to charges around the mesh holes, x-component of the E-field was small at inner holes than outer holes. Mesh electrodes result in smaller polarization effect while maintaining similar electrical properties (E Field, potential etc.) at macro-scale, including dependence of these parameters on distance from the electrodes. We present a mathematical model of the double layer incorporating permittivity variations in its different regions. This was used in COMSOL simulations of electrode polarization. In COMSOL simulations, we also included analogous surface polarization and diffused layers for cells in our biosamples where their non-spherical shape and small size enhance interfacial effects.

9:00 PM - DD5.27

Micro Methanol Steam Reformer based on Cu/ZnO Nanostructure Catalysts.

Chan Hyeok Yeo ¹, Dae Hyeon Choi ¹, Kwang Woo Kim ¹, Won Seok Chang ², Yong Chai Kwon ², Yeon Ho Im ^{1 2}
¹Department of Hydrogen and Fuel Cells Engineering, Specialized Graduate School, Chonbuk National University, Jeonju Korea (the Republic of), ²School of Semiconductor and Chemical Engineering, Chonbuk National University, Jeonju Korea (the Republic of)

Show Abstract

9:00 PM - DD5.28

Chemical Transformation of Bimetallic ``Janus" Particles.

Shengrong Ye ¹, R. Lloyd Carroll ¹
¹Chemistry, West Virginia University, Morgantown, West Virginia, United States

Show Abstract

9:00 PM - DD5.29

A Low-Range Drift-Free Bio-compatible Pressure Sensor Based on P(VDF-TrFE) Piezoelectric Thin Film.

Xiaoyang Li ¹, Timothy Reissman ², Fan Yu ¹, Edwin Kan ¹
¹School of Electrical and Computer Engineering, Cornell University, Ithaca, New York, United States, ²School of Mechanical and Aerospace Engineering, Cornell University, Ithaca, New York, United States

Show Abstract

A low-range highly sensitive pressure sensor (0-100kPa) based on the P(VDF-TrFE) piezoelectric thin film is proposed, where the long-term drift is eliminated by operating near the piezoelectric resonance. The pressure sensor is designed for blood pressure and body tissue swelling pressure monitoring. The poled 50µm±1µm P(VDF-TrFE) copolymer thin film has been used as the sensing element, with all fabrication and assembly materials biocompatible.A modified Butterworth-Van Dyke (BVD) [1] equivalent circuit model is used to characterize the sensor behavior. A series capacitor is used with the resonator circuit to enhance the sensitivity. The simulations for device performance and electromechanical analysis are performed with the BVD model to compare with measurement. The resonance frequencies are found to agree with theoretical simulation. The pressure sensor exhibits non long-term drift in weeks of operation. The device shows a sensitivity of 0.038MHz/kPa resonance frequency shift under stress. Furthermore, the pressure transducing is performed at a fixed frequency around the 3rd harmonic by measuring the RMS voltage magnitude, which has larger mass loading effect than the fundamental resonance frequency. With the new measurement setup, it leads to a maximum readout change of 1.1%/kPa at the narrow band around 3rd harmonic mode. A low-power system-on-chip integration solution with chemoreceptive MOS transistors (CνMOS) has been proposed. Measuring the AC voltage magnitude requires less extensive circuits than impedance analysis, which also provides an easier solution for on pixel pressure sensor integration. CνMOS satisfies this requirement with thin films directly deposited on the extended floating gate, and provides the first-stage low-noise transconductance amplification required for further processing [2].Background: In vivo pressure sensing has many critical applications in biomedical sensors and human-machine interface, which require high sensitivity in the 0-100kPa range, no long-term drift and robust integration platform. The capacitive pressure sensor in this range is fragile, vulnerable to external shock, and requires external reference pressure. Piezoresistive pressure sensor is limited for its low sensitivity. P(VDF-TrFE) is preferred for its high charge constant, flexible thin film structure, low modulus, high yield strength and resistance to corrosive chemicals.[1] S. J. Martin, et al., Anal Chem, 63, 20, 1991[2] B. Jacquot, et al., MRS Symp. Proc. Vol. 952, 2007

9:00 PM - DD5.3

Densely-Packed Microbowl Array with Balanced Dielectrophoretic Forces for Single-Cell Microarray.

Maesoon Im ¹, Dong-Haan Kim ¹, Joo-Hyung Lee ¹, Jun-Bo Yoon ¹, Yang-Kyu Choi ¹
¹Division of Electrical Engineering, School of Electrical Engineering and Computer Science, KAIST, Daejeon Korea (the Republic of)

Show Abstract

Since cellular responses are heterogeneous even under identical environmental conditions, many cell biologists want to analyze a large number of individual cells at once to get stochastic distributions of the cellular responses. For a high-throughput single-cell analysis, microfabrication techniques have been used to implement various platforms such as a single-cell microarray [1], which consists of microchambers accommodating single cells; a dielectrophoretic cell-trapping array [2], a microfluidic device for single-cell assay [3] and so on. However, in aforementioned techniques, there are disadvantages including slow and less-efficient cell-capture by gravitational force [1], possible cell-movement after disconnecting electrical signals [2], and necessity of microfluidic components [3]. For high-throughput and ease of use, it is essential to develop a new array-based technique to confine single cells with higher density as well as without any microfluidic component.In this paper, we demonstrate a perfectly-ordered microbowl array with balanced dielectrophoresis (DEP) for a high-throughput single-cell analysis. In order to fabricate well-ordered microbowl array in a large area, we utilized three-dimensional diffuser lithography [4] for photoresist mold, and nickel electroplating technique for final microbowl structures on a silicon substrate. Single microbowl has six sharp apexes surrounding the microbowl perimeter [4], a diameter of 10 μm, and a height of 9 μm, which can be controllable by patterns on mask and lithography conditions. To investigate the feasibility for application to the microbowl array as a single-cell microarray, we used latex beads of 6.4 μm in diameter to be captured by dielectrophoretic force. The nickel microbowl array densely packed with a hexagonal geometry played as a bottom electrode, and an ITO-coated glass covered the nickel microbowl array as a top electrode while keeping a uniform gap between two electrodes. After injecting solution containing latex beads through the gap, we applied an AC signal (2 V, 10 kHz) between two electrodes to induce high electric field near the sharp apexes of the single microbowl. A negative DEP trap is formed at the center of the single microbowl with balanced DEP force from the six apexes. The experimental result shows that injected latex beads had been successfully and uniformly aligned and trapped at the microbowl array sustained by negative DEP. With the aid of the microbowl array, high-throughput single-cell analysis is expected with high density of single cells as well as increased efficiency of capturing of cells by DEP force guidance. [1] S. Yamamura et al., Analytical Chemistry, 77 (2005) p.8050.[2] B.M. Taff and J. Voldman, MicroTAS2005, (2005) p.865.[3] K.-S. Yun and E. Yoon, Biomedical Microdevices, 7, (2005), p.35.[4] X.-J. Huang et al., Small, 5 (2009), p.90.

9:00 PM - DD5.30

Sputtered SiO2 Film on Cyclic Olefin Copolymer (COC) for Surface Functionalization and BioMEMS Applications.

Kuosheng Ma ¹, Ishtiaq Saaem ¹, Jingdong Tian ¹
¹, Duke University, Durham, North Carolina, United States

Show Abstract

9:00 PM - DD5.31

Towards NEMS Fluid Sensors Based on Suspended Nanomaterials.

Luis Jauregui ^{1 3}, Hao Lu ^{4 5}, Jun Lou ^{4 5}, Yong Chen ^{2 3 1}
¹School of Electrical Engineering , Purdue University, West Lafayette, Indiana, United States, ³Birck Nanotechnology Center, Purdue University, West Lafayette, Indiana, United States, ⁴Department of Mechanical Engineering & Materials Science, Rice University, Houston, Texas, United States, ⁵Richard E. Smalley Institute for Nanoscale Science and Technology, Rice University, Houston, Texas, United States, ²Department of Physics , Purdue University, West Lafayette, Indiana, United States

Show Abstract

9:00 PM - DD5.32

Grayscale Lithography Based on Chalcogenide Glass Photoresists for Microsystems Applications.

Andriy Kovalskiy ¹, Jiri Cech ¹, Miroslav Vlcek ^{2 1}, Himanshu Jain ¹, C. Mike Waits ³, Madan Dubey ³
¹Center for Optical Technologies, Lehigh University, Bethlehem, Pennsylvania, United States, ²Department of General and Inorganic Chemistry, University of Pardubice, Pardubice Czechia, ³, Army Research Laboratory, Adelphi, Maryland, United States

Show Abstract

9:00 PM - DD5.33

Integrated Transducer and Surface Modifications on SAW Device for Enhanced Biosensing.

Reetu Singh ¹, Subramanian Sankaranarayanan ², Venkat Bhethanabotla ¹
¹Chemical and Biomedical Engineering, University of South Florida, Tampa, Florida, United States, ²School of Engineering and Applied Sciences, Harvard University, Cambridge, Massachusetts, United States

Show Abstract

Challenges relating to biomolecular sensing include high sensitivity for early detection and engineering high device specificity to avoid non-specific interactions that can impede accurate detection. The latter issue presents a significant impediment to the development of efficient sensors and accurate detection of pathologies. Although biological sensors have specific interactions that are used to detect the target analyte, they also have non-specific interactions with undesired proteins which manifest themselves as bifouling. In this work, we have investigated combined transducer and delay path modifications in a Langasite (LGS) based surface acoustic wave (SAW) device, using finite element structural and fluid-structure interaction simulations. The combined transducer and delay path modifications are expected to improve sensor performance via enhanced sensitivity, reduced power loss, and increased biofouling removal efficiency, compared to a standard unidirectional SAW device.Microcavities were etched in the delay path of a micron sized LGS SAW device with mutually interacting, orthogonal IDTs. We show different acoustic modes propagating in the two orthogonal directions, which can be selectively exploited for biosensing and acoustic streaming induced biofouling elimination. Our computations reveal that the shear horizontal mode along (0, 22, 90) allows for biosensing and the mixed mode with prominent surface normal component along (0, 22, 0) allow for acoustic streaming which can overwhelm the weak non-specific adhesive force to eliminate biofouling. The mutually interacting orthogonal IDT configuration gives rise to constructive wave interference between the interacting waves, which translates to higher surface displacements and therefore higher device sensitivity as well as higher acoustic streaming force. The incorporation of polystyrene filled microcavities in the delay path leads to acoustic energy confinement near the surface, further enhancing the device sensitivity and intensity of acoustic streaming. Thus, we show that the device with integrated mutually interacting IDTs and surface modifications in the form of microcavities provides superior performance compared to the standard device with unidirectional IDTs and no microcavities. The results are of tremendous importance in the development of efficient and reliable MEMS based biosensors.

9:00 PM - DD5.34

Formation of Hierarchical Structures of Molecular Assembly with Nanotube Structures by Dielectrophoreicsis.

Shigenori Fujikawa ^{1 2}, Wakana Kubo ², Hiromi Takemoto ¹
¹Interfacial Nanostructure Research Lab., RIKEN, Wako, Saitama, Japan, ², JST, CREST, Wako, Saitama, Japan

Show Abstract

9:00 PM - DD5.35

Integrity of Small-Scale Systems at High Temperatures and in Reactive Environments.

Yiu Wai Lai ¹, Robert Zarnetta ¹, Sven Hamann ², Michael Ehmann ², Robert Meyer ², Hayo Brunken ², Alfred Ludwig ^{2 1}
¹Research Department “Integrity of Small-Scale Systems and High-Temperature Materials (IS3/HTM), Ruhr-University Bochum, Bochum Germany, ²Institute for Materials, Ruhr-University Bochum, Bochum Germany

Show Abstract

The increasing trend for miniaturization in technology requires highly integrated small scale multifunctional devices which at the same time ensure integrity (high reliability, constant functionality over lifetime) even in harsh environments like high temperature levels or high mechanical loads. For advanced applications such devices have to work in non vacuum environment, i.e. reactive atmospheres, which increase the requirements for the used materials and the system design. This contribution discusses integrity issues of small-scale devices on the examples micro-hotplates (MHP) and multifunctional sensors. Micro-hotplates are used at high temperatures both as gas sensors and small scale research platforms for materials research [e.g. 1]. They are fabricated by micro-machining processes and consist of several layers of very different materials (Si, SiO2, Si3N4, Pt, functional thin films). The extreme temperature rates which can be achieved for heating (up to 1000°C) and cooling with MHPs already lead to high thermo-mechanical loads, since thermal stresses develop, due to the mismatch of the thermal expansion coefficient of the different materials. The system has to withstand these stresses without losing its integrity. Only when integrity is secured these devices can be used in advanced applications, e.g. for investigating the fatigue behaviour of thin films deposited on their heater area. In order to ensure the integrity of MHP their electro-thermo-mechanical behavior was characterized and analyzed. Furthermore, concepts for in and ex situ integrity surveillance functionalities in MHPs will be presented. In this context both thin film nanocomposite and laterally micro-structured magneto-elastic thin film sensor materials will be addressed. These materials are intended to integrate e.g. magneto-elastic sensor functionalities with wear resistance and high temperature compatibility.Acknowledgments: Financial support from the Research Department IS3/HTM at the Ruhr-University Bochum and from the DFG (Heisenberg programme, SPP1239 & 1299) is acknowledged.References: S. Hamann, M. Ehmann, S. Thienhaus, A. Savan, A. Ludwig (2008), Micro-hotplates for high-throughput thin film processing and in situ phase transformation characterization, Sensors and Actuators A, 147, 576 - 582.

9:00 PM - DD5.4

Advanced Wet Etch Bulk Micromachining in {100} Silicon Wafers.

Prem Pal ¹, Kazuo Sato ¹
¹Micro & Nano Systems Engieering, Nagoya University, Nagoya Japan

Show Abstract

Most micromachined structures applied in MEMS/MOEMS are fabricated on {100} oriented silicon wafers. Other orientations are used for specific applications. Although numerous wet anisotropic etchants have been and are used currently, tetramethyl ammonium hydroxide (TMAH) has gained wide popularity due to its high selectivity with silicon dioxide, and complementary metal-oxide-semiconductor (CMOS) process compatibility. Recently, the addition of surfactants (e.g. NC-200, Triton, PEG, ASPEG, etc...) to TMAH has been reported by several research groups. TMAH solution provides different etching characteristics when surfactant is added into it. Particularly in high concentration TMAH (25 wt%), a small amount of surfactant (0.01-0.1% v/v) suppresses the etch rates of {110} and its vicinal planes to a significantly low level, resulting in a reduction of undercutting at convex and rounded-concave corners.In this work, we have developed some microfabrication techniques by combining wet anisotropic etching and local oxidation of silicon (LOCOS) processes to perform advanced bulk micromachining in {100}Si wafers for the realization of MEMS structures with new shapes. The etching is performed in two steps in pure and surfactant (Triton X-100) added 25 wt% TMAH solutions. Two types of structures (fixed and freestanding) are fabricated. In the case of fixed structures, various types such as circular and square ashtray shapes, concentric rectangular and circular corrugations, etc. are fabricated. These structures comprise perfectly sharp cross sectional shapes. The concentric rectangular and circular corrugations are promising structures for the formation of corrugated diaphragm of P+-Si (or other materials) for the realization of MEMS based devices such as capacitive pressure sensor, microphone, etc. Thermally grown silicon dioxide (SiO2) is used for the fabrication of different shapes freestanding structures such as microfluidic channels.The P-type CZ grown of three inch diameter of {100} orientated silicon wafers are used to measure the etching characteristics and for the fabrication of advanced MEMS structures in pure and surfactant added TMAH. The 25 wt% TMAH and Triton X-100 [C14H22O(C2H4O)n, n = 9-10] are used as main etchant and surfactant, respectively. The 0.1% v/v Triton X-100 is used to prepare the surfactant added TMAH solution. Present research has wide scope including, but not limited to, the realization soft MEMS structures (e.g. PDMS (polydimethylsiloxane)), corrugated diaphragms, compact size liquid (or gas) delivery systems, overhanging microfluidic channels, etc. It enables to extend the range of 3D structures fabricated by silicon anisotropic etching. The proposed silicon micromachining is fully compatible to CMOS process as it utilizes pure and non-ionic surfactant added TMAH solutions.

9:00 PM - DD5.5

Microscale Patterning of Single Crystal Diamond Using Sidero-Metal/Diamond Thermochemical Reaction.

Yuko Morofushi ¹, Hajime Matsushita ¹, Norihisa Miki ¹
¹Mechanical Engineering, Keio University, Kouhoku-ku Yokohama, Kanagawa Japan

Show Abstract

In this paper we propose and demonstrate micro patterning processes of single crystal diamond using thermochemical reaction of diamond with a sidero-metal and elucidate the reaction involved. Single crystal diamond processes a variety of excellent characteristics, such as hardness and wear resistance, and hence, is expected to be a new material for not only micro machining tools but also innovative micro devices. It is mandatory to develop a patterning process of diamond with high precision and at low cost. Laser processing is currently widely used, but it is a serial process and costly. Film deposition and plasma etching are other effective methods while they are time consuming. Thermochemical reaction between diamond and sidero-metals is well known in the field of mechanical machining. Diamond tools cannot machine sidero-metals, such as iron, nickel, and cobalt, when the diamond tools wear instead of the sidero-metals. We used this reaction to micro pattern single crystal diamond. We used nickel as the sidero-metal that was patterned either directly on a bulk of single crystal diamond or on a silicon substrate. We term the former and the latter processes as direct and indirect patterning processes, respectively. In the indirect patterning a bulk of single crystal diamond was placed on the substrate. The pattern was negatively transferred to the diamond after thermal treatment in the air at ~1000K in both processes. The sidero-metal layer can be patterned by photolithography, which enables precise manufacturing and mass production. The direct and indirect patterning achieved etching rates of ~0.5µm/min and ~0.2µm/min, respectively, both of which increased as the annealing temperature increased while the indirect patterning did not require micro patterning of photoresist on a bulk diamond several millimeters squire and involved much less difficult processes than the direct patterning.The thermochemical reaction is reported to be caused by; diffusion of carbon from diamond into the sidero-metal, oxidation-reduction reaction between diamond and the metal, oxidation of diamond, and carbide formation. Carbide formation was not observed when we used nickel as the sidero-metal. When the direct and indirect patterning was conducted in nitrogen, the pattern transfer was observed but the etching rate was extremely low. Therefore, oxidation-reduction reaction is dominant in the direct patterning. In the indirect patterning, the etching continued even after the nickel and diamond was not in contact. The thickness of removed diamond was by far greater than the thickness of the nickel layer. Hence, we consider that diamond was etched by oxide-reduction reaction between the diamond and the metal while they were in contact at the beginning and then, the oxidation of diamond became dominant in the indirect patterning.The process proposed herein is readily applicable to manufacture micro devices that exploit excellent characteristics of single crystal diamond.

9:00 PM - DD5.6

Synthesis and Characterization of Nanocomposite Thin Films for MEMS Applications.

Kaushik Das ¹, Cheol Park ^{2 3}, Pascal Hubert ¹, Srikar Vengallatore ¹
¹Mechanical Engineering, McGill University, Montreal, Quebec, Canada, ², National Institute of Aerospace, Hampton, Virginia, United States, ³Mechanical and Aerospace Engineering, University of Virginia, Charlottesville, Virginia, United States

Show Abstract

9:00 PM - DD5.7

Fabrication of Microfluidic Actuating Device by Thermoset Polyester.

Young Seuck Yoo ^{1 3}, Si-Hyeong Cho ¹, Hyung-Bum Kang ¹, Yoon Sok Park ³, Jae Woo Joung ³, Jin-Goo Park ^{1 2}
¹Bio-Nano technology, Hanyang University, Ansan, Gyeonggi-do, Korea (the Republic of), ³Inkjet Business Group, Samsung Electro Mechanics, Suwon, Gyeonggi-do, Korea (the Republic of), ²Materials Engineering, Hanyang University, Ansan, Gyeonggi-do, Korea (the Republic of)

Show Abstract

9:00 PM - DD5.8

A Technique for 1-10 nm Planar Nanofluidic Channel Fabrication.

Chunrong Song ¹, Pingshan Wang ¹
¹, Clemson University, Clemson, South Carolina, United States

Show Abstract

Planar nanofluidic channels are powerful tools for the investigation of materials under confinement, such as confined water, DNA and proteins. The channels are also essential components in nano-fluidic systems. Different techniques, such as wet-etch and dry etch of silicon and silicon dioxide thin films, have been developed to fabricate trenches of ~ 20 nm or deeper. Then wafer bonding is generally used to form the desired nanofluidic channels. Nano-printing techniques and Poly Dimethyl Siloxane (PDMS) are also used for nanochannel fabrications. Despite the progress, there is still a lack of approaches for 1-10 nm nanofluidic channel fabrication. In this paper, we report a novel, yet simple, technique to fabricate 1-10 nm planar nanochannels.The technique etches native oxide on silicon surface to form nano-trenches of desired depth. Then wafer bonding is used to form 1-10 nm nanofluidic channels. The fabrication process starts with a silicon wafer with patterned spin-on photo-resist as native SiO2 etch mask. Wet HF etch is used to remove native SiO2 layers. Due to the limited native SiO2 thickness, multiple HF etch steps are used to achieve different channel depth. Trench depths from ~ 3 nm to 15 nm were obtained. The trenches are uniform across the wafer. It is found that each HF dip increases trench depth by ~ 1 nm. Surface roughness of the trenches was measured with atomic force microscopy (AFM). An RMS roughness of ~ 0.6 nm is obtained. The roughness does not deteriorate when deeper trenches are formed with multiple HF dips. Additionally, silicon wafer surfaces, hence roughness, are not affected in the fabrication process. Wafer surface conditions are critical for Si-Si, Si-glass (quartz) wafer bonding and nanofluidic channel formation. TetraMethyl Ammonium Hydroxide (TMAH) etch of Si are also used to obtain nano-trenches. The obtained smallest trench depth is about 6 nm with TMAH etch procedures. It is difficult to control shallow trench depth and uniformity due to its start-up delay. The obtained surface roughness is ~ 0.6 nm to 1.1 nm. Nanochannels are then obtained through Si-Si fusing bonding and Si-SiO2 fusion bonding. Water wetting of the obtained nanofluidic channels is used to confirm that the channels are not collapsed. The developed nanochannels are used to study confined DI water properties.

9:00 PM - DD5.9

Microwave Absorbing Ferrite Thin Films for Microwave Heating of Microstructured Reactors.

Pengzhao Gao ¹, Evgeny Rebrov ¹, Jaap Schouten ¹, Richard Kleismit ², John Cetnar ², Guru Subramanyam ³, Gregory Kozlowski ²
¹Chemistry and Chemical Engineering, Eindhoven University of Technology, Eindhoven Netherlands, ²Physics, Wright State University, Dayton, Ohio, United States, ³Electrical and Computer Engineering, University of Dayton, Dayton, Ohio, United States

Show Abstract

Introduction. The concept of microwave heating enables optimal control of the rate of heat transfer in a microstructured reactor by eliminating conductive and/or convective heat transport resistances. In this way the controlled heating of a microreactor can be achieved using selective microwave absorption into a functional coating deposited on the microchannel walls. Ni-Co or Ni-Zn spinel type ferrites are possible candidates for microwave heating due to their moderate dielectric losses in the range between 2 and 15 GHz as well as large magnetic losses. Until now, most of ferrite films have been prepared using sputtering and PLD techniques. In this study, we applied a sol–gel method which allows to obtain chemical homogeneity, provides easy adjustment of chemical composition, and has relatively low cost. Synthesis. Required amounts of Fe(NO₃)₃, Ni(NO₃)₂, and Zn(NO₃)₂ were dissolved in ethanol with a subsequent addition of citric acid. The resulting sol was dropped on a 1x1 cm² Si substrate, which was spin-coated under a relative humidity of 60% for 25 s at 3500 rpm. Drying was performed in an oven at 80^oC for 30 min, which was followed by calcinations at 873 or 1073 K for 1 or 3 h, with a heating rate of 2 K/min. The films with the following compositions have been prepared: NiFe₂O₄, CoFe₂O₄, Ni_0.5Zn_0.5Fe₂O₄, Co_0.2Ni_0.4Zn_0.4Fe₂O₄, Co_0.4Ni_0.3Zn_0.3Fe₂O₄.Characterization. The crystalline structure of the films was determined by XRD in the 2θ range from 20^o to 60^o. CoFe₂O₄ films calcined at 1073 K have a cubic inverse-spinel structure with a Oh7 Fd3m space group with a unit cell parameter of 8.377Å. The composite zinc ferrite films have a cubic spinel structure with a unit cell parameter of 8.339Å. The density of the thin films is close to that of the corresponding powders. A minor amount of the iron oxide (Fe₂O₃) phase was observed in the samples calcined at 873 K. The morphology of the films was investigated with SEM and AFM. The average grain size and the grain size distribution were determined from the AFM image analysis. The average grain size increases from 30 to 68 nm with increasing duration of the calcination. The relative complex permittivity of the films was studied in the 2-10 GHz range at 295, 315, 335, and 355 K. The real part declines from 3 to 2 and the imaginary part increases from 0.2 to 1.1 as the frequency increases from 2 to 10 GHz at 295 K. The loss tangent increases from 0.066 to 0.491 with increasing the frequency and it decreases from 0.066 to 0.014 as the temperature increases, respectively. The heating behavior of the synthesized materials was tested in a multimode microwave cavity operated at 2.45 GHz using an infrared detector for temperature measurement. The Ni_0.5Zn_0.5Fe₂O₄, films demonstrated the highest microwave absorption corresponding to an initial heating rate of 12.5 K/min. The microwave absorption of the Co-containing films was lower at this particular frequency.

2009-12-02 Show All Abstracts

Symposium Organizers

Joerg Bagdahn Fraunhofer Center for Silicon Photovoltaics

Norman Sheppard MicroCHIPS, Inc.

Srikar Vengallatore McGill University

Kevin Turner University of Wisconsin-Madison

DD6/FF6: Joint Session: Nanomechanics for MEMS/NEMS Applications

Session Chairs

Brad Boyce

Henry Padilla

Kevin Turner

Wednesday AM, December 02, 2009

Room 304 (Hynes)

9:30 AM - **DD6.1/FF6.1

Measuring and Mapping Mechanical Properties with Nano-Scale Resolution.

Robert Cook ¹
¹Ceramics Division, National Institute of Standards and Technology, Gaithersburg, Maryland, United States

Show Abstract

10:00 AM - DD6.2/FF6.2

Imperfection and Size Dependent Ductility of Thin Freestanding Metallic Films.

Thomas Pardoen ¹, Charles Brugger ¹, Michael Coulombier ¹, Alexandre Boe ¹, Marie-Stephane Colla ¹, Joris Proost ¹, Thierry Massart ², Jean-Pierre Raskin ¹
¹, Université catholique de Louvain, Louvain-la-Neuve Belgium, ², Université libre de Bruxelles, Brussels Belgium

Show Abstract

10:15 AM - DD6.3/FF6.3

Improved In-situ On-chip Microelectromechanical Uniaxial Tensile Tester.

Siddharth Hazra ¹, Jack Beuth ¹, Maarten de Boer ²
¹Mechanical engineering, Carnegie Mellon University, Pittsburgh, Pennsylvania, United States, ², Sandia National Laboratories, Albuquerque, New Mexico, United States

Show Abstract

10:45 AM - DD6.5/FF6.5

Size-dependent Mechanical Properties of Polymer-nanowires Fabricated by Two-photon Lithography.

Satoru Shoji ¹, Sana Nakanishi ¹, Tomoki Hamano ¹, Satoshi Kawata ^{1 2}
¹Department of Applied Physics, Osaka University, Osaka Japan, ²Nanophotonics Laboratory, RIKEN, Saitama Japan

Show Abstract

Two-photon lithography allows us to fabricate arbitrary three-dimensional structures with micro/nano-spatial resolution. The intrinsic three-dimensional spatial resolution of two-photon lithography is a promising tool for developing a variety of novel photonic and mechanical nano-devices. In addition, two-photon lithography makes it possible to study the properties of polymers of micro/nano-meter dimension [1-4]. In this presentation, we show the evidence that the elasticity and the transition temperature of polymers start to show size-dependent characteristics when the size of the polymer decreases down to a few hundreds of nanometers. We fabricated free-standing polymer nano-wires in the shape of coil spring by two-photon lithography, and measured the elasticity of the nano-wires by applying a mechanical tension onto the springs through laser trapping technique under temperature control. From the stretching length of the spring generated by a certain optical force, the spring constant of the spring and the shear modulus of polymer were calculated from simple Hook’s law. The material we used is a compound of methyl-methacrylate, dipentaerythritol hexaacrylate, benzil, and 2-benzyl-2-(dimethylamino)-4’-morpholino-butyrophenone. We scanned a light spot of a Ti:Sapphire laser at 780 nm with a pulse width of 80 fs focused by a 1.4 NA microscope objective to shape the polymer nano-wires suspended by a thick pillar. We observed that the elasticity of the polymer, which is usually an invariable coefficient, changes according to the thickness of the polymer wire. Furthermore, we changed the temperature of the entire polymer structures from -20 to 40 °C with measuring the elasticity of the springs. We observed phase transition of polymer wires with a rapid change of the share modulus, which also shows a size-dependent behavior. Recently, the similar experimental results, i.e. the drop of the glass transition temperature, were also reported in thin polymer nano-films. Our result is a clear evidence of such a nano size-effect of mechanical properties in polymers confirmed from free-standing polymer nanostructures. [1]S. Nakanishi, H. Yoshikawa, S. Shoji, Z. Sekkat, and S. Kawata, J. Phys. Chem. B 112, 3586-3589 (2008). [2] S. Nakanishi, S. Shoji, S. Kawata, and H.-B. Sun, Appl. Phys. Lett. 91, 063112 (2007). [3]H. Ishitobi, S. Shoji, T. Hiramatsu, H.-B. Sun, Z. Sekkat, and S. Kawata, Opt. Express 16, 14106-14114 (2008). [4]S. Kawata, H.-B. Sun, T. Tanaka, and K. Takada, Nature, 412, 697 (2001).

11:00 AM - DD6/FF6

BREAK

11:30 AM - **DD6.6/FF6.6

Nano- and Micron-Scale Fracture and Fatigue: From MEMS to OLED Structures.

Winston Soboyejo ^{1 2}
¹Mechanical and Aerospace Engineering, Princeton University, Princeton, New Jersey, United States, ²Princeton Institute of Science and Technology of Materials, Princeton University, Princeton, New Jersey, United States

Show Abstract

12:00 PM - DD6.7/FF6.7

Performance and Reliability of Ru-coated Microrelays.

Maarten de Boer ¹, David Czaplewski ^{1 3}, Michael Baker ¹, Steven Wolfley ¹, Gary Patrizi ¹, David Tallant ², James Ohlhausen ²
¹Microsystems Science, Technology and Components, Sandia National Labs, Albuquerque, New Mexico, United States, ³Center for Nanoscale Materials, Argonne National Laboratory, Argonne, Illinois, United States, ²Materials Science and Engineering, Sandia National Labs, Albuquerque, New Mexico, United States

Show Abstract

12:15 PM - DD6.8/FF6.8

Investigating the Dynamics of Adhesion and Separation of Microscale Silicon Contacts.

David Grierson ¹, John Nguyen ², Kevin Turner ¹
¹, UW-Madison, Madison, Wisconsin, United States, ², Sandia National Laboratories, Albuquerque, New Mexico, United States

Show Abstract

Controlling adhesion is a primary challenge for improving the performance and reliability of micro-electromechanical systems (MEMS). Microscale components are highly susceptible to the influence of surface forces, and systematic micromechanical investigations into the nature of adhesion between MEMS materials need to be undertaken in order to fully characterize and ultimately surmount the constraints that adhesive forces impose upon successful device creation. Interfacial behavior is typically described by a single quantity such as the work of adhesion, but the adhesion and separation behavior of an interface is more complex, and a more complete description of the mechanical interaction between contacting surfaces is needed. To investigate this and other adhesion-related issues, we have constructed a novel micromachined test structure to characterize room-temperature adhesion of MEMS-material surfaces with nanoscale roughness. The measurement approach shares some similarities with on-chip micromachined beam arrays that have been employed previously, but is unique in that the microscale beams are not fabricated directly on the test substrate. Instead, a chip with micromachined silicon cantilever beams is controllably contacted with a separate test surface to form the interface to be studied. The separation gap between the beams and the substrate is controlled with a piezoelectric nanopositioner, and the deformed profiles of the adhered beams are measured with an interferometer. The use of a separate beam and test surface allows for the investigation of a broader range of materials and facilitates the inspection and analysis of the contacting surfaces before and after testing. The test method allows for dynamic, time-dependent characterization of both the adhesion and separation of microscale contacts. Here, we describe the technique, report theoretical and finite element analysis of the test structure, and present measurements of adhesion between hydrophilic, single-crystal silicon surfaces with nanoscale roughness (r.m.s. ~0.3 nm) as a function of relative humidity (RH). We find that the work of separation is three to seven times higher than the work of adhesion, which is considerably different than the conventional assumption that the work of adhesion and fracture are the same for room temperature contacts, and we find that the adhesion energy and hysteretic behavior are influenced by the RH. The underlying mechanisms that lead to adhesion and hysteresis in these contacts will be discussed. This work was partially supported by AFOSR MURI Contract #FA9550-08-1-0337.

12:30 PM - DD6.9/FF6.9

Mapping Nanoscale Wear Field by Combined Atomic Force Microscopy and Digital Image Correlation Techniques.

Zhi-Hui Xu ¹, Michael Sutton ¹, Xiaodong Li ¹
¹Department of Mechanical Engineering, University of South Carolina, Columbia, South Carolina, United States

Show Abstract

Meetings & Events

Publishing Alliance

2009 MRS Fall Meeting & Exhibit

Symposium DD : Microelectromechanical Systems--Materials and Devices III

Symposium Organizers

Symposium Organizers

Symposium Organizers

Materials Research Society

Meetings & Events

Careers & Advancement

Publications & News

Advocacy & Policy

Programs & Outreach

About MRS