Christoph Eberl Karlsruhe Institute of Technology
Frank W. DelRio National Institute of Standards and Technology
Maarten P. de Boer Carnegie Mellon University
Chris Keimel GE Global Research
TT1: Devices and Processing: Si and Non-Si
Monday PM, November 28, 2011
9:30 AM - **TT1.1
Texas Instruments Digital Micro-Mirror Device: A Massively Paralleled Micro-Opto-Electro-Mechanical System (MOEMS) - Materials, Metrology and Reliability Considerations.
Patrick Oden 1 Show Abstract
1 DLP(R) Products, Texas Instruments, Inc., Plano, Texas, United States
Through the 15+ years DLP® Products has been producing salable digital micro-mirror devices (DMD) chipsets in the marketplace, over 30 million systems have been offered with between 0.5 – 8 million mirrors on each part working uniformly at the level of human vision perception. Products offerings range from business class projectors to HDTV systems to 3-D Digital Cinema Systems utilize the same basic technology – albeit with a form-factor ranging from a ~cm3 size for pico-projection needs to ~ft3 for high light output cinema requirements.. With this, a significant breadth and depth of competence has been amassed in the specific surface-MOEMS spatial light modulator system we produce with metals and sacrificial resists over a CMOS drive circuitry. In this talk, I will share some of the basics on what we produce (MEMS structure), how we interrogate the device (metrology) and how long it will last (reliability) with some details on the materials aspect to the system.
10:00 AM - TT1.2
Ultrathin Large Circular Membranes with High Quality Factors.
Vivekananda Adiga 1 , Rob Ilic 1 , Rob Barton 1 , Isaac Storch 2 , Paul McEuen 2 , Jeevak Parpia 2 , Harold Craighead 1 Show Abstract
1 applied and engineering physics, Cornell university, Ithaca, New York, United States, 2 Department of Physics, Cornell university, Ithaca, New York, United States
We have fabricated up to 1 mm diameter high tensile stress (1.2 GPa, thickness ∼15 nm) circular SiN membranes and upto 100 μm diameter self tensioned single layer graphene drum resonators. We used electron beam lithography to define uniform ultrathin stoichiometric silicon nitride circular membranes and optical lithography to define through holes in the silicon which marked the circular boundaries for suspended graphene. Resonant frequency and quality factors are measured using optical interferometric detection technique. We have measured extremely high quality factors (upto 4,000,000 for ultrathin silicon nitride membranes and upto 5,000 for graphene) at room temperature. High quality factors observed in these resonators indicate that the active dissipation mechanisms are different from conventional high surface to volume ratio resonators which show very low quality factors. The measured mechanical Q shows a systematic modal and size dependence, possibly indicating the influence of clamping losses. These findings pave the way for identifying optimum device geometries and modes for achieving high Q resonators for applications in mass sensing and fundamental research in optomechanics.
10:15 AM - TT1.3
Low Profile Packaging for MEMS Aeroacoustic Sensors.
John Burns 1 , Robert White 2 , Joshua Krause 3 Show Abstract
1 Mechanical Engineering, Tufts University, Boston, Massachusetts, United States, 2 Mechanical Engineering, Tufts University, Medford, Massachusetts, United States, 3 Mechanical Engineering, Tufts University, Medford, Massachusetts, United States
This paper describes a semi-automated conductive ink process used for packaging MEMS devices. The method is applied to packaging of MEMS acoustic sensors for wind tunnel testing. The primary advantage of the method is a reduction in surface topology between the package and the integrated MEMS sensors. In this particular application, the sensor records surface pressure and shear stress under a turbulent boundary layer. In order to avoid self-noise effects or other modifications to the boundary layer structure associated with surface roughness, the interface between the MEMS sensor and its package must be as close to planar as possible.The thickness of the viscous sub layer below the turbulent boundary layer is the upper bound on allowable surface topology. For wind tunnel flows at free stream velocities between 20 and 200 m/s and plate lengths on the order of half a meter, the Reynolds number is between 10^5 and 10^7. This suggests that the viscous wall unit will be on the order of 1 to 35 micrometers. The viscous sub layer is approximately 5 wall units thick, so surface topologies of between 5 and 175 micrometers are desired. A previous packaging approach using gold wire bonds with CNC machined epoxy fill resulted in a minimum surface topology of greater than 100 micrometers. In addition, for large arrays of MEMS microphones, yield issues were dominated by wire bond integrity problems. These two issues were the primary motivation for developing the low profile conductive ink process. However, the process is generally useful and can be applied to the packaging of various types of sensor systems that require low profile interconnects.The conductive ink process consists of a low profile, low volumetric resistivity silver conductive ink which is pneumatically dispensed from a syringe between the pads of the printed circuit board package and its integrated sensor. The package and sensor are mounted to an aluminum fixture on top of computer controlled micro-positioning stages for accurate and precise placement of the silver traces. Ongoing tests are being conducted to map the resistivity, contact resistance, and trace geometry to the process parameters of feed rate, syringe diameter, dispense pressure, and dispense duty cycle. The impedance and surface topology of the conductive ink packaging scheme will be compared to previous wire bonded hybrid packaging schemes.The ability to produce low-profile connections easily and effectively allows for a simple integration of MEMS devices into low topology packages. By optimizing the flow characteristics of the conductive ink through the syringe, the velocity of the micro-positioning stages, and the graphic user interface, the height of the silver traces can be reduced to fewer than 50 micrometers in order to achieve low profile, low resistance, and versatile connections for a wide variety of applications. In addition, process compatibility with a variety of release etchants is under investigation.
10:30 AM - TT1.4
Piezotronic Nanowire Based Resistive Switches as Programmable Electromechanical Memories.
Wenzhuo Wu 1 , Zhong Lin Wang 1 Show Abstract
1 School of Material Science and Engineering, Georgia Institute of Technology, Atlanta, Georgia, United States
The concept of complementing field effect transistors (FETs) with two-terminal hysteretic resistive switches has recently attracted a great interest in implementing and scaling novel nonvolatile resistive memories for ultrahigh density memory storage and logic applications. Notably, previous existing non-volatile resistive memories are all based on electrically switchable resistance change in various oxides and ionic conductors. These devices are electrically programmed and they are not suitable for direct interfacing with actuation/triggering other than electrical inputs. For applications such as human-computer interfacing, sensing/actuating in nanorobotics, and smart MEMS/NEMS, a direct interfacing of electronics with mechanical actions is required. Here we present the first piezoelectrically-modulated resistive switching device based on piezotronic ZnO nanowire (NW), through which the write/read access of the memory cell is programmed via electromechanical modulation. Adjusted by the strain-induced polarization charges created at the semiconductor/metal interface under externally applied deformation by the piezoelectric effect, the resistive switching characteristics of the cell can be modulated in a controlled manner, and the logic levels of the strain stored in the cell can be recorded and read out, which has the potential for integrating with NEMS technology to achieve micro/nano-systems capable for intelligent and self-sufficient multi-dimensional operations.
10:45 AM - TT1.5
A Detailed Study of a Novel Wafer Separation Method for Surface Sensitive MEMS Wafers.
Karl Malachowski 1 , Simone Severi 1 , Rita van Hoof 1 , Sandeep Sangameswaran 1 , Satoshi Genda 2 , Tomotaka Tabuchi 2 , Naoki Uchiyama 3 , Ann Witvrouw 1 Show Abstract
1 , IMEC Belgium, Leuven Belgium, 2 , DISCO Hi-Tec Europe GmbH, Munich Germany, 3 , Hamamatsu Photonics K.K., Shimokanzo Iwata-City Japan
Wafers containing MEMS (Micro Electro Mechanical Systems) devices cannot be diced and handled like wafers containing DRAM or Logic devices. Indeed, special die and wafer handling procedures are required due to the sensitive and fragile freestanding structures on the MEMS wafers. Ideally, MEMS structures are protected by a wafer-level permanent cap before reaching the die separation step, which is usually abrasive blade dicing. However, as most capping processes lead to non-transparent caps, this approach is not always possible for Micro Opto-Electro Mechanical Systems (MOEMS). This paper focuses on an alternative dry and non-abrasive die separation method, which does not harm the fragile MEMS devices and also does not introduce extra processing steps in the MEMS process flow. The dicing method presented here is known as "Stealth Dicing", developed by Hamamatsu Photonics. The dicing performance and capability of the system is stepwise investigated on 200mm full thickness wafers, starting with stealth dicing of blanket silicon wafers, adding further process steps and material layers until finally reaching wafers with released and fully functional structures (like mirrors, clamped-clamped beams and cantilevers). The diced wafers are analyzed with respect to the silicon cutting quality, possible particle contamination, the intactness of functional structures, as seen in the optical microscope, and their mechanical and electrical functionality. Also the performance and limitations of two different Stealth Dicing Engine (SDE) types, SDE01 and SDE03, are compared to each other. In conclusion, from our work, design rules and proper dimensions of the scribe line could be determined. Also integration solutions, describing steps before and after the stealth dicing process including the contact-less dicing tape application to the wafer back side and the final die separation method by tape stretching, are presented. It was found that the SDE03 laser with its outstanding performance in terms of process speed and separation quality could finally bring a break-through for this type of technology for MEMS wafer separation.
11:30 AM - **TT1.6
Subtractive Microfabrication Processes for Metal Alloys and Ceramic Materials.
Yogesh Gianchandani 1 Show Abstract
1 EECS Department, University of Michigan, Ann Arbor, Michigan, United States
The possible applications of MEMS are constrained by the properties of materials that are available for fabricating microstructures. A variety of insulating, semiconductor, and metallic films can be deposited on silicon, glass, or other substrates by chemical vapor deposition, physical vapor deposition, electroplating, or other techniques. Such additive methods expand the material diversity, but not to the fullest extent – for example, stainless steel is rarely used in MEMS despite its attractive properties. In addition, the material properties of the deposited films can be difficult to control – for example, residual stress, film composition, etc., can present challenges. This presentation will discuss two lithography-compatible subtractive processes that allow bulk metal alloys and ceramic composites to be used as structural materials. The first is micro electrodischarge machining (micro-EDM), which can be performed with lithographically fabricated cutting tools. This method has been used to microfabricate high power DC relays and RF switches, radiation detectors, magnetic micromotors, etc. These devices have been fabricated from metals such as stainless steel, platinum-rhodium, iron-nickel magnetic alloys, etc. The second is micro ultrasonic machining (micro-USM), which can be used for the batch mode machining of brittle materials by micro-tools. (These micro-tools, in turn, can be easily fabricated by micro-EDM.) Micro-USM has been used for machining glass, ruby, glass-mica ceramic, and piezoelectric lead zirconate titanate (PZT). Applications range from micro-heaters to 3-dimensional mechanical resonators. Prospects for further research and new applications will also be discussed.
12:00 PM - TT1.7
Acoustic Energy: A New Tool for MEMS Manufacturing.
Don Dussault 1 , Viorel Dragoi 2 Show Abstract
1 , Product Systems Inc., Campbell, California, United States, 2 , EV Group, Sankt Florian/Inn Austria
The use of non-standard materials (e.g. specific substrates or polymer materials) for MEMS applications impose a requirement for the development of new techniques for even well-established process steps. Acoustic energy in the MHz frequency range has been used in the semiconductor industry for various processes such as photoresist development, substrate cleaning and electro-plating enhancement. Our work presented in this paper is focused on cleaning and resist development. Initial work on megasonic-enhanced wet processing was done in immersion type configurations in which the substrate to be processed was submerged in a tank filled with the process fluid. Megasonic energy was introduced to the fluid either indirectly through a coupling fluid layer, or directly with a transducer in direct contact with the process fluid.These methods were limited by irregularities in field caused by acoustic wave reflections endemic to a tank type system with a wafer present. This non-uniformity effected both cleaning and resist development results.A single wafer spin processer with triangular-shape transducer, the MegPie, was used for both clean and resist process. This transducer covers a sector of the wafer and is placed in proximity to substrate surface. The small gap between transducer and substrate is filled with the required process fluid. The shape and positioning of the transducer assure optimal acoustic energy transfer to the process fluid resulting in high uniformity.Substrate cleaning during a process flow plays a crucial role for the final success of the process. The use of acoustic energy in the megasonic range in either batch or single wafer configurations is known to significantly enhance the cleaning results.Wafer bonding is one of the MEMs processes that is very sensitive to particle contamination and demands the use of high efficiency cleaning processes with good uniformity and repeatability.Acoustic energy was used in the past in photolithography for enhancing the uniformity of, and accelerate resist development process, initially in LIGA applications.The new MEMS applications, particularly the ones based on microfluidic structures built in polymer layers coated on a substrate raised the demand of new photolithography processes able to resolve high aspect ratio structures. Such an optimized process addresses new coating/soft baking procedures, new UV exposure techniques as well as new development methods.Experimental results for single wafer cleaning and enhanced photoresist development will be presented. Particle neutrality and high particle removal efficiency were demonstrated for single wafer cleaning.Fabrication of high aspect ratio structures using thick resists and a significantly reduced process time for standard resists is introduced.
12:15 PM - TT1.8
One- and Two-Photons Micro-Structuring of Polymers Containing Photoluminescent-Chromophores and Azobenzene-Moieties.
Antonio Ambrosio 1 , Pasqualino Maddalena 1 , Andrea Camposeo 2 , Marco Polo 2 , Antonio Neves 2 , Dario Pisignano 2 , Antonio Carella 3 , Fabio Borbone 3 , Antonio Roviello 3 Show Abstract
1 , CNR-SPIN Napoli and Dipartimento di Scienze Fisiche, Università degli Studi Federico II, Napoli Italy, 2 , NNL, Istituto Nanoscienze-CNR and Dipartimento di Ingegneria dell’Innovazione, Università del Salento, Lecce Italy, 3 , Dipartimento di Chimica, Università degli Studi Federico II, Napoli Italy
The possibility of fully-optical structuring of azo-polymers films surface has recently attracted interest for its potential application in optoelectronics. For instance, Ubukata et al.  have reported on the fabrication of a double-layered distributed feedback (DFB) laser based on a diffractive element constituted by a one-dimensionally structured azo-polymer, obtained by exposing to the interference pattern of two visible laser beams from an Argon ions laser.In fact, mixing active compounds with azobenzene molecules or polymers can represent a strategic way for obtaining systems that combine patterning capability (through the azobenzene photo-isomerization) with light emission. Our results show the possibility to pattern the mixture by means of a laser scanning technique, thus realizing photoluminescent features whose shapes and spatial distribution can be arranged by exploiting the light-polarization dependence of the mass migration process. The mass migration phenomenon occurring on the free surface of azobenzene-containing polymers illuminated by light of appropriate wavelength is employed to pattern polymeric films constituted by an azo-polymer containing a photoluminescent chromophore. Different topographical features are obtained by adjusting the laser scanning parameters, in particular the laser polarization direction .Furthermore, we report about the possibility to drive mass migration into an azo-based polymer by means of two-photon absorption. For this experiment, the azo-polymer we used (PU-AS-Y-DCV) is a polyurethane obtained containing, in the main chain, a symmetric azo-derivative of 4-(dicyanomethylene)-2-methyl-6-[p-(dimethylamino)-styryl]-4H-pyran (DCM. Our azo-chromophore, characterized by a symmetric donor-acceptor-donor π-conjugated structure, is expected to show good TPA features. It is, in fact, well known that a symmetric groups distribution into a molecule may favor TPA properties due to symmetric charge transfer from the end of the molecule to the middle. The samples we used for TPL are high optical quality films, with thickness of 550 nm, prepared by spin coating a pyridine solution (polymer concentration 5% wt) on glass coverslips. By this way we have realized structures down to 250 nm wide employing the diffraction limited spot of the 800 nm wavelength pulsed laser, far below the half-wavelength diffraction limit of the focused laser beam. T. Ubukata, T. Isoshime, and M. Hara, Adv. Mater. 17, 1630 (2005);. Ambrosio, A. Camposeo, A. Carella, F. Borbone, D. Pisignano, A. Roviello, P. Maddalena, Journal of Applied Physics 107, 083110 (2010) A. Ambrosio, E. Orabona, P. Maddalena, A. Camposeo, M. Polo, A.A.R. Neves, D. Pisignano, A. Carella, F. Borbone, A. Roviello, Applied Physics Letters 94, 011115 (2009).
12:30 PM - TT1.9
Patterned Piezoelectric PZT Thin Films Using Non-Lithographic Ink-Jet Printing Method for MEMS Applications.
Seung-Hyun Kim 1 , Jinkee Lee 1 , Chang Young Koo 2 , Jooho Moon 3 , Angus Kingon 1 Show Abstract
1 School of Engineering, Brown University, Providence, Rhode Island, United States, 2 R&D Center, INOSTEK Inc., Ansan, Gyeonggi, Korea (the Republic of), 3 Materials Science and Engineering, Yonsei University, Seoul Korea (the Republic of)
Recently, MEMS (Micro-Electro-Mechanical Systems) with integrated piezoelectric films have been applied to electrical and mechanical devices such as RF switches, ink jet printer heads, and various MEMS sensors. MEMS with integrated piezoelectric lead zirconate titanate (PZT) 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. Integration of PZT-based piezoelectric films into MEMS devices is not straightforward, due to materials and process compatibility issues. The common approach is to utilize thin film process methods such as chemical solution deposition processing, but finding patterning and etching methods that are effective for, and compatible with, both the 3D MEMS structuring and the oxide films remains a challenge, particularly at the manufacturing level. Screen printing of the PZT after MEMS processing has been widely adopted for the rapid manufacture of simple MEMS structures. However, annealing temperatures are constrained to ~ 900°C and in general PZT films by the screen printing method have displayed porous and non-uniform microstructures and poor electrical properties. Furthermore, the method is limited to thick films of > ~ 5μm, and the necessary high process temperatures lead to reactions with the MEMS structure. In this work, we systematically investigated ink-jet printed PZT thin films in the sub-μm thickness range and low process temperatures in the range of 650°C. The method direct prints precursor chemical solution in predetermined locations on the preprocessed MEMS device or wafer. We have successfully demonstrated adequate patterning resolution for piezoelectric applications and excellent electrical properties of PZT thin films using this cost effective and simple non-lithographic process. This method can be applied for both MEMS sensor and actuator applications.
12:45 PM - TT1.10
Development and Characterization of Ni-Al Superalloys for High Temperature LIGA MEMS Materials.
Devin Burns 1 , Michael Teutsch 2 , Sara Perez-Bergquist 4 , Klaus Bade 3 , Tresa Pollock 5 , Jarir Aktaa 2 , Kevin Hemker 1 Show Abstract
1 Mechanical Engineering, Johns Hopkins University, Baltimore, Maryland, United States, 2 Institute for Material Research II (IMF II), Karlsruhe Institute of Technology (KIT), Karlsruhe Germany, 4 , Los Alamos National Laboratory, Los Alamos, New Mexico, United States, 3 Institute for Microstructure Technology (IMT), Karlsruhe Institute of Technology (KIT), Karlsruhe Germany, 5 Material Science, University of California, Santa Barbara, Santa Barbara, California, United States
Nickel is one of the most electrodeposited structural LIGA MEMS materials (LIGA is a German acronym for lithography, electroplating and molding). The LIGA technique produces metallic structures with submicron resolution and high aspect ratios. However, LIGA nickel’s mechanical properties degrade when subjected to elevated temperatures. Bulk Ni-based superalloys have shown great success in land and aero based turbine applications. The manufacturing of Ni superalloy microcomponents could therefore increase their application range towards higher temperatures.In this work, existing LIGA Ni microtensile specimens are modified by a vapor phase aluminization process. Aluminum additions in excess of 20 at.% are possible. To transform these composite Ni-Al materials into superalloys, different heat treatment cycles are investigated. The effect of heat treatment on the structural and mechanical properties of these LIGA superalloys is then studied through microtensile testing and microstructural investigations.
TT2: Experimental Techniques and Mechanical Properties
Monday PM, November 28, 2011
2:30 PM - **TT2.1
In Situ Observation and Mechanical Criterion on Interface Cracking in Nano-Components.
Takayuki Kitamura 1 , Takashi Sumigawa 1 Show Abstract
1 Department of Mechanical Engineering & Science, Kyoto University, Kyoto, Kyoto, Japan
We have investigated the criterion of interfacial crack initiation in nanometer-scale components (nano-components) by means of a loading facility built in a transmission electron microscope (TEM). Three types of experiments are conducted in this project.(1) In order to clarify the applicability of conventional continuum mechanics to the nano-components, we prepare cantilever specimens with different shape, which introduce different stress fields, containing an interface between a 20-nm-thick copper (Cu) thin film and a silicon (Si) substrate. One specimen with the straight arm possesses the stress concentration in the scale of 20nm-30nm near the interface edge (junction between the interface and the surface), while the other with a step on the arm does it in the interior. In the both, a crack is initiated at the stress-concentrated location of Cu/Si interface. These demonstrate the validity of the “stress” criterion even for the nano-scale fracture.(2) In order to examine the effect of microscopic structure on the mechanical property, we fabricate a bending specimen in the nano-scale with thin Cu bi-crystal (the thickness of about 100nm) formed on Si substrate, of which understructure can be observed in situ by means of a TEM during the mechanical experiment. The initial plastic deformation takes place near the interface edge in a grain with a high critical resolved shear stress and expands preferentially in the grain. Then, the plasticity appears near the between Cu grain boundary and Cu/Si interface, and this development brings about the interfacial cracking from the junction. These indicate the governing influence of understructure on the mechanical property in the nano-components.(3) In order to investigate the fatigue behavior of metal in a nano-component, a cyclic bending experiment is carried out using nano-cantilever specimens with a 20 nm- or 200nm thick Cu constrained by highly rigid materials (Si and SiN). The high strain region is in the size of 20-40nm near the interface edge. The specimen breaks along the Cu/Si interface before the maximum load under the fatigue loading. The load-displacement curve shows nonlinear behavior and a distinct hysteresis loop, indicating plasticity in the Cu film. Reverse yielding appearing after the 2nd cycle suggests the development of a cyclic substructure in the Cu film. The cumulative plastic strain in the Cu film at fracture is more than three times larger than that under monotonic loading. These indicate that the crack is caused by characteristic understructure owing to fatigue cycles. The S–N curve shows clear dependence of fatigue life on the applied stress in the high-stress and the fatigue limit in the low-stress.
3:00 PM - TT2.2
Deformation and Fracture of Single-Crystal Silicon Theta-like Specimens.
Frank DelRio 1 , Michael Gaither 1 , Richard Gates 1 , Robert Cook 1 Show Abstract
1 Ceramics Division, National Institute of Standards and Technology, Gaithersburg, Maryland, United States
Many advanced materials are intended for use in small scale applications, for example, microelectronics, microelectromechanical systems (MEMS), photonics, and magnetic storage, or may be available only in small volumes, for example during materials development. Developing or optimizing such materials and their processing thus requires measurements of structure and properties at these scales. A pervasive measurement requirement is that of measuring mechanical properties, and relating them to processing and structure, in order to optimize manufacturing yield and operational performance, especially reliability. However, establishing processing-structure-mechanical properties linkages at small scales is difficult due to problems associated with specimen gripping, alignment, and post test sample collection. The micro-scale NIST theta specimen, shaped like the Greek-letter theta, allows tensile strength of brittle materials to be measured at small scales while minimizing these problems. In this talk, we summarize recent work on the deformation and fracture of single-crystal silicon theta specimens, with a particular emphasis on using the specimens to elucidate the effects of sample geometry and fabrication technique on the tensile strength of single-crystal silicon. Three sets of MEMS-scale single-crystal silicon theta specimens are fabricated using two deep reactive ion etching (DRIE) recipes and a temperature-controlled cryogenic plasma etching recipe, each set resulting in a different specimen surface quality. DRIE is a silicon etch technique that results in high-aspect-ratio sidewalls and nearly uniform etch steps, or scallops, along these sidewalls. The cryogenic plasma etching technique results in smooth, high-aspect-ratio sidewalls. Each sample is tested by instrumented indentation and finite element analysis (FEA) is used to determine sample strength. Equations developed via FEA translate load-displacement response at the load-point into stress-strain behavior across the theta web region. Strength values for each set of specimens are examined via Weibull statistics. The resulting surface roughness for each etching recipe is determined by atomic force microscopy and sample fragments are examined via field-emission scanning electron microscopy. Surface roughness topography and fracture origins located during fractographic analysis of tested samples are compared with strength-limiting flaw size calculations.
3:15 PM - TT2.3
Fatigue Testing of Al and Cu Thin Films Using a Resonance Cantilever Bending Setup.
Sofie Burger 1 , Christoph Eberl 1 , Alexander Siegel 2 , Alfred Ludwig 2 , Oliver Kraft 1 Show Abstract
1 Institute for Applied Materials, Karlsruhe Institute of Technology, Eggenstein-Leopoldshafen, Baden-Wuerttemberg, Germany, 2 Institut für Werkstoffe/Werkstoffe in der Mikrotechnik, Ruhr University of Bochum, Bochum, Nordrhein-Westfalen, Germany
Mechanical properties and fatigue lifetime of thin films differ significantly from those of bulk materials. With decreasing film thickness the fatigue failure changes from a dislocation based mechanism to pore formation and grain boundary cracking.To obtain a better understanding of the governing mechanisms, a new fatigue setup for micro cantilever bending was developed. A micro-machined Si cantilever coated with Al or Cu thin films is excited at its resonance frequency by a piezo actuator. The thin film undergoes cycles of different strain amplitudes along the cantilever from a maximum near the shoulder to zero at the free end. With a reflected laser beam from the surface the bending amplitude as well as the reflectivity is detected. Thus, by observing the damage front on the thin film after a certain number of cycles, it is possible to compile a fatigue lifetime diagram from one cantilever.Experimental results of cycled thin films of sputter deposited Al, Cu, and Cu with a Ti seed layer will be presented. The micro- and damage structure of these samples was qualitatively analyzed by a SEM/FIB Dual Beam. Additionally, mechanical characterization of these thin films was performed by nanoindentation. The results will be discussed in the light of current models for fatigue in thin films.
3:30 PM - TT2.4
Indentation and Scratch Device with In Situ Raman and Optical Capabilities.
Yvonne Gerbig 1 , Chris Michaels 2 , Aaron Forster 3 , Robert Cook 1 Show Abstract
1 Ceramics Division, NIST, Gaithersburg, Maryland, United States, 2 Surface and Microanalysis Science, NIST, Gaithersburg, Maryland, United States, 3 Materials and Construction Research Division, NIST, Gaithersburg, Maryland, United States
Instrumented indentation and scratch experiments are widely used techniques to study the mechanical behavior of materials at small scales and thus have been exploited in particular as models for processes such as deformation and fracture. In combination with mechanical tests, microscopic and spectroscopic studies enable determination at the crystallographic and molecular level of the kinetics and processes involved in the mechanical deformation of materials, e.g. strain build up in crystal lattice, phase transformations, and changes in crystallinity. However, many of these phenomena occurring during indentation and scratching can only be observed in their entirety and analyzed in depth under in situ conditions. This talk describes the design, calibration, and operation of an indentation and scratch device that is coupled with a Raman microscope to conduct in situ spectroscopic and optical analysis of mechanically deformed regions under contact loading. The presently open-loop controlled force transducer of the device allows adjustment of crucial experimental parameters, such as indentation strains and strain rates. An incorporated displacement sensor is employed for simultaneous displacement measurements, and thus collection of force-displacement curves comparable to conventional instrumented indentation testers. The device is designed to mount on the sample stage of an inverted optical microscope that is configured for Raman microscopy, allowing optical access to the mechanically deformed regions of transparent samples. The modular design allows conversion of the basic indentation device into a scratch tester through the addition of a fiber optical sensor for measurements of lateral forces. The lateral sample movement is generated by a piezoelectric stage containing an optical encoder. The capabilities of the presented device are demonstrated by in situ studies of the indentation-induced phase transformations of semiconductors and shear-induced modifications of molecular alignments in polymer films. Such measurements are critical in the design and optimization of microelectromechanical systems that contain contacting, moving parts.
3:45 PM - TT2.5
A High Strength Creep Resistant Alloy for MEMS Applications.
Andrew Detor 1 , Leon Beha 1 , Christopher Keimel 1 Show Abstract
1 , General Electric Global Research, Niskayuna, New York, United States
MEMS available on the market today have been built from a rather narrow set of materials; the structural components typically fabricated from silicon or pure metals such as gold, nickel, or aluminum. Looking forward to more aggressive applications, new materials are required with properties beyond those available from the traditional materials set. For applications requiring a combination of high strength and electrical conductivity we have investigated the use of an electrodeposited nickel alloy. This alloy enables long-term operation of a new high power (>1kW) switching device that would otherwise be impossible with existing MEMS materials. The present study focuses on detailed understanding of the time-dependent mechanical behavior of this alloy in comparison to other materials. Key properties are extracted that are directly relevant to the successful operation of the high power switch, and potentially other MEMS devices.We use standard plating solutions to fabricate miniature dogbone specimens of the nickel alloy, along with gold and hard gold, on patterned wafers. A custom built mechanical test rig is then used to measure the stress relaxation performance of the material over a range of temperatures. Our analysis procedure extracts key material characteristics, such as the activation energy and activation volume associated with stress relaxation, revealing fundamental differences between these materials. A lifing approach is then outlined to predict performance of a simple cantilever beam. It is shown that the nickel alloy outperforms gold and hard gold by orders of magnitude, extending the operating window and life of MEMS actuators. The ability to deposit a mechanically robust and electrically conductive material at low temperature has the potential to replace polysilicon, enabling a new class of MEMS devices.
4:30 PM - **TT2.6
A MEMS Stage for Testing Nano to Micro Scale Samples.
Taher Saif 1 Show Abstract
1 Mechanical Science and Engineering Department, University of Illinois at Urbana-Champaign, Urbana, Illinois, United States
We present a MEMS based versatile stage that allows uniaxial testing of micro to nano scale samples in situ in scanning and transmission electron microscopes. The stage has built-in self-alignment mechanism to correct for any misalignment error, and a temperature sensor. The stage is used to test nanograined aluminium and gold films, and micro scale single crystal silicon samples at various temperatures. The thin film samples are cofabricated with the stage, silicon samples are fabricated separately. The experiments reveal that ultra small grained (50-100nm) metals can recover a significant part of their plastic strain under macroscopic stress free condition. Slightly larger grained (100-200nm) metals show yielding during unloading. Both of these behaviors originate from the heterogeneity of microstructure in the samples due to variation in grain size. During deformation, smaller grains deform elastically while relatively larger grains deform plastically. Upon unloading, the smaller grains induce reverse stress in the larger grains resulting in strain recovery and yielding. This mechanism due to microstructural heterogeneity is confirmed by in situ tests in TEM. For the micro scale silicon samples, we recover the known elastic modulus of silicon at temperatures up to 400C.
5:00 PM - TT2.7
Analysis of Thin-Film Ni-Ti-Hf Shape Memory Alloys by Combinatorial nanoCalorimetry.
Patrick McCluskey 1 , Yahya Motemani 1 2 , Chunwang Zhao 1 3 , Ming Tan 2 , Joost Vlassak 1 Show Abstract
1 School of Engineering and Applied Sciences, Harvard University, Cambridge, MA, Massachusetts, United States, 2 School of Mechanical and Aerospace Engineering, Nanyang Technological University, Singapore Singapore, 3 College of Science, Inner Mongolia University of Technology, Hohhot China
The recently developed parallel nano-scanning calorimeter (PnSC) has been used to analyze a combinatorial library of thin-film Ni-Ti-Hf samples. The PnSC consists of an array of nanocalorimeters, each with sensitivity in the nJ/K range. The array design allows for 25 samples to be created simultaneously, while the fine sensitivity allows for calorimetry measurements on nanoscale samples. The sensitivity is enabled by the minute thermal mass of the membrane sensors, which also facilitates fast heating and cooling rates. The fast heating/cooling rates allow high-throughput measurements of combinatorial libraries and novel thermal treatments. Here the PnSC is used to crystallize the as-deposited amorphous Ni-Ti-Hf samples resulting in samples with an ultra-fine grain structure (18 ± 5 nm). The martensite-austenite transformation temperature of the samples increases linearly with Hf content at a rate comparable to bulk Ni-Ti-Hf, although the transformation temperature is significantly lower than for bulk Ni-Ti-Hf because of the small grain size of the samples. Thermal cycling is used to investigate the thermal stability of the martensitic transformation. The response to high-temperature cycling (22°C < T < 850°C) changes with Ni concentration. For Ni ≤ 47 at%, the transformation temperature increases during high-temperature cycling because precipitation of (Ti1-x,Hfx)2Ni enriches the surrounding matrix in Hf; for Ni ≥ 47.7 at%, precipitation of the same phase gradually suppresses the transformation. Low-temperature cycling (22°C < T < 450°C) causes the transformation temperature to initially decrease and then stabilize. Relaxation of internal stresses by dislocations generated during thermal cycling is suggested as the active mechanism. Low-temperature thermal cycling stability of the films is improved compared to studies on bulk Ti-Ni-Hf. This result agrees with previous PnSC analysis on the Ni-Ti-Zr system, in both cases the nanoscale grain structure reduces dislocation activity.
5:15 PM - TT2.8
Scalable Fabrication of Folded Carbon Nanotube Microstructures by Iteration of CVD Growth and Capillary Densification.
Michael De Volder 1 2 , Sameh Tawfick 1 , Sei Jin Park 1 , A. John Hart 1 Show Abstract
1 , University of Michigan, Ann Arbor, Michigan, United States, 2 , IMEC, Leuven Belgium
Scalable fabrication methods for microstructures with folded and re-entrant topologies would be highly beneficial for the development of engineered surfaces, metamaterials, and lab-on-a-chip devices. Known methods of making such complex geometries are highly limited, and typically require multi-step lithography and etching, or methods such as stereolithography, multiphoton lithography, and focused ion beam (FIB) processing. While these latter approaches can create arbitrary forms, their serial nature is only practical for small production volumes. We demonstrate batch fabrication of 3D corrugated carbon nanotube (CNT) microstructures, via an iterative sequence of vertically aligned CNT growth and capillary densification. Arrays of multi-segment vertical micro-bellows and slanted micro-cantilevers are created over centimeter-scale areas, and these structures can have thin walls with aspect ratios exceeding 100:1. We characterize the mechanical properties of CNT micro-bellows, and demonstrate that their mechanical compliance can be tuned over a wide range based on the wall thickness and the number of folds. The local specificity of capillary forces during the densification process also enables the simultaneous formation of unique multi-directional architectures. The attractive properties of CNTs suggest that these structures could offer novel multifunctionality, such as coupling between mechanical deformation and electrical conductivity, and combinations of corrugated topologies with chemical and biochemical surface functionalities.
5:30 PM - TT2.9
Fabrication of Phonon Transducers on Micromachined Silicon Structures for Investigating Nanoscale Thermal Transport.
Obafemi Otelaja 1 , Jared Hertzberg 1 , Richard Robinson 1 Show Abstract
1 Materials Science and Engineering, Cornell University, Ithaca, New York, United States
Phonons are the major heat carriers in semiconductors and insulators, and a better understanding of their nanoscale transport will inform the thermal engineering of materials. To this end, we demonstrate a scalable method of fabricating phonon transducers on silicon microstructures, and we show acoustic phonon confinement effects via silicon micromachining. Al-AlxOy-Al superconducting tunnel junctions were utilized for generation and detection of non-equilibrium phonons with frequency in the order of 100 GHz and that travel ballistically along the <110> direction of a silicon mesa . The mesas were formed by a silicon nitride-masked shallow depth anisotropic etching of silicon (100) in KOH, and were 250 micron long, 10-50 microns wide and 1.5 micron high. A bilayer of photoresist is utilized with stepper lithography to form a mask for lift-of processing of tunnel junction and wiring traces onto the mesa sidewalls. “Dolan Bridges”  allow double-angle shadow evaporation of aluminum with an oxidation step in between to form the tunnel barrier. The phonon generator and detector pair formed on the sidewalls of the mesa allows the study of phonon transport across the mesa. The ballistic transport path between the generator and detector can be blocked by etching a trench into the mesa. Using dry etching, we have also etched vertical sheets of silicon into the mesas to study phonon confinement effects. We show a simple and scalable configuration for studying phonon transport in silicon microstructures, and in this talk we will present on-going efforts to extend this technique to fabricate a phonon spectrometer that would improve understanding of thermal transport in nanostructures and nanomaterials. References  W. Eisenmenger, A. H. Dayem, Phys. Rev. Lett. 18, 125 (1967).  G. J. Dolan, Appl. Phys, Lett. 31, 337-339 (1977).
5:45 PM - TT2.10
Photo-Thermal Spectroscopic Imaging of MEMS Structures with Sub Micron Spatial Resolution.
Robert Furstenberg 1 , Christopher Kendziora 1 , Michael Papantonakis 1 , Viet Nguyen 1 , R. McGill 1 Show Abstract
1 , NAVAL RESEARCH LABORATORY, WASHINGTON, District of Columbia, United States
With the increasing materials complexity of MEMS devices, there is a growing need for new characterization techniques that provide chemical composition with improved spatial resolution. Existing established techniques are not always well suited for the length scales involved in MEMS devices. For example, FTIR spectroscopy provides the chemical composition but is principally limited to bulk samples. While FTIR micro-spectroscopy addresses this problem, the practical resolution limit is limited to about 20 micrometers. Similar resolution restrictions apply to various X-ray techniques as well. On the other hand, well-developed imaging techniques at the nanometer scale (TEM/EELS, SPM, AFM etc.) may be impractical at the micron-scale. The emerging technique of Raman micro-spectroscopy provides adequate spatial resolution (~1 micrometer), but may not always be useful due to its low throughput and in cases where strong fluorescence suppresses the weak Raman signature. We are developing a new non-contact and non-destructive technique that provides similar information as FTIR or Raman spectroscopy while being immune to fluorescence and at a potentially faster scan rate and/or higher spatial resolution. It involves photo-thermal heating of the sample with a quantum cascade laser (or other suitable infrared laser) and measuring the resulting increase in thermal emissions (at wavelengths other than the laser wavelength itself) by either an infrared detector or a laser probe consisting of a visible laser reflected from the sample. The latter case allows for further increase in the spatial resolution from ~10 micrometers to ~1 micrometer or better, with the right experimental conditions. Since the thermal emission signal from the surface is directly proportional to the absorbed laser energy which is in turn proportional to the absorption coefficient, by tuning the wavelength of the infrared laser we can directly measure the FTIR spectrum of the sample. By raster scanning over the surface of the sample and applying standard chemometric algorithms, we can obtain maps of the chemical composition of the sample surface. We demonstrate this technique by imaging the surface of a microfabricated flow-through chemical vapor preconcentrator consisting of a silicon frame and a suspended-perforated polyimide membrane with a pair of platinum meander heater traces, coated with a custom sorbent polymer for selective sorption of analyte. We derive the spatial resolution of our photo-thermal imaging system as well as discuss the conditions under which the spatial resolution can be further increased from the far-field diffraction limited resolution given by the combination of the imaging optic and infrared excitation laser wavelength. In addition, we compare the relative merits of measuring the infrared photo-thermal and the infrared back-reflectance signal.
Christoph Eberl Karlsruhe Institute of Technology
Frank W. DelRio National Institute of Standards and Technology
Maarten P. de Boer Carnegie Mellon University
Chris Keimel GE Global Research
TT5: Poster Session: MEMS
Tuesday PM, November 29, 2011
Exhibition Hall C (Hynes)
TT3: ADL, Complex Lithography and Adhesion
Tuesday PM, November 29, 2011
9:30 AM - **TT3.1
Atomic Layer Deposition for N/MEMS Devices and Systems.
Victor Bright 1 Show Abstract
1 Mechanical Engineering, University of Colorado at Boulder, Boulder, Colorado, United States
Atomic Layer Deposition (ALD) and Molecular Layer Deposition (MLD) can be effectively used to deposit custom-designed, multi-material layers with atomic resolution on any micro- or nano-scale device surface. The nano-scale ALD coating can protect the devices from electrical short, charge accumulation, moisture-induced adhesion, wear, corrosion, creep, fatigue and anodic oxidation during short-term prototyping or long-term product life. The nano- and micro-electro-mechanical systems (N/MEMS) community has been looking for effective antistiction and environmental protections coatings for many years. ALD films for MEMS achieve these goals similar to what CVD Si3N4 has been for CMOS. As devices further shrink toward nano-scale, ALD-based processes offer a new strategy for depositing conformal and precise films that may have important applications as a novel dielectric, a sacrificial layer for gap control in nanofabrication, or as a structural layer for NEMS realization. ALD relies on sequential, self-limiting surface reactions to deposit ultra thin, conformal films with the following characteristics: ALD film thicknesses can be precisely deposited from a few angstroms to hundreds of nanometers; ALD films can be deposited at low temperatures compatible with CMOS; ALD films are pinhole-free, dense, smooth and highly conformal; ALD films can be deposited on silicon, polysilicon, silicon nitride, metals, polymers, and ceramics; ALD films can be conformally deposited on any size or shape device or any substrate; ALD can coat high surface area to volume ratio structures with complex geometries; ALD can deposit dielectric or conductive layers; ALD can deposit hydrophobic or hydrophilic layers covalently bonded to the surface; ALD can deposit on lithographically defined selective areas; ALD films can be micromachined to create nano-scale gaps and free standing structures. The ALD techniques for N/MEMS, pioneered at the University of Colorado – Boulder, represent breakthrough in nano-scale processes that can be used to fabricate custom-designed, multi-material layers with atomic resolution. The ALD processes developed are proven, mature, and are available to serve the N/MEMS community.
10:00 AM - TT3.2
Micro-Assembly Using Elastomeric Surfaces with Switchable Dry Adhesion.
Seok Kim 1 , John Rogers 2 Show Abstract
1 Mechanical Science and Engineering, University of Illinois at Urbana-Champaign, Urbana, Illinois, United States, 2 Materials Science and Engineering, University of Illinois at Urbana-Champaign, Urbana, Illinois, United States
Most of three-dimensional sub-milimeter scale manufacturing relies on top-down microelectromechanical systems (MEMS) fabrication which includes photolithography, deposition, and etching as primary process steps. Although conventional MEMS fabrication is one of the most powerful micromanufacturing tools, it natively requires sophisticated equipments and wasting relatively much building materials during process, particularly for fabricating three-dimensional structures and devices as opposed to macroscale manufacturing such as assembling. For last decade, many other micro-manufacturing strategies including self-assembly and self-folding have been developed to reduce manufacturing cost and make more challenging three dimensional structures which are not easily achieved with conventional MEMS fabrication. In addition, we present our effort to develop a novel micro-manufacturing method for building three-dimensional micro-structures and devices as an alternative to MEMS fabrication. We are developing an innovative microscale deterministic assembly manufacturing process which is partly inspired by and similar to ‘bricklayering’ in macroscale. The key of this micro-assembly involves fabricating microscale unit elements (‘bricks’) on donor substrates; retrieving, transferring, assembling them at a target location using elastomeric surfaces with switchable dry adhesion; bonding stacked or assembled elements. This method enables to manufacture arbitrary shaped microscale three-dimensional structures and devices in a deterministic way. Preliminary results with silicon square plates implied that the micro-assembly made possible cost-effective and rapid microfabrication in conjunction with additional manufacturing design flexibility.
10:15 AM - TT3.3
Processing-Structure Relationships for Ultra-Violet Assisted Direct-Write Fabrication Technique and Potential Applications.
Matthew Becker 1 2 , Daniel Therriault 1 2 Show Abstract
1 Mechanical Engineering, École Polytechnique de Montréal, Montreal, Quebec, Canada, 2 Center for Applied Research on Polymers and Composites, École Polytechnique de Montréal, Montreal, Quebec, Canada
The rapid and cost effective fabrication of intricate microdevices still faces significant technological and scientific challenges. Microfabrication techniques used for the creation of planar (2D) structures such as photolithography have significantly improved their precision and cycle-times in recent years and some techniques for the creation of three-dimensional (3D) structures based on stereolithography are readily available, but the complexity of the possible geometries is limited. Direct-Write (DW) fabrication techniques can be used to create miniature 3D devices using ink filaments robotically deposited in layer-by-layer sequences, but rely on specific ink viscosities to flow through micronozzles while still maintaining the desired shape after extrusion. DW techniques have so far been mainly limited to straight-spanning filaments between support points. The UV light assisted direct-write (UV-DW) procedure overcomes the layer-by-layer building limitation by robotic extrusion and polymerisation with immediate targeted exposure to UV light, thus fabricating freeform 3D filament structures. This technique has been previously used to create 3D microcoiled geometries with extrusion diameters of 100 µm, but the size of the geometries is limited by the precision of the positioning systems used. Smaller geometries, from tens of microns, to submicron and ideally to the nanometer scale, would allow for the application of the UV-DW technique to a whole new spectrum of applications. Here, processing-structure-property relationships were determined systematic experimentation. A deposition speed and UV-ink curing time (controlled by UV light intensity) were held constant and the ink flow rate was varied during fabrication first of 2D patterns, then vertical lines and suspended vertical to horizontal angle depositions to determine the optimal range of applied pressures. Following this process map, the optimal parameters were used to create complex structures such as 3D scaffolds, microcoils and square spiral towers with overall sizes two to ten times smaller than previously obtained. Applications for the UV-DW technique at the micron and submicron scales include energy storage using carbon nanotube reinforced microsprings, 3D porous scaffolding for tissue engineering using biocompatible UV hydro-gels, and tetragonal square spiral photonic band gap crystals.
10:30 AM - TT3.4
Actuation of Liquid Metal Droplets Using Electrowetting-on-Dielectric (EWOD) for Reconfigurable Antenna Applications.
Engin Cagatay 1 , Yasin Damgaci 2 , Bedri Cetiner 2 , Necmi Biyikli 1 Show Abstract
1 UNAM - Institute of Materials Science and Nanotechnology, Bilkent University, Ankara Turkey, 2 Electrical and Computer Engineering, Utah State University, Logan, Utah, United States
Electrowetting-on-dielectric (EWOD) is the phenomenon in which the wetting properties of a liquid are modified by the influence of an applied electric field. By EWOD, a conducting or non-conducting liquid can be moved using a device consisting of a dielectric layer on top of a metal electrode to which the electrostatic actuation voltage is applied. Electrowetting has many uses in different fields and disciplines such as lab-on-a-chip systems, electrical switches and reconfigurable antennas. Our aim in this study is to characterize and examine the actuation of metallic liquid droplets, namely EGaIn and Galinstan, using EWOD. EGaIn and Galinstan are used instead of Hg because they are non-toxic and liquid at room temperature. We have investigated the effect of different actuation electrode geometries like rectangular, inter-digitated fingers and crescent-shaped electrodes on the droplet actuation. In this study we have fabricated microfluidic platforms using a two-mask microfabrication process. Quartz wafers were used as substrates on which the test devices were built. First, 20nm/200nm Cr/Au was deposited using thermal evaporation to define the actuation and ground electrodes. Then, 300nm-thick SiO2 layer was deposited PECVD as the dielectric layer. Finally a 50nm-thick hydrophobic layer, Teflon AF2400, was spin-coated to achieve optimum contact angles. Characterization of the test devices involved visual inspection with optical microscope and SEM, I-V measurement with parameter analyzer and contact angle measurements of DI-water on hydrophobic surfaces. During actuation experiments, two different metallic liquids have been used: Eutectic Gallium Indium (EGaIn) and Gallium Indium Tin alloy (Galinstan). With the application of 80-100V voltage difference across the actuation electrode and the ground electrode, the metallic liquid droplets were observed to be actuated. Moreover, we observed significant differences in the actuation behavior of the metal droplets with the application of AC and DC voltages. Actuation experiments in closed-pack microfluidic channels were also successful with slightly higher actuation voltages. The effect of dielectric layer thickness on the actuation voltage was also studied. Results of this study may be potentially used in microfluidic reconfigurable antenna structures where liquid metal droplets/slugs will act as movable active antenna parts.
11:15 AM - **TT3.5
Characterizing Adhesion of Micro- and Nano-Scale Contacts.
Kevin Turner 1 Show Abstract
1 Mechanical Engineering and Applied Mechanics, University of Pennsylvania, Philadelphia, Pennsylvania, United States
Micro- and nano-scale contacts are ubiquitous in MEMS, AFM processes, and semiconductor manufacturing processes. Characterizing the adhesion mechanics of these contacts is essential to understanding the behavior of devices and many fabrication processes. This presentation will discuss two different experimental techniques to characterize the adhesion of small-scale contacts in micro- and nanosystems. First, a microbeam-based method that was used to characterize adhesion hysteresis of smooth single crystal silicon contacts will be described. The results, which show significant hysteresis between adhesion and separation, have implications for single crystal silicon MEMS and direct bonding processes. Second, the adhesion of single asperity nanoscale contacts, such as those formed by an AFM tip in contact with a surface, will be examined. Specifically, the characterization of these contacts and the role of tip geometry and wear on the adhesion will be discussed. The implications of both studies on MEMS design and operation will be reviewed.
11:45 AM - TT3.6
Contamination Thresholds of Pt- and RuO2-Coated Microrelays.
Vitali Brand 1 , Michael Baker 2 , Maarten de Boer 1 Show Abstract
1 Mechanical Engineering, Carnegie Mellon University, Pittsburgh, Pennsylvania, United States, 2 MEMS, Sandia National Labs, Albuquerque, New Mexico, United States
Microelectromechanical relays are of great interest in radio-frequency and power switching applications, and also in filtering circuits. Ohmic DC relays must maintain low electrical resistance over millions to billions of cycles. As such devices are cycled, soft metal coatings such as Au either tend to (i) cold-weld, causing failure due to adhesion, or (ii) accumulate hydrocarbons, leading to an increase in contact resistance. Pt-coated surfaces are also susceptible to hydrocarbons. We are exploring conducting metal oxides, which are hard yet have low affinity for hydrocarbons. In this work, we address quantitatively the effect of background contamination to determine the immunity of Pt-coated microrelays relative to RuO2-coated microrelays. We conjecture that due to its higher catalytic activity, the Pt will exhibit a lower contamination threshold, above which contact resistance degrades. A vacuum chamber with plasma cleaning and residual gas analysis capabilities has recently been constructed to carry out the study. The microrelay is an in-plane design employing polysilicon thermal actuators attached to contact bars. A self-shadowing design protects the thermal actuators and isolates traces, but allows coating of the contacting sidewalls. To obtain initially pristine surfaces, devices are vacuum-baked at 200 degrees Celsius. The chamber is then cooled to room temperature to attain ultra-high vacuum levels. Contaminant gases are introduced at highly controlled levels through a precision leak valve. Next, in-situ testing is conducted. In this presentation, we will detail the newly designed chamber, the switch design, its fabrication, packaging and testing. Then, electrical test results for different materials and contaminant levels will be reported. This work addresses the critical issue of the effect of environment and materials on the reliability of micromachined ohmic contact switches.
12:00 PM - TT3.7
Study of Electronic Nano-Size Contact Mechanism in Au-Based MEMS Contact Switches.
Yinxuan Yang 1 , Scott Hoffmann 1 , Benjamin Gaddy 2 , Xiaoyin Ji 2 , Christopher Freeze 2 , Douglas Irving 2 , Angus Kingon 1 Show Abstract
1 Engineering, Brown University, Providence, Rhode Island, United States, 2 Materials Science & Engineering, North Carolina State University, Raleigh, North Carolina, United States
MEMS contact switches play a very important role in mobile devices due to their low insertion losses and high on-off ratios. Gold has been used as the primary contact material for its beneficial physical and electrical properties such as negligible insulating oxide formation on the surface and very low resistance. In general, however, the performance and the reliability of the nano-size contacts in MEMS switches are very difficult to predict because the contact resistance is a strong function of the surface topography, surface chemistry (contamination), switching current density and voltage, and the number of times the surface has been previously contacted (cycling contacts). In this study, we focus on the nature of electronic transportation behaviors in single asperity nano-size contacts, both the metal-metal ohmic contacts and the metal-insulator-metal tunneling contacts. Controlled small-scale contact experiments are systematically performed to understand the behaviors of the contaminant layer on gold surfaces using the voltage, the current, and the contact force, and the mechanical changes in the gold surface resulting from tip-substrate currents. Particular attention is focused on the hydro-carbon contamination on the gold surface and its interaction with the probe under low bias and low force conditions. To address and to model this hydrocarbon layer, we applied a uniform self-assembled-monolayer on the surface of the gold-based sample. In addition, we briefly describe the characteristics of gold-nickel alloy as an attractive candidate for the contact materials with the reliable and desirable mechanical and electrical properties. This work is supported by a grant from the Office of Naval Research (N00014-10-1-0402).
12:15 PM - **TT3.8
Probing Interfacial Contact via MEMS-Based Microinstrumentation.
Roya Maboudian 1 Show Abstract
1 Chemical & Biomolecular Eng, UC Berkeley, Berkeley, California, United States
Recent developments in the micro- and nanoelectromechanical systems (M/NEMS) field have created a growing interest in evaluating the reliability of these miniaturized devices. The critical reliability issues include adhesion, friction, wear and corrosion. In this presentation, I will discuss the impact of these interactions in the M/NEMS technology. I will also present the unique opportunities provided by the MEMS processing techniques to interrogate surfaces at a length scale not easily accessible by other techniques, namely in the mesoscopic length scale. With this view, I will introduce a number of MEMS-based microinstruments that we have developed to study these interactions, and some of the insights we have gained using them about the nature of surface interactions involved in M/NEMS.
TT4: Contact Mechanics and Active MEMS Materials
Tuesday PM, November 29, 2011
2:30 PM - **TT4.1
Contacts for MEMS and NEMS Switches.
Nicol McGruer 1 2 , George Adams 2 1 Show Abstract
1 Electrical and Computer Engineering, Northeastern University, Boston, Massachusetts, United States, 2 Mechanical and Industrial Engineering, Northeastern University, Boston, Massachusetts, United States
MEMS switches are smaller and switch faster than existing mechanical relays, including reed relays. They offer improved switching performance relative to semiconductor devices, having better Ron/Coff ratios, better linearity, and lower power consumption than their solid-state counterparts. Applications in cell phones aim to reduce power consumption by switching signals more efficiently and by achieving a better match between the power amplifier and the antenna. Other applications are in communication systems, where the greatly improved linearity of the MEMS switch allows the rejection of strong unwanted signals, and in a variety of other low and high-frequency switching applications where small form factor and batch manufacturing offer economic advantages.The principal barrier to adoption of the technology is reliability. The most common failure modes are “stuck-closed,” when the switch conducts all of the time and “stuck-open,” when the switch no longer conducts well. These failures can be mitigated or exacerbated by actuator and system level design but are fundamentally limited by the performance of the materials making up the contact. The contact force for MEMS actuators of desirable size for manufacturing is generally from 100 to about 1000 micro Newtons, and this force range constrains contact choices. For example, because of the low force, gold has often been chosen as the contact material for MEMS switches. Gold does produce a low contact resistance at a given force (small Ron), and is resistant to contamination. However, as a clean gold contact is operated, the resistance generally decreases and the adhesion between the two sides of the contact increases, sometimes to the point that the actuator cannot pull the contacts apart. In our experiments we have measured the adhesion force in gold contacts after 100,000 switching cycles to be as large as 3 times the contact force (contact force 100 micro Newtons, adhesion force 300 micro Newtons), making actuator design difficult. Other contact materials, such as Ru, exhibit much lower adhesion forces, but have higher contact resistances and are more susceptible to build-up of contamination during operation. The contact forces of MEMS switches may not be large enough to break through the contaminant film, leading to the “stuck-open” type of failure. Both types of failures are strongly influenced by the contact material, the contact preparation, the switch packaging, the contact force, the current flowing in the contact, the voltage and current during the switching operation (hot switching). Many types of materials are being tested, including identical pure metals, different pure metals for the two sides, layered metals, alloys, particle-hardened metals, and various combinations of the above.Nanoswitches, with applications in memory, logic, and power management, are seen to exhibit the same kinds of limitations due to adhesion and contamination, perhaps to a greater degree.
3:00 PM - TT4.2
Simulated Nano-Indentation Testing of Au-Ni Contacts for RF-MEMS Switches.
Benjamin Gaddy 1 , Yinxuan Yang 2 , Angus Kingon 2 , Douglas Irving 1 Show Abstract
1 Materials Science & Engineering, North Carolina State University, Raleigh, North Carolina, United States, 2 Division of Engineering, Brown University, Providence, Rhode Island, United States
The next generation of high-performance wireless communication devices will be transformed by ohmic radio-frequency microelectromechanical (RF-MEMS) switches, if long-term degradation mechanisms can be understood and overcome. The widespread use of these devices is limited by stiction and increased contact resistance. Understanding how these mechanisms inhibit the device performance will lead to improved rules for RF-MEMS materials selection. A multi-scale atomistic and continuum simulation method is used to explore the “in-use” conditions of these devices. Simulations of nano-indentation hardness tests on several AuNi (0-15% Ni) alloys are performed to determine the effects of local order, grain boundaries, and surface segregation. A modified AuNi interatomic potential is presented which better matches first principles and experimental data. Local ordering and surface segregation are probed by comparing the results of random solid solutions and Monte-Carlo equilibrated structures. Crystal orientation and grain boundary effects are examined by moving position of the nano-indenter. In-use conditions are studied via the application of voltage (0 – 0.3 V) across the system to determine the effects of localized Joule heating at the contact point on the mechanical properties of the substrate. The results presented will enable the creation of design rules for the selection of materials that will improve RF-MEMS device performance. This work was supported by ONR grant number N00014-10-1-0402, a U.S. Dept. of Defense NDSEG Fellowship, and a U.S. Dept. of Education GAANN Fellowship.
3:15 PM - TT4.3
Effect of Dry Cleaning on Device Characteristics in MLC Nand Flash Memory.
Young-Taek Song 1 Show Abstract
1 , Hynix Semoiconductor Inc., Icheon-si Korea (the Republic of)
With extreme technology shrinkage, pattern leaning occurs by conventional wet bath wafer cleaning below sub 3X nm technology. The surface tension during the rinse process is responsible for pattern leaning. To avoid this problem, we have adopted dry cleaning technique without rinse process. To minimize surface tension during the dry cleaning process, hydrofluoric acid and ammonia chemical are used in low pressure at room temperature. To verify the dry cleaning effect on electrical properties, PGM and Erase according to the cleaning process are compared. PGM Vt of conventional wet cleaning to Dry cleaning results was 3.10 to 3.11 and ERASE Vt results was -5.06 to -5.08. This paper present to avoid pattern leaning, we have adopted dry cleaning technique and verify a device characteristics of dry-cleaning technique showed similar results of conventional wet cleaning.
3:30 PM - TT4.4
Origin of the Tetragonal Distortion in Fe–Pd Magnetic Shape Memory Alloys.
Ingo Opahle 1 2 3 , Klaus Koepernik 2 , Ulrike Nitzsche 2 , Manuel Richter 2 Show Abstract
1 , Goethe-Universität Frankfurt, Frankfurt/Main Germany, 2 , IFW Dresden, Dresden Germany, 3 ICAMS, Ruhr-Universität Bochum, Bochum Germany
Magnetic shape memory alloys (MSMA) have attracted considerable attention as materials for actuator and sensor applications, due to large magnetically induced strains of up to 10%. A promising MSMA is disordered Fe70Pd30 with an induced strain of about 6% and a relatively high blocking stress.We have calculated the electronic structure of disordered Fe-Pd alloys in the framework of density functional theory using the full potential local orbital (FPLO) code. The magnetocrystalline anisotropy energy is calculated as a function of c/a-ratio and Pd concentration. Disorder is found to be essential for the experimentally observed easy a-axis anisotropy in fct-Fe70Pd30. The origin of the tetragonal distortion in these completely disordered alloys is found to be a Jahn-Teller like effect, which allows the system to reduce its band energy in a narrow composition range. On the basis of our results, we discuss the prospects for an optimization of the alloys' properties by adding third elements, including effects on the magnetocrystalline anisotropy energy.
3:45 PM - TT4.5
Non-Stoichiometric Lead Free (K0.5Na0.5)1-xNbxO3 and Its Electromechanical Behaviour.
Maria Costa 1 , Muhammad Rafiq 1 , Paula Vilarinho 1 Show Abstract
1 , University of Aveiro, Aveiro Portugal
Lead free piezoelectrics and ferroelectrics fall in the class of functional materials which have potential applications as optical displays, acceleration sensors, chemical sensors and power generation and storage devices, among others. K0.5Na0.5NbO3 (KNN) is one of the leading lead free piezoelectric materials, being considered as a candidate to replace Pb(Zrx,Ti1-x)O3 (PZT), which is currently the most widely used piezoelectric. Pure KNN has inferior electromechanical properties as compared to PZT, but various strategies have been adopted to improve the properties including substitutions at A and B-site, especially in combination with texturing to improve the properties to a level comparable to the PZT response. The volatility of its constituents i.e. K and Na, is one of the main problems in KNN based system. Several studies on the effect of K/Na ratio (A-site occupancy) on the properties have been conducted, but no systematic study has been reported about the relationship between the properties and the non-stiochiometry at A and B site of KNN. In this work, non-stoichiometric KNN ceramics with precisely controlled A/B ratio, i.e. K/Na (A-site) and Nb (B-site), from 0.9 to 1.1 were synthesised by solid state reaction. Structural properties and microstructure development of the sintered ceramics were examined by X-Ray Diffraction and electron microscopy (SEM/TEM). Dilatometric analysis combined with electron microscopy followed the densification behaviour and microstructure development. The dielectric properties were evaluated as a function of the temperature in the radio frequency range. The electromechanical response was assessed at room temperature. The Curie temperature (Tc) exhibits a tendency to increase as the non-stoichiometric ratio increases within the solid solubility limit. High dielectric constant values are observed for the compositions with A/B ratio close to 1.0. The role of non-stiochiometry and the associated defect chemistry on the structure and microstrucure and electromechanical properties of KNN is established.
4:30 PM - TT4.6
Metal-Oxide-Metal (MOM) Diode Optimization for Heat Energy Recovery.
David Wood 1 , Linzi Dodd 1 , Andrew Gallant 1 Show Abstract
1 School of Engineering and Computing Sciences, Durham University, Durham United Kingdom
This work presents details of the optimization of metal-oxide-metal (MOM) diodes through analysis of their electrical performance and physical structure. These diodes have great potential to rectify terahertz radiation collected by micron-scaled antennas from waste heat sources and successfully converting this into a useful DC current.The main focus of this work is on the production of a uniform thin oxide layer in the MOM diode. The oxidation techniques studied range from simply exposing metals in both controlled and uncontrolled atmospheric environments, to careful etching and re-growth of a metal’s native oxide. The aim is to achieve a high level of electrical performance, as measured by diode behaviour, breakdown voltage and reliability, as well as uniformity across devices. Complementary physical analysis includes AFM, SEM and RBS to help elucidate the oxide structure. The MOM diode structure consists of two dissimilar metals separated by a native oxide layer, which is sufficiently thin to allow electron tunnelling to occur, i.e. is less than 5 nm thick. A functional diode is best achieved by using one metal which will oxidize readily and another which is inert. Maximising the work function difference between these metals is also beneficial as it increases the asymmetry of the resulting diode. Many metal combinations have been tried, e.g. Ni/Al; here we have chosen readily oxidizable titanium, and maximised the work function difference by using inert platinum as the other metal.The oxidation of titanium has taken place via three different methods; exposure to a cleanroom ambient atmosphere, controlled oxidation in a moist, heated environment and via reactive ion etching and subsequent in situ oxidation.The I-V curves of the resulting diodes show significant asymmetry and non-linearity. dI/dV and d2I/dV2 can then be calculated, with the latter being a universal figure of merit known as the curvature coefficient (CC) of the diode, which is a measured of how well the diode can rectify a signal. For this application, it is beneficial for the diodes to operate at zero bias, where CC values of up to 6.6 V-1 have been achieved. Biasing the diodes to +30 mV significantly increases the CC up to 22.7 V-1. These results are competitive with others published to date and successfully validate the performance and production methods of these diodes.The diode performance is variable, and physical analysis – SEM, AFM and RBS – has been used to try and correlate the electrical performance with the compositional nature of the oxide. This is both across a wafer of devices, and in checking for run to run repeatability.
4:45 PM - TT4.7
Harmonic Excitation of Surface Acoustic Wave Devices on Gallium Nitride for Chemical Sensor Applications.
J. Justice 1 , K. Lee 1 , O. Mukdadi 2 , D. Korakakis 1 Show Abstract
1 Lane Department of Computer Science and Electrical Engineering, West Virginia University, Morgantown, West Virginia, United States, 2 Department of Mechanical and Aerospace Engineering, West Virginia University, Morgantown, West Virginia, United States
Gallium nitride (GaN) is a robust piezoelectric semiconductor material with excellent thermal and chemical stability, making it an attractive material for surface acoustic wave (SAW) devices operating in harsh chemical environments . In addition, AlGaN/GaN based HEMT devices can be integrated with SAW devices to make very sensitive piezoelectric microbalances which can be used in wireless sensor arrays [1, 2]. However, there is an anisotropic wave propagation associated with SAWs in GaN and traditional sapphire substrates . By designing SAW devices with smaller wavelengths, the SAW energy can be confined to the GaN layer, eliminating the substrate effects and increasing the sensitivity of the device. By operating at the 5th and higher harmonics, gigahertz operation can be realized with larger IDTs and thinner GaN layers, resulting in more cost effective solutions for GaN based SAW chemical sensors. Devices have previously been designed to operate at the 5th harmonic on lithium niobate (LiNbO3) , but there are no reports of using this technique on GaN in the literature. In this study, GaN thin films have been grown via metal organic vapor phase epitaxy (MOVPE) on sapphire substrates. SAW devices have been designed to operate at the fundamental frequency and higher harmonics. SAW response of fundamental and harmonic operation with respect to crystal orientation is experimentally measured and compared. Dispersion curves for SAWs in GaN and GaN/sapphire are simulated and compared to experimental results.S.J. Pearton, B.S. Kang, S. Kim, F. Ren, B.P. Gila, C.R. Abernathy, J. Lin and S.N.G. Chu, J. Phys.: Condens. Matter, 16 (2004) R961–R994.T. Lalinskýa, I. Rýgera, G. Vankoa, M. Tomáškab, I. Kostic, Š. Hašíka and M. Valloa, Procedia Engineering, 5 (2010) 152–155.J. Pedrós, F. Calle, J. Grajal, R.J. Jiménez Riobóo, C. Prieto, J.L. Paua, J. Pereiro, M. Hermann, M. Eickhoff and Z. Bougrioua, Superlattices and Microstructure, 36 (2004) 815–823.P.M. Naraine and C.K. Campbell, IEEE Ultrason. Symp., 0090-5607/84, (1980) 322-325.
5:00 PM - TT4.8
Ultrananocrystalline Diamond (UNCD) Films as Fast Charging/Discharging Dielectric Layers to Enable Robust Commercial RF MEMS Capacitive Switches without Dielectric Charging Failure.
Orlando Auciello 1 2 , Charles Goldsmith 3 , Anirudha Sumant 2 , Suresh Sampath 4 , Chris Gudeman 4 , Wei Wang 5 , James Hwang 5 , Hongjun Zeng 6 , John Carlisle 6 Show Abstract
1 Materials Science, Argonne National Laboratory, Argonne, Illinois, United States, 2 Center for Nanoscale Materials, Argonne National Laboratory, Argonne, Illinois, United States, 3 , MEMtronics Corporation, Plano, Texas, United States, 4 , Innovative Micro Technology, Goleta, California, United States, 5 Electrical Engineering, Lehigh University, Bethlehem, Pennsylvania, United States, 6 , Advanced Diamond Technology, Romeoville, Illinois, United States
Radio frequency RF MEMS systems have many benefits over conventional semiconductor devices for controlling and routing microwave and millimeter-wave signals. RF MEMS switches exhibit very low insertion loss and power consumption, and ultrahigh linearity, which enable high efficiency phase shifters or tunable filters for mobile communication and radar systems at ≥ 1 GHz frequencies. Reliability inhibited deployment of RF MEMS switches into military or commercial systems, mainly due to failure related to electrical charging of the dielectric layer on top of the bottom electrode. Proposed solutions to charging include hermetic packaging, minimizing the electric field across the dielectric, and tailoring the polarity and waveform of bias control signals. These approaches provided significant improvements in reliability, but did not eliminate RF MEMS switch failure due to slow electrical charging (10s-100s of sec) and slow discharging (10s-100s of sec) of the oxide or nitride dielectric layers used in RF MEMS capacitive switches. This presentation shows that a novel ultrananocrystalline (UNCD) diamond film dielectric layer with 2-5 nm grains and 0.4 nm wide grain boundaries exhibit unique fast charging (~ 100 µsec) but also fast discharging (~ 100 µsec) behavior, the latter being 5-6 orders of magnitude faster than the discharging of oxide and nitride dielectric layers. Thus, the UNCD dielectric layer eliminates RF MEMS switch charging-induced failure, due to fast charge motion in and out of nanograins though atomic wide grain boundaries. The switch pulls down charging the dielectric layer to failure, but the latter discharges quickly when the voltage is removed. Thus, the switch can recover fully from charging before the next switch operation, providing for the first time an operating RF MEMS switch near-continuously “on” without adverse impact on switch reliability. The results presented here demonstrate a new paradigm in RF MEMS switch operation with no dielectric charging failure,which mitigates the dielectric charging failures often demonstrated in conventional dielectric materials.This work was supported by US Department of Energy, Office of Science, Office of Basic Energy Sciences-Materials Science, under contract DE-AC02-06CH11357 and DARPA under contracts MIPR 06-W238. Use of the Center for Nanoscale Materials was supported by the U. S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357.
5:15 PM - TT4.9
Understanding Interfacial Structure of Transient Liquid Phase Bonded Cu Microchannel Devices.
Ke Chen 1 , Wen Meng 1 , Fanghua Mei 2 Show Abstract
1 Mechanical Engineering, Louisiana State University, Baton Rouge, Louisiana, United States, 2 , Enervana Technologies LLC, Baton Rouge, Louisiana, United States
Metal-based microdevices can have advantages over Si-based counterparts, one example being microchannel heat exchangers (MHEs). To form functional metallic microchannel devices from metallic high-aspect-ratio microscale structures (HARMS), proper assembly and packaging are critical.The transient liquid phase (TLP) bonding approach to microchannel device assembly is of particular interest. The TLP approach to bonding has been studied extensively . The use of an intermediate layer to depress the melting point of the bonding interface region alleviates high temperature and high pressure requirements. Such a TLP approach can bond metallic microchannel devices without causing excessive structural distortions or damage to the final device. We have shown that low-profile, Cu-based, completely enclosed, microchannel devices can be built by Cu/Al/Cu TLP bonding [2, 3].In the case of Cu/Al/Cu TLP bonding, a detailed study of how the bonding protocol influences the structure of the bonding interface region and how this structure influences the mechanical properties of the bond has not been carried out. In this paper, we present experimental results on TLP bonding of Cu-based structures using thin Al foil intermediate layers. Structure of the bonding interface region within Cu-based microchannel specimens TLP bonded with thin Al foil intermediate layers was examined, and shown to be similar to TLP bonded Cu/Al/Cu sandwich-like specimen assemblies. For the latter, further structural and compositional characterizations of the bonding interface region were achieved through conventional and focused ion beam (FIB) sectioning, followed by examination with scanning electron microscopy (SEM), transmission electron microscopy (TEM), and X-ray energy dispersive spectroscopy (EDS). By combining FIB microscopy with specimen lift-out and TEM analysis, the development of bonding interfacial structure was examined and understood. A number of TLP bonded Cu/Al/Cu specimen assemblies were further tensile tested until fracture occurred across the bonding interface. Measured average tensile strength was correlated to the structure of the bonding interface region. Our results highlight the importance of tuning the TLP bonding protocol in order to achieve the desired structure of the bonding interface region and consequently the optimal physical properties for the bonded structure, and illustrate the potential of applying TLP bonding to fabrication of metal-based microdevices.References: MacDonald WD, Eagar TW. Annu. Rev. Mater. Sci., 1992; 22:23. Mei Fanghua, Jiang J, Meng WJ. J. Vac. Sci. Technol. A, 2008; 26:798. Lu Bin, Chen Ke, Meng WJ, Mei Fanghua. J. Micromech. Microeng. 2010; 20:115002.
5:30 PM - TT4.10
The Development of BZT-BCT Lead-Free Thin Films for Piezoelectric Applications.
Baris Celtikci 1 , Ahmet Ozenbas 1 Show Abstract
1 Metallurgical and Materials Engineering, Middle East Technical University, Ankara Turkey
For the second half of century, the piezoelectric materials have been considered as an important functional material for various applications, from the microphones to the high technology scanning electron microscopes, actuators, sonar sensors, cell phones, MEMs and etc... Lead-based phases have dominated almost all these applications. However, lead based materials are problematic for health because of its toxicity and it will be prohibited to use within short period of time. Among various lead-free piezoelectric materials, BZT-BCT based phases are good candidates instead of Pb-based materials due to having a tricritical point compared to other lead-free alternatives. In this study, lead-free BZT-BCT thin films were grown on (111)-Pt/Ti/SiO2/Si-(100) substrates by CSD (Chemical Solution Deposition) technique. Barium acetate, calcium acetate, zirconium (IV) propoxide, and titanium(IV) isopropoxide were used as starting materials. Acetic acid and 2-methoxyethanol were used as solvents for 0.4 M precursor solution. Structural, morphological and chemical analyses were conducted using XRD, SEM and XPS techniques. Optimum film thickness was determined as 400 nm by cross-section studies via SEM. TC-DTA and XRD analysis were used in correlation for determining the critical temperatures for pure perovskite phase formation. The pyrolysis temperature was determined as 500°C for organic removal and the pure perovskite phase was obtained at 750°C after 1 hour sintering. The electrical properties like dielectric constant, tangent loss, P-E hysteresis loops, fatigue, leakage current and piezoelectric displacement will be determined by using polarization tester, and AFM for possible use of these lead-free thin films for potential applications.
5:45 PM - TT4.11
Method for Achieving CMOS MEMS Accelerometers with Excellent Built-in Thermal Stability and Reduced Charge Damage.
Klaus Hsu 1 , Siew Seong Tan 1 Show Abstract
1 Institute of Electronics Engineering, National Tsing Hua University, Hsinchu Taiwan
Capacitive CMOS MEMS sensors are usually defined by anisotropic dry etching processes (RIE and DRIE). These processes can provide clean and vertical sidewall geometry. However, during the dry-etching processes, charges are added to the gate electrodes of the on-chip MOSFET’s through metal pads and micro-structures, and the voltage may be raised to the level of breaking down the gate oxide, which leads to large leakage current and fails the circuit. On another hand, the thin spring beams in capacitive CMOS MEMS accelerometers suffer from in-plane curling and out-of-plane curling caused by stress gradient. Furthermore, the stress in the layers of MEMS structure is a function of temperature. Therefore, the in-plane curling and out-of-plane curling vary with temperature, leading to varying electrode coupling area in the sensing beams. This in turn causes variation in the sensitivity and the DC offset of sensors, meaning that usually the thermal stability of CMOS MEMS capacitive accelerometers is very poor. To cope with these problems, this work develops a new wafer-level post-CMOS process for fabricating thermally stable capacitive accelerometers. The resultant MEMS structures have high aspect ratio (e.g. 2-2.5 μm gaps versus 57 μm depth) and are insensitive to residual stress and temperature change. Excellent thermal stability was achieved intrinsically by making the crystalline Si layer in the sensors thick. Moreover, this process totally avoids the charge damage problem during the dry-etching procedure. For demonstration, an accelerometer sensor was fabricated by using the proposed process and was integrated with an on-chip sensing circuit in commercial 0.35 μm 2P4M CMOS process. High detection sensitivity of 595 mV/g and very low thermal variation of 1.68 mg/°C were successfully achieved.
TT5: Poster Session: MEMS
Wednesday AM, November 30, 2011
Exhibition Hall C (Hynes)
9:00 PM - TT5.1
Large Area Electrodeposited Nickel Single Crystal Substrates for Microelectromechanical Systems.
Meifang Li 1 , Deepa Vairavapandian 1 , Eric Chason 1 Show Abstract
1 Division of Engineering, Brown University, Providence, Rhode Island, United States
Electrodeposited Ni films are widely used in microelectromechanical systems (MEMS) due to their high elasticity and flexibility, in applications such as micro-needles, micro-pumps, inertial switches and energy-harvesting device. Residual stress in these films is a problem because it can lead to unwanted elastic deformation. Here we describe a method for producing large-area, inexpensive single crystal Ni foils, which can be patterned directly or used as substrates on which to grow stress-free Ni layers for MEMS processing. The method uses a sequence of electropolishing, epitaxial electrodeposition and selective etching to create freestanding foils from an initial template crystal that can then be reused to make more material. Pole figure and XRD results indicate high crystal quality, with FWHM of the (200) peak less than 1 deg. The process has been demonstrated for both Ni(100) and Ni(111) orientations. The starting single crystal nickel substrate can be reconditioned after growth so that we can produce multiple generations of freestanding films from the same starting sample with the same crystalline quality. This will allow our method to be turned into a continuous process for making long ribbons or large areas of single crystal films that can provide inexpensive substrates for microelectromechanical systems device fabrication.
9:00 PM - TT5.11
The Role of Trapped Contaminant Pockets in the Degradation of Ohmic Contact in RF-MEMS Switches.
Christopher Freeze 1 , Yinxuan Yang 2 , Angus Kingon 2 , Douglas Irving 1 Show Abstract
1 Department of Materials Science and Engineering, North Carolina State University, Raleigh, North Carolina, United States, 2 , Brown University, Providence, Rhode Island, United States
Ohmic radio frequency microelectromechanical systems (RF-MEMS) switches are currently limited by reliability. Failure primarily occurs by stiction of contacts or a rapid increase in the contact resistance above a 1 to 2 Ohm target regime. Although environmental hydrocarbons play an important role in preventing stiction, the resistance spike is often attributed to the presence of contaminants. The mechanisms of this resistance spike are not understood and tend to be ignored. Previous trial-and-error studies to find better contact materials have proven unsuccessful due to disregard of contaminant influences and complications from the non-equilibrium conditions associated with switch usage. Understanding the fundamental mechanisms of the spike in contact resistance is imperative in the search to prevent or delay switch failure. This work uses coupled atomistic and continuum simulation to study the dynamics of contact loading in the presence of hydrocarbon contaminants. Pressure buildup is studied by compressing a pocket of hydrocarbons trapped on the surface of a flat gold substrate. A simplified system provides insight into the basic mechanisms of releasing pressure buildup in the hydrocarbon pocket. A more representative surface is then studied to understand the dynamics of contact evolution through the loading process and the roles that contaminants play. Varying loads and local thermal conditions caused by Joule heating are explored to account for further complications associated with the in-use environment. This work is supported by ONR grant N00014-10-1-0402.
9:00 PM - TT5.12
Ferroelectric and Dielectric Response of RF Sputtered Sr3Bi4Ti6O21 Thin Films.
Maria Costa 1 , Venkata Saravanan 1 , Erik Pandiana 1 , Isabel Miranda Salvado 1 Show Abstract
1 , University of Aveiro, CICECO, Aveiro Portugal
Bismuth Layered Ferroelectrics also known as Aurivillus oxides (AO) such as Srm-3Bi4TimO3m+3 (SBT, m= integer) have attracted much attention in the recent past due to their low operating voltage, fast switching speed, negligible fatigue up to 1012 switching cycles, excellent retention characteristics, and low leakage current density on Pt electrodes. SBT thin films are one of the forerunner materials for integrated device applications in non-volatile ferroelectric random access memory (NVFeRAM). Though studies on m=1 to 5 are quite voluminous, sparse efforts have been taken to synthesize and understand the properties of SBT thin films with m≥6. The current interest on lead free materials due to toxicity concerns renews the interest for these still unexploited oxides which potential applications remain obscure. Aiming to contribute to a better knowledge of AO with high m, in particular as thin films for microelectronic applications, a systematic study of the synthesis and properties of Sr3Bi4Ti6O21 (SBT6) thin films was undertaken in the present work. SBT6 thin films were deposited by RF magnetron sputtering on Pt/Ti/SiO2/Si substrates as a function of various sputter deposition parameters including power density, working pressure, oxygen mixing percentage (OMP) and ex-situ annealing treatment. X-ray diffraction analysis and scanning electron microscopy were used to follow up the crystalline phase formation and the microstructure development. Power density and working pressure played a vital role in altering the film thickness, while the oxygen partial pressure and the annealing temperature/time profoundly affected the films microstructure. Oxygen rich atmosphere favoured a fine grain microstructure while increasing the temperature from 600 to 750°C markedly enhanced the grain growth. The dielectric properties of the films were found to be voltage dependant, pointing out a dielectric tunability of about 20% for the films annealed at 700°C for 1 h. The ferroelectric response of the obtained films under the various conditions and its fatigue dependence were macroscopically characterized by P-E hysteresis loops. Local domain structure and piezoelectric properties were assessed by Piezoforce Response Microscopy (PFM). The results are discussed in the framework of the potential technological usefulness of high m Aurivillius oxide thin films.
9:00 PM - TT5.14
Coarse-Grained Molecular Dynamics Simulation of Epoxy-Based Chemically-Amplified Resist for MEMS Application.
Hiromasa Yagyu 1 , Yoshikazu Hirai 2 , Akio Uesugi 2 , Yoshihide Makino 2 , Koji Sugano 2 , Toshiyuki Tsuchiya 2 , Osamu Tabata 2 Show Abstract
1 R&D Division, Mitsuboshi Belting Ltd., Kobe Japan, 2 Department of Micro Engineering, Kyoto University, Kyoto Japan
A unique simulation method of epoxy-based chemically-amplified resist (hereafter resist) mechanical property by coarse-grained molecular dynamics (CGMD) were proposed. Our goal is a prediction of filtration function which is a novel function of the resist using the CGMD model. We have reported a new fabrication technique of an embedded microchannel in the resist by varying photolithography parameters . To date, we also have demonstrated experimentally that the semi-cross-linked resist fabricated using this technique acts as a nanofilteration membrane with nano-scale pores . In order to carry out a quantitative evaluation of the filtration function of the resist, the properties of the resist must be evaluated from the viewpoint of a cross-linked structure in the meso-scale. Along this line, we investigated the modeling of the resist using molecular dynamics, and the evaluation of mechanical property using the developed model was carried out as a first step. Due to expensive computational costs, it is difficult to evaluate the properties of the resist in the molecular dynamics simulation of a polymer material using a full-atomistic model. To address these issues, we investigate the modeling of the resist using CGMD, which a bead of chain was treated as several monomer units. Although CGMD model using beads-spring model cannot take into account the chemical effects of chain linkage in a model, the dynamics of chains at the micro-scale level and long time scale can be simulated using appropriate potential energy between the beads . In this work, CGMD models with different numbers of cross-linked ratio including FENE, Lennard-Jones, and Theta angle bending potentials were applied to the resist, and uniaxial elongation simulations were carried out to confirm the validity of the parameters used for the modeling. From the elongation simulation results, it was confirmed that the calculated elastic modulus of the model with large angle bending spring constant at the strain below 5% depends on the cross-linked ratio, and its elastic modulus increased with increase in the cross-linked ratio. The dependency of the model with the angle bending spring constant of 25ε/rad2 was good agreement with that measured with nanoindentation test using the resists on silicon wafer with different numbers of cross-linked ratio prepared by varying UV dose and post exposure bake temperature . Therefore, the validity of the developed simulation model and its applicability for the resist were confirmed.  Y. Hirai et al., J. Microelectromech. Syst. 19, 1058 (2010).  Y. Hirai et al., in Tech. Dig. Transducers2011, p. 2706 (2011).  K. Kremer et al., J. Chem. Phys. 92, 5057 (1990).
9:00 PM - TT5.15
In Situ TEM Observation of AlGaN/GaN HEMT Degradation under Applied DC Bias.
Andrew Lang 1 , Chris Winkler 1 , Jen Sloppy 1 , Craig Johnson 1 , Mitra Taheri 1 Show Abstract
1 Materials Science and Engineering, Drexel University, Philadelphia, Pennsylvania, United States
GaN HEMTs are the backbones of many high frequency applications. There are multiple types of HEMTs, most commonly AlGaAs/GaAs heterostructures, but AlGaN/GaN heterostructures are of interest due to the wide band gap of GaN and the ability of AlGaN/GaN to carry high voltage and current. The resulting heterostructure exhibits superior carrier saturation velocity, thermal conductivity, and high breakdown field all of which are required for high temperature and high speed applications. AlGaN/GaN HEMTs are significantly different from other HEMT structures due to a very strong piezoelectric effect caused by the lattice mismatch between AlGaN and GaN. This lattice mismatch in addition to spontaneous polarization fields give rise to an interfacial charge that is referred to as a two-dimensional electron gas (2DEG). In order to increase reliability and performance of AlGaN/GaN heterostructures, the electrical degradation of these HEMTs need to be examined under an applied bias. By utilizing in situ TEM the degradation mechanisms and the associated defects are studied at the 2DEG by direct observation of their nucleation under applied DC bias. The interface region at the drain is investigated, as well as the piezoelectric-induced strain and relaxation trap region in the AlGaN layer beneath the gate region. TEM samples are prepared by removing a section of the device that includes the source, gate, and drain and connecting electrodes in a custom in situ TEM holder. Samples are tested below, at, and above critical voltage values. The experiments allow for the monitoring of the evolution of defects associated with GaN HEMT electrical degradation. Specifically, these results will provide information towards determining predictive mechanistic models of the evolution of defects under various electrical stress conditions, which will contribute to improved reliability of AlGaN/GaN HEMT devices.This research was funded by the Office of Naval Research under contract #N00014-11-1-0296
9:00 PM - TT5.16
Lensless Imaging for Simultaneous Space-Constrained Microfluidic Sperm Sorting.
Xiaohui Zhang 1 , Imran Khimji 1 , Hooman Safaee 1 , Paolo Catalano 1 , Umut Gurkan 1 , Aida Nureddin 1 , Raymond Anchan 2 , Emre Kayaalp 3 , Richard Maas 4 , Utkan Demirci 1 5 Show Abstract
1 Center for Bioengineering, Department of Medicin, Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts, United States, 2 Center for Infertility and Reproductive Surgery, Obstetrics Gynecology and Reproductive Biology, Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts, United States, 3 Department of Obstetrics and Gynecology, Jamaica Hospital Medical Center, Queens, Massachusetts, United States, 4 Division of Genetics, Department of Medicine, Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts, United States, 5 Harvard-MIT Health Sciences and Technology, Harvard-MIT Health Sciences and Technology, Cambridge, Massachusetts, United States
5.3 million American couples of reproductive age (9%) are affected by infertility, among which male factors accounts for up to 50% of cases, which necessitates the identification of parameters defining sperm quality, including sperm count and motility. In vitro fertilization (IVF) with or without intra cytoplasmic sperm injection (ICSI) has become the most widely used assisted reproductive technology (ART) in modern clinical practice to overcome male infertility challenges. One of the obstacles of IVF and ICSI lies in isolating and analyzing the most motile, and presumably healthiest sperm from semen samples that have low sperm counts (oligozoospermia) and/or low sperm motility (oligospermaesthenia). Manual selection of motile sperm using the microdrop technique is still a major approach utilized in fertility clinics, which can be inefficient and time consuming. Microfluidics offers an alternative to improve the selection process efficiency through accurately manipulating samples in microscale volumes. Moreover, the small field of view (FOV) of conventional microscopes commonly used to image sperm motion presents challenges in tracking a large number of sperm cells simultaneously. To address these challenges, we have designed a simple and cost-effective microfluidic sorting system integrated with a lensless charge coupled device (CCD) to enable wide FOV and automatic recording as the sperm are sorted inside a microfluidic channel. In this study, we optimized the sorting system experimentally for channel length and incubation time to maximize its sorting capability for samples with low sperm count (< 4 million sperm/ml) based on sperm motility, percentage of motile sperm and collectable sperm percentage. The integrated system enables sorting and tracking of a population of sperm that have been placed in a microfluidic channel. This channel can be monitored in both a horizontal and vertical configuration similar to a swim-up column method used clinically. Sperm motilities can be quantified by tracing the shadow paths for individual sperm. Moreover, as the sperm are sorted by swimming out from the inlet towards the outlet of a microfluidic channel, motile sperm that reach the outlet can be extracted from the channel at the end of the process. This technology can lead to methods to evaluate each sperm individually in terms of motility response in a wide field of view, which could prove especially useful when working with oligozoospermic or oligospermaesthenic samples, in which the most motile sperm need to be isolated from a pool of small number of sperm.
9:00 PM - TT5.17
Fabrication and Performance of Beryllium Copper Micro LGA Contact Array by Cosputtering on Sacrificial Polymer Template, 3D Lithography and Final Thermal Processing.
Gareth Hougham 1 , Gerard McVicker 1 , Frank Libsch 1 , Xiaoxiong Gu 1 , Pavan Samudrala 1 , Jay Pogemiller 1 , Mike Gaynes 1 , Maurice Mason 1 , Peter Sorce 1 , Eugene O'Sullivan 1 , Sung Kang 1 , Robert Meinel 1 , Marinus Hopstaken 1 , Michael Gordon 1 , Michael Gedeon 4 , Jimmy Johnson 4 , Heather Wilkins 3 , Hongwei Xu 3 , Janet Okada 5 , Leah Langsdorf 2 , Chuan Yue 6 , Allison Tripp 6 , Jeffrey Mason 6 , Kurt Swanson 7 Show Abstract
1 , IBM, T.J.Watson Research Center, Yorktown Heights, New York, United States, 4 , Materion Brush Performance Alloys, Mayfield Heights, Ohio, United States, 3 , General Atomics, San Diego, California, United States, 5 , Dow Chemical, Marlborough, Massachusetts, United States, 2 , Promerus, Bakelite Sumitomo, Brecksvile, Ohio, United States, 6 , TE Connectivity, Norwood, Massachusetts, United States, 7 , Hutchinson Technology, Hutchinson, Minnesota, United States
A new method for the fabrication of high density beryllium copper electrical contacts is described. A micro land grid array (uLGA) was fabricated on a 0.74 mm pitch by cosputtering copper and beryllium onto an array of sacrificial polymer cones which were molded in a prior step onto a ceramic substrate with pre-metallized through vias. 3D photolithography on these high aspect ratio cones defined several contact geometries from helical to linear cantilever beam. Subsequent thermal annealing in forming gas simultaneously vaporized the polymer template and solutionized the beryllium and copper before rapid quench to room temperature. A post quench thermal treatment at lower temperature hardened the contacts by controlled precipitation of a beryllide phase to approximately 0.7 of a full hard alloy 25 HT (C17200). Process details including polymer over-molding, cosputtering, 3D lithography, etching and thermal processing as well as final measured properties including micro hardness, tensile strength, contact spring force and high speed signal integrity of functional contacts will be reported. Note: Required publication permissions are pending.
9:00 PM - TT5.2
Characterization and Application of Reactively Sputtered Iron Oxide Thin Films.
Xiaoning Wang 1 , Congshun Wang 1 , Stephan Anderson 2 , Xin Zhang 1 Show Abstract
1 Mechanical Engineering, Boston Univ, Boston, Massachusetts, United States, 2 Radiology, Boston University, Boston, Massachusetts, United States
Magnetic resonance imaging (MRI) technology has been rapidly gaining research and clinical attentions recently due to its non-radiation nature, higher soft tissue contrast, and versatile imaging capabilities. A new type of micro-engineered MRI contrast agent particles incorporates magnetic biocompatible materials, such as iron and iron oxide, to bring multi-spectral and functional capabilities to MRI, similar to what quantum dot technology has brought to optical imaging, were presented in this work. A microfabrication compatible process for depositing different phases of iron oxide thin films (hematite, maghemite, magnetite, etc.) were developed, as well as characterized, for the aforementioned application to fabricate the MRI contrast agent particles.Approximately 1µm thick of iron oxide thin films were reactively sputtered onto silicon substrates from elemental iron target within oxygen-rich environment using a RF magnetron sputtering system (Denton Vacuum, Inc., NJ). Oxygen partial pressure, RF power, and substrate temperature were controlled to yield different iron oxide thin films with various chemical compositions and crystalline structures, therefore different magnetic properties. The effects of rapid thermal annealing of these thin films samples were also studied.The crystalline structures of these thin films were then characterized by X-ray diffraction (XRD). In combination with energy dispersive X-ray spectroscopy (EDX), we were able to identify the chemical composition of these sample thin films. Therefore, the correlation between sputtering parameters and resulting phases of iron oxide thin films could be determined. Superconducting quantum interference device (SQUID) was used to quantify the saturation magnetic polarizations of these iron oxide thin films to characterize their magnetic property.Based on the characterized magnetic property, we were able to choose the ideal saturation magnetic polarization, and thus the corresponding sputtering parameter. The contrast agent particles were then fabricated using the same sputtering process. Start with lift-off patterning the first magnetic disk, followed by spin-coating a polyimide layer and lift-off patterning the second magnetic disk. The polyimide layer was then etched using reactive ion etching to form a slim vertical post separating the two disks. Arrays of these particles were successfully fabricated and validated using MRI to demonstrate their contrast enhancement capability.To summarize, the techniques of depositing different phases of iron oxide thin films using reactive RF sputtering were examined. The resulting thin films were then characterized for their chemical composition, crystalline structure, as well as magnetic and mechanical properties for the use of fabricating a novel type of contrast agents for MRI applications.
9:00 PM - TT5.3
Evolution of Hole Patterns by Surface Diffusion on Si(001).
Koichi Sudoh 1 , Reiko Hiruta 2 , Hitoshi Kuribayashi 2 Show Abstract
1 The Institute of Scientific and Industrial Research, Osaka University, Ibaraki, Osaka, Japan, 2 , Fuji Electric Co., Ltd., Matsumoto, Nagano, Japan
Three dimensional microstructure formation applying spontaneous shape change of high aspect ratio hole patterns on Si substrates during high temperature annealing is a promising technique for manufacturing MEMS devices. Surface-diffusion-driven shape change of hole patterns is capable to yield well-controlled cavity structures in the Si substrate [1,2]. A most promising application of this technique is fabrication of freestanding thin Si membranes. In order to obtain desired freestanding membrane structures by this method, we have investigated the relationship between the initial hole pattern and the structure obtained by surface-diffusion-driven shape change both experimentally and by numerical simulations. For the experiment we fabricated various kinds of periodic square array of high aspect ratio holes on Si(001) substrates by reactive ion etching. The radius and depth of a hole are varied 0.75~1.5 μm and 2.9~3.9 μm, respectively. The pitch of the array is varied from 1.0 to 1.8 μm. The samples were annealed in 10~60 Torr hydrogen gas ambient at 1150 ○C. We show that there is a window in parameter-space for initial hole patterns to obtain diaphragm structures by surface-diffusion-driven shape change. If the initial hole pattern is properly designed, a large plate-shaped cavity is formed through cavity formation by closure of the hole inlets and subsequent coalescence of the cavities. When the aspect ratio of holes is smaller than ~3.0, no cavities are formed while the hole pattern is gradually flattened due to the surface tension. Moreover, there is a critical aperture ratio (ratio of hole diameter to the pitch), above which no cavities are formed. In order to cause coalescence of the cavities formed by hole inlet closure, sufficiently large aperture rations are necessary. However, if the aperture ratio is too large, the thin Si wall between holes becomes unstable during the shape change, resulting in formation of irregular structures. We have also performed numerical simulations of surface-diffusion-driven evolution of a hole pattern, numerically integrating the Cahn-Hilliard equation with degenerate mobility . Our simulations reproduce well the observed dependence of the shape evolution on the initial hole pattern. It is found that our simulation is practically useful to predict the evolution of hole patterns on Si substrates during high temperature annealing. I. Mizushima, T. Sato, S. Taniguchi, and Y. Tsunashima, Appl. Phys. Lett. 77, 3290 (2000). K. Sudoh, H. Iwasaki, R. Hiruta, H. Kuribayashi, and R. Shimizu, J. Appl. Phys. 105, 083536 (2009). J.W. Cahn, C.M. Elliott, A. Novick-Cohen, Euro. J. Appl. Math. 7, 287 (1996).
9:00 PM - TT5.4
Fe(III)Porphyrin Functionalized Piezo-Resistive SU-8 Micro-Cantilever as a CO Sensor.
Vijaya Bhaskar Reddy 1 , Mrunal Khaderbad 2 , Sahir Gandhi 2 , Sheetal Patil 2 , Harshil Raval 2 , K. Narasaiah Chetty 3 , K. Govindha Rajalu 3 , P. C. Krishnama Chary 1 , M. Ravikanth 4 , V. Ramgopal Rao 2 Show Abstract
1 Department of Mechanical Engieering, Sri Kalahasteeswara Institute of Technology, Chittoor (Dt.), Andhra Pradesh, India, 2 Centre of Excellence in Nanoelectronics, Department of Electrical Engineering, Indian Institute of Technology Bombay, Mumbai, Maharashtra, India, 3 Department of Mechanical Engineering, JNTU, Anantapur, Andhra Pradesh, India, 4 Department of Chemistry, Indian Institute of Technology Bombay, Mumbai, Maharashtra, India
Being toxic, carbon monoxide (CO) needs to be detected for various healthcare and environmental applications. Due to the chemical affinity of Iron(III)porphyrin (Fe(III)porphyrin) towards CO, it can be used as a sensing element in CO detectors. In this study, Fe(III)porphyrin functionalized on a piezo-resistive SU-8 microcantilever has been demonstrated as a CO sensor that offers high sensitivity, easy integration and at low costs. Binding of CO to the Fe(III)porphyrin generates a surface stress, which results in a cantilever bending. The change in resistance in the piezoresistive layer due to the cantilever deflection can be read electrically using these polymer composite cantilever sensors. UV-vis spectroscopy, FTIR spectroscopy and AFM measurements on porphyrin modified SU-8 films explain the Fe(III)porphyrin-CO binding/sensing mechanisms. These porphyrin modified microcantilevers have been found to give a strong signal for very low concentrations (2ppm) of CO vapor exposure. Moreover, as our studies show, the sensor surface can be regenerated by purging these CO exposed cantilevers to dry nitrogen. The selectivity of these cantilevers has been tested by exposing them to various gases such as N2, O2, moisture and CO2. Uncoated SU-8 cantilevers have not shown any response to CO or other gases. The sensor response and recovery times have been measured to be 3 ms and 10 ms.
9:00 PM - TT5.5
Digitized Nanobiomedical Device Based on Nanowell Array Electrodes.
Joo-kyung Lee 1 , InRock Hwang 1 , Ah Young Kim 1 , BaeHo Park 1 , HeaYeon Lee 2 1 , Lim Seo Yul 1 , Kim Mi Ran 1 Show Abstract
1 Division of Quantum Phases and Devices, Konkuk University, Seoul Korea (the Republic of), 2 The Institute of Scientific and Industrial Research, Osaka University, Osaka Japan
To develop digitized nanobiomedical device, it is important to develop a nanobiosensor with ultra-high signaling accuracy. To achieve enough sensitivity, we have to solve some critical issues, such as (i) how to immobilize the targeting molecules such as ligand or analysis in a specific position, (ii) how to stabilize targeting molecules and localize biomolecules selectively without non-specific adsorption, and (iii) how to detect and quantitatively measure very small signals that result from the interaction of a single biomolecule with multiplex molecules. We tried to solve basically by integrating technologies onto a nanowell array structure. We reported the nanomatrix geometry of a well-oriented nanowell array derived from nanofabrication technology which can easily be employed for digital detection with a high S/N ratio, miniaturization, integrated assays and single molecule analysis. In this present, we describe the diffusion principles of nanowell array electrochemistry toward digitized nanobiomedical device. Nanowell array electrodes were enhanced diffusion rates and thus can be used to measure the kinetics of faster electrode reactions. The enhanced or diminished redox currents are measured at the nano scaled electrode depending on the redox species, i.e. anionic or cationic. It is envisioned that the miniaturized integrated nanowell array-chip system has excellent advantages over conventional instrumental systems for analysis of biomaterials such as compactness, economical, rapid, and multiplex capability.
9:00 PM - TT5.6
Soft to Hard Magnetization Transition in the Heusler Compounds Mn3−xCoxGa.
Vajiheh Alijani Zamani 1 , Juergen Winterlik 1 , Gerhard Fecher 1 , Claudia Felser 1 Show Abstract
1 Institute of Inorganic Chemistry and Analytical Chemistry , Johannes Gutenberg University, Mainz, Rhineland-Palatinate, Germany
In addition to the well-known cubic structures of Heusler compounds, tetragonally distorted Heusler compounds are currently receiving increased interest in the field of spintronics [1, 2], especially for spin torque applications . Only few tetragonal distorted Heusler materials have been studied thoroughly, Mn3Ga and Mn3−xGa are the most prominent examples [4, 5]. These materials are particularly interesting due to the perpendicular magnetic anisotropy which can be achieved in thin films opening the door to spin-torque devices . Therefore, it is essential to design new materials that fulfill the corresponding criteria, i.e. high spin polarization and Curie temperature but a low saturation magnetization and magnetic damping . Our current research is particularly focused on Mn3-xCoxGa Heusler compounds. These compounds are of exceptional importance due to their large diversity of adaptive magnetic properties and their tunability by variation of electron doping. This work is focused on the electronic, magnetic, and structural properties of Mn3−xCoxGa Heusler compounds. The Mn3−xCoxGa series, with X varying from 0.1 to 1.0 in steps of X = 0.1, was synthesized and investigated. It has been shown that the series Mn3−xCoxGa crystallize in the inverse tetragonal structure (I4-m2, space group no 119), for X= 0.1 − 0.4, in the cubic inverse Heusler CuHg2Ti structure type (space group no 216, F43-m), for X = 0.6 − 1 and in both cubic and tetragonal phases for Mn2.5Co0.5Ga. In this series, while the tetragonal alloys are hard magnets and exhibit the features typically attractive for STT applications (high TC, high PMA, high spin polarization, low MS and Gilbert damping), the cubic systems are soft magnets and present the 100% spin polarized materials (or half-metals) obeying the Slater-Pauling rule. Different type of magnetic properties, hard-magnetic properties for Mn-rich alloys and soft-magnetic hysteresis loops for Co-rich alloys, facilitates the tenability of the magnetic anisotropy by varying the Co concentration.  C. Felser, G. H. Fecher, and B. Balke, Angew Chem Int Ed 46 (2007) 668. I. Galanakis, P. H. Dederichs, N. Papanikolaou, Phys. Rev. B. 66 (2002) 174429. J. Winterlik, S. Chadov, V. Alijani, T. Gasi, K. Filsinger, B. Balke, G. H. Fecher, C. A. Jenkins, J. Kübler, G. Liu, L. Gao, S. Parkin, and Claudia Felser. Submitted to Angew Chem. B. Balke, G. H. Fecher, J. Winterlik, and C. Felser, Appl. Phys. Lett. 90 (2007) 152504. J. Winterlik, B. Balke, G. H. Fecher, and C. Felser, Phys. Rev. B 77 (2008) 054406. T. Graf, C. Felser, and S. Parkin, Solid State Chemistry 39 (2011) 1. S. Wurmehl, G. H. Fecher, H. Kandpal, V. Ksenofontov, C. Felser, and H. Lin, Appl. Phys. Lett. 88 (2006) 032503.
9:00 PM - TT5.7
Lead-Free (Na0.5,K0.5)(Nb0.95,Ta0.05)O3-BiFeO3 Piezoelectric Thin Films for MEMS Vibration Energy Harvesting Devices.
Alice Leung 1 , Seung-Hyun Kim 1 , Lindsay Kuhn 1 , Wenyan Jiang 1 , Chang Young Koo 2 , Dong-Joo Kim 3 , Angus Kingon 1 Show Abstract
1 School of Engineering, Brown University, Providence, Rhode Island, United States, 2 R&D Center, INOSTEK Inc., Ansan, Gyeonggi, Korea (the Republic of), 3 Materials Research and Education Center, Auburn University, Auburn, Alabama, United States
Piezoelectric materials are of importance due to their high energy-conversion efficiency, in particular, for converting from mechanical energy to electrical energy, and vice versa. In recent years, there has been a growing interest in microelectromechanical system (MEMS) piezoelectric vibration energy harvesting for their potential to power microelectronic devices, in particular low power digital signal processors for sensor applications. While many research groups have focused on the study of bulk prototypes of piezoelectric energy harvesters, only a few groups have demonstrated MEMS devices capable of generating useful power. Among many piezoelectrics, PZT films have been considered the most promising candidates for the piezoelectric MEMS energy harvesting devices since they can produce high mechanical strain under an applied electric field and a large output power. However, this system unfortunately contains a toxic component, lead (Pb), which has led to worldwide efforts to identify an effective replacement system. The evaporation of Pb component during annealing and the improper disposal of Pb-based materials present serious health hazards and become a major concern for commercial manufacture and application. In this work, we investigated lead-free piezoelectric thin films from the promising sodium potassium niobate tantalate (Na,K)(Nb,Ta)O3 (NKNT) system with a small concentration of BiFeO3 (BF) dopant as a replacement for the conventional PZT system. It is clearly demonstrated that a small concentration of the BF dopant enhances the densification of the NKNT thin films, which is a great advantage since pure NKNT thin films are very difficult to become highly dense microstructures due to severe volatilization of sodium and potassium components. We thoroughly investigated material properties of NKNT-BF lead-free thin films. Then, based upon the optimal parameters from material property analysis, a NKNT-BF piezoelectric cantilevered energy harvesting device with a proof mass was successfully fabricated. Progress on fabrication and testing of a lead-free piezoelectric thin film MEMS harvester is discussed along with material property and device performance improvements.
9:00 PM - TT5.8
Multi-Scaling Simulation Study of the Influence of Contaminant Films Interaction Strength on Performance of Ohmic RF MEMS.
Xiaoyin Ji 1 , Yinxuan Yang 2 , Angus Kingon 2 , Douglas Irving 1 Show Abstract
1 Materials Science and Engineering, North Carolina State University, Raleigh, North Carolina, United States, 2 School of Engineering, Brown University, Providence, Rhode Island, United States
Despite many advantages over traditional solid-state devices, RF MEMS (radio frequency micro-electrical-mechanical systems) switches all suffer from issues of long-term reliability. Environmental contaminants such as hydrocarbons are believed to prevent metal/metal contacts under mechanical actuation of ohmic RF MEMS, which reduces performance of the device over its lifetime. It is difficult to address the issue of the surface contamination layer experimentally, thus we investigate the influence of contaminant films on metal/metal contacts in ohmic RF MEMS via a computational approach. We have developed and applied a novel coupled continuum and atomistic method that simultaneously solves thermal and electric transport equations and couples this information to an underlying molecular dynamics simulation. Environmental contaminants are often a random mixture of hydrocarbons that have varying interaction strengths with the underlying gold contact. We will present results of our simulations that vary the interaction strength of contaminant films with the substrate studying the response of films that are made up of alkanes, alkane-thiols, and a blend of both. The thiol-based alkanes have a strong interaction with the Au substrate while pure alkanes have a much weaker interaction. We study the response of these contaminant films under a variety of mechanical conditions such as external shear force and pressure. We also explore the influence of Joule heating with applied voltage. Additionally, we will present results on our efforts to optimize the performance of this novel method to handle these materials in a high throughput fashion. This work is supported by ONR under Contract No. N00014-10-1-0402.
9:00 PM - TT5.9
The Suitability of Proton Beam Writing as a Tool to Fabricate MEMS Devices in Semiconductors.
Martina Schulte-Borchers 1 , Ulrich Vetter 1 , Hans Hofsaess 1 Show Abstract
1 II. Physikalisches Institut, Georg-August-Universitaet Goettingen, Goettingen Germany
Proton beam writing is a fast direct-write method which allows for the fabrication of structures in the micrometer range having high aspect ratios simply by scanning the proton beam over the semiconductor at positions where structuring is desired. After irradiation of selected areas of a near surface region of a semiconductor with protons in the energy range of a few MeV, the structures reveal after etching under appropriate conditions. In this work we will discuss the suitability of proton beam writing for MEMS fabrication in some potentially interesting semiconductors such as GaAs, InP, GaP, AlN and ZnO. Different etching methods such as electrochemical and ion beam etching are compared with respect to their usefulness when etching proton irradiated semiconductors. It will also be shown that three dimensional structuring may be achieved by simply varying the proton fluence in some cases.
Christoph Eberl Karlsruhe Institute of Technology
Frank W. DelRio National Institute of Standards and Technology
Maarten P. de Boer Carnegie Mellon University
Chris Keimel GE Global Research
TT6: Bio MEMS and Sensors
Wednesday AM, November 30, 2011
9:30 AM - **TT6.1
Micromachined CMOS Probes for High-Density Intracortical Recording: Materials and Technologies.
Oliver Paul 1 , Patrick Ruther 1 , Hercules Neves 2 Show Abstract
1 IMTEK, University of Freiburg, Freiburg Germany, 2 , IMEC, Leuven Belgium
The paper reviews recent progress in highly integrated silicon-based penetrating neural probes opening a new perspective on the spatial control of the recorded neural signals. The slender-shaft-shaped probes are fabricated using commercial CMOS technology followed by dedicated micromachining. By integrating multiplexers on the probe’ shaft, as opposed to previous approaches, where multiplexing structures were constrained to the probe base, an unprecedented number of electrode locations can be screened for optimal neural signal quality. At the same time the interconnection overhead on each shaft and to the external instrumentation remains minimal. As an example, a 4-mm long four-shaft comb structure contains 752 electrode sites, i.e. 188 sites per shaft arranged as two parallel columns, with an electrode pitch of only 40 µm. Eight sites per shaft can be freely selected for signal recording after a careful screening of the signal quality at all locations. This method of selecting the desired recording location has been termed ‘electronic depth control’. Even in case of brain micromotion with ensuing signal loss at given electrodes, signals can be recovered by selecting nearby recording sites.Pt and IrOx electrodes are deposited in a post-CMOS process, connected to aluminum features of the CMOS structure, and protected by a thick silicon nitride passivation. The probes are structured by etching before grinding (EBG), where trenches outlining the geometry of the shafts and their base are first deep reactive ion etched into the CMOS substrates from the front. The wafers are then ground from the rear until the probes are released. Probes with thicknesses from 150 µm down to 25 µm have thus been realized. Such processing defines probes with highly controlled geometry and a minimum of the artefacts produced by alternative techniques such as dicing or double-sided ion etching. The assembly and interconnection of the probes with PCB boards and highly flexible polyimide cables and mounting methods suitable for acute, sub-chronic and chronic recording experiments have been implemented. The electronic depth control has been successfully demonstrated in controlled in-vitro experiments and in in-vivo recordings in anaesthetized rats for several weeks.
10:00 AM - TT6.2
Controlled Release of Selectively Captured Label-Free Cells in Thermoresponsive Microfluidic Channels.
Umut Gurkan 1 , Tarini Anand 1 , David Elkan 1 , Altug Akay 1 , Hasan Keles 1 , Utkan Demirci 1 Show Abstract
1 Center for Biomedical Engineering, Brigham and Women’s Hospital, Harvard Medical School, Harvard-MIT Health Sciences and Technology, Cambridge, Massachusetts, United States
Selective capture of cells from bodily fluids in microchannels has enabled circulating tumor cell isolation, rapid CD4+ cell counting for HIV monitoring, and diagnosis of infectious diseases. Although cell capture methods have been demonstrated in microfluidic systems, release of the captured cells remains as a challenge. Nondestructive retrieval of captured cells in microchannels will enable a new era in biological sciences by allowing post-processing and recultivation of isolated cells with easily accessible rapid microfluidic methods. The significant challenge in release comes from the fact that the cells adhere strongly to the microchannel surface especially with immuno-based immobilization methods. Even though physical (fluid shear) and chemical (enzymes) means have been used to detach cells in microchannels, these methods harm cells and affect cellular characteristics. Here, we describe a new approach to release the selectively captured label-free cells in microchannels without the use of fluid shear or enzymes. We have integrated temperature responsive polymer, biotin-binding protein and biotinylated antibody based surface chemistry with microfluidics to release the captured CD4+ cells from blood with high specificity (89% ± 8%), viability (94% ± 4%), and release efficiency (59% ± 4%). This method utilizes inexpensive, easy to fabricate microchannels, and allows simple manual selective cell isolation in less than 10 minutes, which can also enable resource-limited applications. The presented technology can be used to isolate and purify a wide variety of cells from mixed populations offering broad applications in: tissue engineering, regenerative medicine, clonal/population studies, downstream proteomic/genomic studies, re-cultivation studies, and rare cell isolation.
10:15 AM - TT6.3
Rapid, Label-Free Sample Preparation Microchip for Sorting Blood Components at the POC.
Shuqi Wang 1 , Dusan Sarenac 1 , Francoise Giguel 2 , Daniel Kuritzkes 3 , Utkan Demirci 1 Show Abstract
1 Harvard-MIT Health Sciences and Technology, Harvard-MIT Health Sciences and Technology, Cambridge, Massachusetts, United States, 2 Infectious Diseases Unit, Massachusetts General Hospital, Boston, Massachusetts, United States, 3 Harvard Medical School, Harvard University, Cambridge, Massachusetts, United States
Rapid and label-free separation of viruses and serum on-chip is essential for a variety of point-of-care (POC) applications such microfluidic PCR and ELISA systems. A POC device should include three modules that can perform sample preparation, detection and quantification read out on-chip. However, most current diagnostic technologies require sample pre-processing steps such as centrifugation, which is not suitable for POC applications. Although some microfluidic systems have shown separation of blood components, these systems require microfluidic pumps and control over the flow rates. Separation of viruses in whole blood using a microfluidic pump free system has not been reported. Here, we present a microfluidic device with size-based filtration to isolate plasma and viruses from the rest of the blood components, i.e., white blood cells (WBCs) and red blood cells (RBCs). The results showed that 73.4% of RBCs and 91.8% of WBCs were captured on the filter using a membrane filter at a pore size of 2 µm. Off-chip PCR analysis showed up to 95.8% of virus recovery from whole blood in which cultured HIV virus was spiked. The results demonstrate that the size-based filtration microchip can be used as an effective pump-free system to perform sample preparation for viral separation facilitating POC diagnosis in resource-limited settings.
10:30 AM - TT6.4
A Novel High Yield Microfabrication Process and Porphyrin SAM Functionalization for Polymeric Piezoresistive Microcantilevers for Biochemical Sensing.
V. Seena 1 , Prasenjit Ray 1 , Rohit Pandharipande 1 , Prakash Apte 1 , Ramgopal Rao 1 Show Abstract
1 Centre for Excellence in Nanoelectronics, Department of Electrical Engineering, Indian Institute of Technology Bombay, Mumbai, Maharashtra, India
Microfabrication of polymeric piezoresistive microcantilevers using a novel fabrication process with an improved process yield is presented in this work. SU-8 microcantilevers with SU-8/conducting nanoparticle composites such as SU-8/Carbon Black or SU-8/CNT composites as piezoresistor are potential candidates for many (bio) chemical sensing applications due to their very high sensitivity, cost effectiveness, simple fabrication process and mechanical stability. However, one of the major concerns in connection with the microfabrication of polymer/nanoparticle composite structures is the device to device variability and low process yield. For a well dispersed SU-8/CB nanocomposite, one of the main sources of device variability is the non-uniform distribution of CB in SU-8/CB layer after spin coating the nanocomposite solution on patterned samples. Another source of device variability is the breakage of SU-8/CB patterns during the sonication step which is required as part of the device fabrication. A novel microfabrication process is reported in this work for fabrication of SU-8 nanocomposite microcantilevers that could address both these issues. In this modified process the first layer and SU-8 and SU-8/CB layer are patterned together. The photolithographic steps for the first layer of SU-8 (950 nm) and SU-8/CB nanocomposite piezoresistive layer were done separately using the same mask, whereas the development step of both the layers in SU-8 developer solution was done together with sonication. In this process, the presence of an undeveloped bottom SU-8 layer acts as the lift-off layer for easy removal of the undeveloped SU-8/CB layer and it provided a uniform base layer for spin coating of SU-8/CB composite without any problem of step coverage and hence thickness variations. The yield of SU-8/CB patterning process was improved by about 40 %. The fabricated devices provided better thermal stability in comparison to the previously reported SU-8 nanocomposite microcantilevers. A process for functionalizing SU-8 microcantilevers with Zn-Porphyrin was developed. The application of porphyrin SAM functionalized SU-8 nanocomposite microcantilevers for the detection of trinitro toluene (TNT) explosive vapours is also presented. The sensor exhibited an output voltage change of 40 mV for TNT vapour concentration of 30 ppb and the sensor was found to be reusable for multiple TNT vapour exposure cycles.
10:45 AM - TT6.5
PVDF Microbelt for Harvesting Energy from Respiration.
Chengliang Sun 1 , Jian Shi 1 , Matthew Starr 1 , Xudong Wang 1 Show Abstract
1 , University of Wisconsin - Madison, Madison, Wisconsin, United States
Harvesting energy from ambient mechanical energy sources has recently been shown as a promising strategy for powering small electronics and eventually achieving self-powered or self-sufficient electronic devices. The self-powering capability is particularly important for implantable biological devices which are desired to scavenge energy from the surroundings while maintaining minimum sizes and harmless compositions. By using piezoelectric nanomaterials as the functional elements, very low-level energies from heart beats, muscle stretching, or blood circulation could be converted into electricity via the direct piezoelectric effect. Compared to above biological energy sources, respiration possesses the highest energy density in the form of air flow. Thus, to scavenge energy from respiration is promising and advantageous for providing relatively high power to biomedical devices. Here, we report the design of a micrometer-thick piezoelectric β-phase polyvinylidene fluoride (PVDF) belt that can effectively convert the energy from low-speed air flow to electricity via its resonant oscillation. We employed an etching-based technique for thinning commercial β-phase PVDF thin films with persevered phase purity. The film thickness-related energy conversion capability was characterized. We demonstrated that such a thin film could provide sufficient power for driving a small electronic device. Thus, utilizing a phase-preserving PVDF film thickness-reduction technique we have produced an energy harvester which in principle is capable of scavenge energy from the biological process of respiration.
11:30 AM - TT6.6
Novel Integration of Nanofluidic and Nanophotonics for Single Molecule Detection.
Mehrsa Raeiszadeh 1 , Ehsan Shah Hosseini 2 , Ali Adibi 2 , Paul Kohl 1 Show Abstract
1 Chemical and Biomolecular Engineering, Georgia Institute of Technolog, Atlanta, Georgia, United States, 2 Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, Georgia, United States
Incorporating advanced micro/nanofluidics with high sensitivity photonic sensors will provide compact, effective sensors for lab-on-a-chip systems. Integration of micro/nanofluidics with nanophotonics leads to inexpensive, small sample volume, and reduced analysis time point of care sensors. Thus, optofluidic sensors are gaining widespread use in biosensing and chemical analysis applications. Some potential applications of optofluidic sensors include clinical screening, medical diagnostics, screening of chemical compounds in drug discovery and development, and toxic detection. The microfluidic integration of optical chips with the usually aqueous solutions can be done by decomposable polymers. Epoxy-functionalized polymers, such as polynorbornene (PNB), are valuable for forming micrometer-size structures due to their case of reaction, and can be used as sacrificial polymers. In this work, a negative-tone, PNB-based sacrificial material has been used to make microchannels. The developed sacrificial polymers patterns can be encapsulated with a thick layer of SiO2, which will not affect the optical performance of the resonators. The thermal decomposition products of the polymer are able to diffuse through the encapsulating SiO2 to leave clean channels of an exact shape.The device consists of a Si micro-ring resonator covered by a microfluidic channel. A range of 500 nm to 40 µm wide lines is successfully fabricated. The SiN waveguides and resonators are fabricated with electron-beam lithography and plasma etching. These ultra-high quality factor SiN resonators are demonstrated in the visible range for the first time. The device is designed in a way that the evanescent light traveling in the ring resonator interacts with the upper fluidic cladding. At resonance, light circulates many times within the ring, which leads to a large enhancement in the interaction length between the evanescent field and the cladding liquid (with micro/nanochannels, the interaction with the cladding fluid will be maximized). High quality optical resonators can demonstrate very sharp resonances, which can be the core of very sensitive and accurate integrated sensing devices based on refractive index, fluorescence and Raman. In this investigation, different fluids are successfully passed through the channels, and the fluidic structures are interfaced with photonics for index and florescence sensing.These channels made with a low-temperature decomposable polymer are promising for future applications. Considering possible size reduction of the photonic devices (especially through photonic crystal cavities), it is possible to shrink the size of the channels even further and achieve ultra-small sample sizes and multi-mode sensing functionalities through florescence and Raman signals. The Raman signal can be excited and collected through metallic nanoparticles fabricated on top of the photonic devices and integrated with nanofluidic channels for single molecule sensing.
11:45 AM - TT6.7
Electrowetting on Dielectric for Poro-Vascular Composite Skins with Active Surface Morphology.
Marriner Merrill 1 , Kristin Metkus 2 , James Thomas 1 Show Abstract
1 Multifunctional Materials, Code 6350, Naval Research Laboratory, Washington, District of Columbia, United States, 2 , NOVA Research, Inc., Alexandria, Virginia, United States
A new class of multifunctional materials is being developed at the US Naval Research Laboratory (NRL) combining a structural skin with controllable surface morphology. These new materials are called ‘poro-vascular’ (PV) composites and consist of a two-phase (solid / liquid) network of channels and pores with sub-millimeter length scales. Active surface morphology is proposed by controlling the shape of the liquid phase at the pore exits from domed to dimpled through a combination of electrowetting on dielectric (EWOD) and pumping. However, overall design requirements create a challenging environment for EWOD. The substrate must be flexible, conductive, lightweight, robust, capable of being precisely machined by laser micro-machining, and still have good specific stiffness and strength. Additionally, the fluid phase must have a very low vapor pressure, so traditional aqueous solutions cannot be used. This paper presents the experimental development of a substrate / ionic liquid combination that provides EWOD control of drop shape within these unique constraints, focusing first on flat, non-machined substrates. Two different electrode materials (gold coated and conductive polyimides), three different dielectric materials (Parylene HT, Parylene C, and PTFE, all vapor deposited), and an additional spin-coated layer of Teflon AF were examined. The evolution of the nano- and micro-scale surface roughness was characterized at the different processing stages. Then, complete electrode-dielectric-hydrophobic layers were tested with EWOD using an ionic liquid and the results related to the material and processing choices. The optimal PV composite substrate found in this process will be presented together with design rules for future development.
12:00 PM - TT6.8
Simulation of Surface Effects in the Dissipative Dynamics of Silicon Nanoresonators.
Behrouz Shiari 1 Show Abstract
1 EECS, University of Michigan, Ann Arbor, Michigan, United States
Recent advances in nanotechnology have led to the development of nano-scale mechanical resonators, which have drawn great attentions due to their new applications such as ultrafast sensors and high-frequency signal processing with high stability, high resolution output. The quality factor (Q-factor) is frequently used to measure the resonators’ performance. This work investigates the quality factors of resonance associated with the axial and transverse vibrations of resonators through the use of molecular dynamics (MD) simulation.We have adapted the parallel molecular dynamics (MD) code LAMMPS to simulate and analyze the nanoresonator oscillation, making use of the Stillinger-Weber (SW) potential to describe the interatomic interaction. The nano-beams are initially thermalized and mechanically equilibrated at 10 K, 100 K, 200 K, 300 K and 800 K. We investigate the energy dissipation for clamped-clamped silicon nano-beam resonators. The intrinsic losses arising from anharmonic interatomic potential and surface roughness are considered. The resonators are simulated in vacuum. The extrinsic loss mechanism is taken account through the attachment cross section of resonators to their support substrate.The dependence of the quality factor (Q) on temperature and the size of the resonator is calculated from direct simulation of the oscillation of a series of Si <100> nano-beam with patterned rough surfaces.The quality factors of the fundamental vibrational modes have been observed to decrease approximately linearly with decreasing the surface area to volume ratio of the resonator. This shows that surface losses play a significant role in determining the quality factor of nanoresonators. The quality factor also decreases with increasing temperature from 10 K to 800 K.
12:15 PM - TT6.9
Low Power Consumption MEMS Microhotplates for the Integration of Sensing Nanowires.
Roman Jimenez-Diaz 1 , Jordi Sama 1 , J. Daniel Prades 1 , Albert Romano-Rodriguez 1 , Francisco Hernandez-Ramirez 2 , Joaquin Santander 3 , Carlos Calaza 3 , Luis Fonseca 3 , Carles Cane 3 Show Abstract
1 , University of Barcelona, Barcelona Spain, 2 , IREC, Catalonia Institute for Energy Research, Barcelona Spain, 3 , Instituto de Microelectronica de Barcelona, IMB-CNM-CSIC, Bellaterra Spain
Nanowires have emerged as potential building blocks for future electronic devices. In this work, a methodology for the fabrication of gas sensors by integrating individual metal oxide nanowires (NWs) as sensing elements in microhotplates for low power consumption and fast operation is presented. The integration of nanowires in sensing devices requires the possibility of modulating their temperature with an easy-to-control and low power consumption system. The fabrication of MEMS microhotplates that include microheaters with reduced dimensions allows setting the working temperature up to 600 K and guarantees thermal dynamics response much faster than the bulky counterparts, as well as having extremely low power consumption requirements. In these sensing platforms, the size-reduction of the microhotplates, which are suspended by a few arms contribute to an important reduction in the power consumption. Moreover, the fabrication of small hotplates (100 µm diameter) satisfies the requirement for low heat dissipation which lead to ultra-lower power consumption. In our devices, a temperature as high as 600K can be achieved with voltages below 3V and power around 15 mW. These values are within the range of power consumption of most actual semiconductor electronic components.Some nanowires can be dispersed in propylene glycol and, afterwards, a droplet of this solution is spread onto suspended the microhotplates. To guarantee the formation of good electrical contacts between pre-patterned microelectrodes and nanowires, Electron Beam Assisted Deposition and Ion Beam Assisted Deposition processes were performed. These nanowires were electrically contacted using a FEI Strata 235 dual beam instrument equipped with an injector to deposit Pt.Finally, dc electrical measurements were done using a Keithley 2602 Source Measure Unit, enabling the estimation of the key-parameters of these nanowires. The devices were tested using well-controlled environmental conditions of gas and temperature. Metal oxide materials exhibit a temperature dependent response to different gas species when used as chemiresistors. The integration of microheaters in the measuring platform enabled an optimal control of the working temperature, thus enhancing the mechanism of adsorption and desorption of gas species in metal oxides materials and expediting the use of these nanowires as gas sensors. The obtained results demonstrated that working with microhotplates simplifies the experimental setup and also decreases the power consumption of the devices. Moreover, the huge potential of nanowires as building-blocks of a new generation of devices with improved performances has been exposed. For this reason microhotplate-based technologies are a promising approach for the fabrication of nanosensors in a scalable process.
12:30 PM - TT6.10
Size Effect of Flexible Proof Mass on the Mechanical Behavior of MEMS-Scale Cantilevers for Energy Harvesting Applications.
Miso Kim 1 2 , Seungbum Hong 2 , Dean Miller 2 , John Dugundji 1 , Brian Wardle 1 Show Abstract
1 , MIT, Cambridge, Massachusetts, United States, 2 Materials Science Divsion, Argonne National Laboratory, Lemont, Illinois, United States
Mechanical behavior of micron-scale cantilevers with a distributed, flexible proof mass is investigated to understand proof mass size effects on the performance of microelectromechanical system (MEMS) energy harvesters. Single-crystal silicon beams with proof masses of various lengths were fabricated using focused ion beam milling and tested using atomic force microscopy. Comparison of three different modeling results with measured data reveals that a ‘two-beam’ method has the most accurate predictive capability in terms of both resonant frequency and strain. Accurate strain prediction is essential because energy harvested scales with strain squared and maximum strain will be a design limit in fatigue.