Nano-Manufacturing Technology

IMECE2008-66825 pp. 1-7; (7 pages)

Piezoelectric nanotubes are applicable to many areas including sensors, actuators, and energy storage. Before these applications can be realized using nanoscale components, a study is needed to characterize the piezoelectric-properties of nanotube geometries versus the piezoelectric-properties of a thin-film of the same material (not shown here). The material used in this study is (PZT) lead-zirconate-titanate, PbZr0.5 Ti0.48 O3 . In this work, 200nm diameter PZT structures are manufactured by a template wetting procedure. These ceramic nanotubes are brittle but hold up well inside the alumina template structure. Having an array of nanometer-sized piezoelectric-tubes has been hypothesized to give comparable results to that of thin-films of the same material. The research described in this paper is a preliminary step towards testing this hypothesis via characterization and later producing piezoelectric-nanoscale-sensors, which are both small and efficient.

Topics: Wetting , Nanotubes
Commentary by Dr. Valentin Fuster
IMECE2008-67155 pp. 9-17; (9 pages)

The ever increasing development of portable electronics leads to a higher demand for compact and reliable power sources. Significant resources are being presently dedicated to the study of micro machined turbines, because of their remarkable power density that suggests that the generation of about 100–300 W with a total device weight of few hundreds grams and a fuel mass flow rate of few grams per second may be feasible in the short range. In this paper a possible configuration of such a nano-GT set is considered, which was defined on the basis of previous thermo-fluid dynamic analysis: the results of a preliminary design study, including some cold-run tests, are reported in this paper. The layout of the device was finalized on the basis of both a CFD and a FEM analysis that identified the “optimal” blade shape, shaft size and rotors arrangements under the point of view of the energy efficiency and of thermo-mechanical material stresses, Some of the problems deriving by the physical construction and preliminary testing of the prototype are analyzed and discussed.

Commentary by Dr. Valentin Fuster
IMECE2008-67372 pp. 19-23; (5 pages)

Hot embossing is an effective technology for reproducing micro-scale features in polymeric materials, especially micron scale patterning. The equipment used for hot embossing to date is often research oriented, (intended to be flexible and provide a wide range of processing conditions), and a dedicated equipment industry has yet to develop. This paper details a hot embossing machine design strategy suitable for large-scale manufacturing. The design is motivated by capital cost reduction, right-size machine design, system simplicity, and production flexibility and scalability. Toward this end, a minimal number of components were used, commercially available off-the-shelf components were chosen where possible, system layout was designed to be modular, and system size was scaled for the intended products (in this case microfluidic devices). Innovative design aspects include the use of new ceramic substrate heaters for electrical heating, choice of a moveable heat sink to minimize heat load during the heating cycle, and the careful design of the thermal elements to minimize cycle time. The capital cost and the cost per part produced with this machine are estimated to be an order of magnitude less than currently available options. This design has a minimum cycle time of two minutes, and replicates microstructures within a 25mm by 75mm area.

Commentary by Dr. Valentin Fuster
IMECE2008-68299 pp. 25-32; (8 pages)

Electrospinning is a method of producing nanometer scale fibers by accelerating a jet of charged polymer solution in an electric field. In many emerging, high value electrospinning applications, such as the biomedical area, the diameter distribution of electrospun polymeric nanofibers has important implications for the product’s performance and process economics (in terms of yield and production rate). However, the current state-of-the-art electrospinning process results in unpredictable and time varying diameter distributions, both during a run and run-to-run. Thus, this work is focused on developing an appropriate control system to achieve consistent and controllable fiber diameters. Another goal of this work is to develop a better understanding of the relation between process physics and the resulting fiber diameter characteristics. To address these problems, a well instrumented and computer based actuator control system has been developed. The effects of process parameters on fiber diameter are investigated for achieving consistent and repeatable process capability. The fundamental process dynamics are identified and the relation between measurable variables and the resulting fiber diameter distribution is analyzed. This relation provides the basis of developing appropriate control strategies in order to reduce both the process variations from run-to-run and during a run.

Commentary by Dr. Valentin Fuster
IMECE2008-68714 pp. 33-40; (8 pages)

The unique heat-releasing characteristics of explosively reactive nanolayers (RN) are used in this study to produce Si/solder/Si joints. The microstructure of the RN in the reacted state as well as the post-joining foil/solder interface is characterized via XRD, SEM, and TEM, which have never been done. Additionally, RN solder joints are mechanically characterized by single lap shear and nanoindentation to obtain a set of optimized processing parameters, specifically initial pressure applied (Pa ) and initial temperature of the system (Ti ). A maximum joint shear strength of ∼ 30MPa at Pa = 15MPa and Ti = 75°C. Furthermore, nanoindentation is used to clarify the mechanical behavior of individual layers and interfaces across the joints as a result of thermal aging.

Topics: Solder joints
Commentary by Dr. Valentin Fuster
IMECE2008-69111 pp. 41-45; (5 pages)

Flexible magnetic nanocomposite fibers were produced by electrospinning method using a polymeric solution containing poly(acrylonitrile) and magnetite nanoparticles. Magnetic nanoparticles (∼10 nm) were prepared by a chemical co-precipitation of ferric and ferrous chloride salts in the presence of an ammonium hydroxide solution. The effect of magnetic particle concentrations (e.g., 0%, 1%, 5%, 10%, 20% and 30%) on nanocomposite fibers, distribution and morphology were studied using scanning electron microscopy (SEM). The experimental study indicated that the average diameters of the magnetic nanocomposite fibers were between 400 nm and 1.08 μm. The magnetic responds were also found to increase linearly with increasing percent loading of the magnetic nanoparticles. It is concluded that this study provides promising results for various applications, such as filtration and separation of micron and nanosize organic and inorganic particles, HF antenna fabrication and biomedical.

Commentary by Dr. Valentin Fuster

Micro and Nano Systems

IMECE2008-66257 pp. 49-60; (12 pages)

Different types of fillers with high electrical and thermal conductivities, e.g. graphite and alumina, have been added to adhesive polymers to create composite materials with improved mechanical and electrical properties. Previous modeling efforts have found that it is relatively difficult to predict the effective thermal conductivity of a composite polymeric material when incorporated with large volume content of fillers. We have performed comprehensive computational analysis that models the thermal contacts between fillers. This unique setup can capture the critical heat conduction path to obtain the effective thermal conductivity of the composite materials. Results of these predictions and its comparison with experimental data will be presented in this paper.

Commentary by Dr. Valentin Fuster
IMECE2008-66833 pp. 61-65; (5 pages)

We studied the rotational Brownian motion of magnetic triaxial ellipsoidal particles (orthotropic particles) suspended in a Newtonian fluid, in the dilute suspension limit, under applied shear and magnetic fields. The algorithm describing the change in the particle magnetization has been derived from the stochastic angular momentum equation using the fluctuation-dissipation theorem and a quaternion formulation of orientation space. Results are presented for the response of dilute suspensions of ellipsoidal particles to constant magnetic and shear flow fields.

Commentary by Dr. Valentin Fuster
IMECE2008-68308 pp. 67-71; (5 pages)

This paper addresses the fast-transient heat conduction phenomena of a gold nanoparticle embedded in cancerous tissue in hyperthermia treatment. Dual phase lag model in spherical coordinates was employed and a semi-analytical solution of 1-D non-homogenous dual phase lag equation was presented. Results show that transient temperature depends dramatically on the lagging characteristic time of the surrounding tissue. Temperature predicted by dual phase lag model greatly exceeds that predicted by a classical diffusion model, with either a constant source or a pulsed source. This phenomenon is mainly attributed by the phase lag of heat flux of tissue. The overheating in short time scale and the consequent biological effect needs to be paid more attention in the related study.

Commentary by Dr. Valentin Fuster
IMECE2008-68980 pp. 73-78; (6 pages)

This paper will examine the holding capability of a single dielectric gate. Bennett et al. have previously developed a single dielectrophoretic (DEP) gate for the front-end device of a bio-detection system [12]. By applying a high frequency AC electric field on the top and bottom electrodes, particles which are negatively polarized are repelled from the high field gradient region near the electrodes, and the particles are held in place due to the balancing of the DEP force and the Stokes drag force. However, the particles will flush through the channel if the fluid flow force becomes larger than that the DEP force. Thus, there is a holding capability for the single dielectrophoretic gate which corresponds to the maximum flow rate at which particles can balanced by the DEP force. In this paper, we present a 3D numerical simulation to investigate the parameters that influence the DEP force such as the width of the electrodes, the applied electric field, the ratio of the width of top electrode to the bottom of electrode as well as the horizontal shift between the top electrode and bottom electrode. A select group of parameters is studied in order to achieve the optimal holding capability of the single dielectrophoretic gate. The influence of the phase variation of the AC electric field between the top electrode and bottom electrode is also analyzed.

Commentary by Dr. Valentin Fuster
IMECE2008-66627 pp. 79-85; (7 pages)

This paper presents a comprehensive investigation of energy loss mechanisms in tuning-fork gyroscopes from both numerical and experimental perspectives. Two main energy loss mechanisms, anchor loss and thermoelastic damping (TED), are investigated through numerical models and later experimental investigation. In order to predict the qualify factor of tuning-fork gyroscopes, a separation-and-transfer method is employed to predict anchor loss and a thermal energy method is utilized to calculate thermoelastic damping. The experimental investigation is conducted to verify the developed models. It is found that HF release in fabricating these devices is a critical factor to the measured quality factor of tuning-fork gyroscopes. The calculated Quality factors are compared with the measured data, showing good agreement.

Commentary by Dr. Valentin Fuster
IMECE2008-66703 pp. 87-96; (10 pages)

The objective of this work is to study the effect of three important parameters such as photostrictive actuator thickness, incident light intensity and convective heat transfer coefficient on a silicon cantilever beam with thin photostrictive optical actuator film surface. The authors have developed a computational method useful for design of systems incorporating thin film photostrictive actuators. The element has been implemented in an in-house finite element code named BAMAFEM. A finite element for static analysis of photostrictive thin films has already been developed and verified with analytical analysis approach of another author. To the best of our knowledge, finite element parametric analysis of the photostrictive thin film has not been extensively studied, if studied at all. Photostrictive materials, such as PLZT, demonstrate significant photostrictive behavior under illumination by high-energy light, which can be considered a superposition of a bulk photovoltaic effect and a converse piezoelectric effect. Photostrictive actuators can directly convert photonic energy to mechanical motion. Photostrictive materials can produce strain as a result of irradiation from high-intensity light. Neither electric lead wires nor electric circuits are required. Thus, photostrictive actuators are relatively immune from electrical interference. They have potential use in numerous MEMS devices. At least from the computational point of view, it would be interesting to investigate the effect of different parameters in the actuation of beam using the thin film photostrictive actuators to develop a finite element model useful for design of numerous MEMS and NANO systems having photostrictive actuators.

Commentary by Dr. Valentin Fuster
IMECE2008-66813 pp. 97-102; (6 pages)

In our specific application of the van der Pauw (VDP) structure as a pressure sensor, where the stress field varies significantly, smaller VDP size is beneficial. The resistivity of the VDP sensor is affected by the stress state of the entire area over which it lies, providing a relation between resistance and the average stress in an area, rather than the stress at a point. The further the area over which the VDP lies is reduced, the closer we approximate the stress at a point rather than the average stress over a large surface. Due to microfabrication and other limitations, small, point-like sensors may not be feasible. In our study, a clear relationship between size and sensitivity could not be made. However, it was apparent that size had little overall effect on sensitivity. In testing the VDP devices for comparison with conventional sensor types, it was found that the VDP devices had a linear response as expected, were the most sensitive, and provided a number of additional advantages. Specifically, the VDP device allows for significant miniaturization, because its resistance value is independent of size, and the measurement technique is independent of line resistance. The simple geometry of the VDP also simplifies fabrication.

Commentary by Dr. Valentin Fuster
IMECE2008-67026 pp. 103-111; (9 pages)

The Hertz contact theory allows the onset of yielding to be predicted for those contacts in which the effect of adhesion can be neglected. However in microscale contacts, such as those which occur in MEMS, yielding will occur for lower loads than predicted by the Hertz theory. For such cases, the JKR, DMT, and Greenwood-Johnson theories extend the Hertz theory to include the effect of adhesion. The present study provides yield conditions for the JKR, DMT, and Greenwood-Johnson theories of adhesion. Attention is first focused on the initiation of yield along the axis of symmetry of the contact. The results show that the critical loads for the three adhesion theories are close together, but differ significantly from that predicted by Hertz. In fact it is possible for yielding to occur due to adhesion alone, without an external applied load. A curve-fit formula is given to express the yield load as a function of an adhesion parameter for different Poisson’s ratios. Results are also obtained for the onset of plastic deformation away from the axis of symmetry using the Greenwood-Johnson theory of adhesion.

Commentary by Dr. Valentin Fuster
IMECE2008-67135 pp. 113-120; (8 pages)

In this paper, a brief summary of some of our recent work [1, 2] is presented, with the goal of developing an engineering science-based process-simulation capability for micro-hot-embossing of amorphous polymers. To achieve this goal: (i) a three-dimensional thermo-mechanically-coupled large deformation constitutive theory has been developed to model the temperature and rate-dependent elastic-viscoplastic response of amorphous polymers; (ii) the material parameters in the theory were calibrated by using new experimental data from a suite of simple compression tests on Zeonex-690R (cyclo-olefin polymer), that covers a wide range of temperatures and strain rates; (iii) the constitutive model was implemented in the finite element program ABAQUS/Explicit; and (iv) the predictive capability of the numerical simulation procedures were validated by comparing results from the simulation of a representative micro-hot-embossing process against corresponding results from a physical experiment.

Commentary by Dr. Valentin Fuster
IMECE2008-67340 pp. 121-127; (7 pages)

An extremely promising microscale processing method for bulk metallic glasses called thermoplastic forming has emerged in recent years. However, most of the recent experimental thermoplastic forming studies have been conducted by trial-and-error. In this paper, the large-deformation constitutive theory of Henann and Anand [1] is used as a numerical simulation tool for the design of micro-hot-embossing processes. This numerical simulation capability is used to determine appropriate processing parameters in order to carry out several successful micron-scale hot-embossing operation on the metallic glass Zr41.2 Ti13.8 Cu12.5 Ni10 Be22.5 (Vitreloy-1). By carrying out the corresponding physical experiments, it is demonstrated that microscale features in Vitreloy-1 may be accurately replicated under the processing conditions determined by use of the numerical simulation capability.

Commentary by Dr. Valentin Fuster
IMECE2008-68214 pp. 129-135; (7 pages)

In this study, the homotopy analysis method (HAM) is used to study dynamic pull-in instability in microbeams considering different sources of nonlinearity. Electrostatic actuation, fringing field effect and midplane stretching causes strong nonlinearity in microbeams. In order to investigate dynamic pull-in behavior, using Galerkin’s decomposition method, the nonlinear partial differential equation of motion is reduced to a single nonlinear ordinary differential equation. The obtained equation is solved analytically in time domain using HAM. The problem is studied by two separate manners: direct use of HAM and indirect use of HAM in conjunction with He’s Modified Lindstedt-Poincaré Method. To demonstrate the effectiveness of the solution methods, results are compared with those in literature. The comparison between obtained results and those available in literature shows good agreement.

Topics: Microbeams
Commentary by Dr. Valentin Fuster
IMECE2008-68531 pp. 137-143; (7 pages)

Detailed analysis of residual stress profile due to laser micro-joining of two dissimilar biocompatible materials, polyimide (PI) and titanium (Ti), is vital for the long-term application of bio-implants. In this work, a comprehensive three dimensional (3D) transient model for sequentially coupled thermo-mechanical analysis of transmission laser micro-joining of two dissimilar materials has been developed by using the finite element (FE) code ABAQUS, along with a moving Gaussian laser heat source. The laser beam (wavelength of 1100 nm and diameter of 0.2 mm), moving at an optimized velocity, passes through the transparent PI, gets absorbed by the absorbing Ti, and eventually melts the PI to form the bond. The laser bonded joint area is 6.5 mm long by 0.3 mm wide. First the transient heat transfer analysis is performed and the nodal temperature profile has been achieved, and then used as an input for the residual stress analysis. Non-uniform mixed meshes have been used and optimized to formulate the 3D FE model and ensure very refined meshing around the bond area. Heat resistance between the two materials has been modeled by using the thermal surface interaction technique, and melting and solidification issues have been approximated in the residual stress analysis by using the appropriate material properties at corresponding temperature. First the model has been used to observe a good bonding condition with the laser parameters like laser traveling speed, power, and beam diameter (burnout temperature of PI > maximum temperature of PI achieved during heating > melting temperature of PI) and a good combination has been found to be 100 mm/min, 3.14 W and 0.2 mm respectively. Using this combination of parameters in heating, the residual stress profile of the laser-micro-joint has been calculated using FE model after cooling the system down to room temperature of 27 °C and analyzed in detail by plotting the stress profiles on the Ti and PI surfaces. Typically the residual stress profiles on the PI surface show low value in the middle, increase to higher values at about 160 μm from the centerline of the laser travel symmetrically at both sides, and to the contrary, on Ti surface show higher values near the centerline of traveling laser beam. The residual stresses have slowly dropped away on both the surfaces as the distance from the bond region increased further. Maximum residual stresses on both the Ti and PI surfaces are at the end of the laser travel, and are in the orders of the yield stresses of respective materials.

Commentary by Dr. Valentin Fuster
IMECE2008-68537 pp. 145-152; (8 pages)

The application of higher order continuum theories, with size effect considerations, have recently been spread in the micro and nano-scale studies. One famous version of these theories is the couple stress theory. This paper utilizes this theory to study the anti-plane problem of an elliptic nano-fiber, embedded in an infinite medium, both made of centrosymmetric isotropic material. In this framework, a characteristic length appears in the formulation, by which examination of the size effect is possible. This work presents an analytical solution for the proposed problem.

Commentary by Dr. Valentin Fuster
IMECE2008-68762 pp. 153-161; (9 pages)

A novel two-parameter area function for determination of near surface properties of Young’s modulus of elasticity and hardness has shown promise for compensating for the imperfection of the tip-end in an instrumented indentation measurement. This paper provides a comprehensive study involving a Berkovitch tip. The tip is utilized in an MTS nanoindentation measurement machine and used to establish load indentation information for fused silica samples. The geometry of the tip is then characterized independently using a highly accurate Atomic Force Microscope. Using the indentation data along with the two-parameter area function methodology, the tip-end radius of curvature is found to provide the most consistent value of modulus of elasticity. Independently, the data from the SEM measurement of the same tip is used to obtain the least squares estimation of the tip curvature. The two approaches yield favorable agreement in the estimation of tip-end radius of curvature. Therefore, the validity of the two-parameter area function method is proved. The method is shown to provide a robust, reliable and accurate measurement of modulus of elasticity and hardness in the nanoscale proximity of a surface.

Commentary by Dr. Valentin Fuster
IMECE2008-68884 pp. 163-170; (8 pages)

A simple method to determine the frictional interaction between a carbon nanotube (CNT) and a substrate is analyzed for feasibility. In this technique an atomic force microscope (AFM) tip is used to drag a CNT along a substrate. Then the deformed shape of the CNT can be viewed either with the AFM or in an SEM. An analysis of the deformed shape allows the determination of the frictional interactions which occurred during dragging. It is important to quantify these interactions in a variety of potential applications of nanotechnology. In one such example, a CNT based nanoswitch consists of a CNT bridging over a trench. Actuation of the CNT causes it to stretch and can lead to partial slip at the interface. This slip causes hysteresis which has been observed in the mechanical actuation of a CNT bridge. In this paper continuum level modeling of the frictional interaction is used to determine the relationship between the shape of the CNT and the frictional interaction which occurred between the CNT and substrate during dragging. The model and analysis indicate that this method should be feasible for CNTs with aspect ratios approximately in the 100–250 range.

Commentary by Dr. Valentin Fuster
IMECE2008-69171 pp. 171-179; (9 pages)

Nanotechnologies are considered to be the driver of the Information-Age engineering. Recent discoveries in practically all aspects of engineering developments indicate that properties at nano-levels are starkly different from properties at bulk levels. These discoveries signal great potentials for nanotechnologies that can revolutionize all technologies, ranging from medicine to energy. However, the same discoveries also point to the fact that conventional laws and theories that have enjoyed long-standing confidence of the scientific community do not apply to nanotechnologies. In absence of such laws that describe nano-scale phenomena, it is difficult if not impossible to predict long-term impacts of nanotechnologies. This paper presents a comprehensive formulation of mass and energy balance equations. This formulation gives rise to a unique set of equations that apply to both nano- and bulk scale natural phenomena. The formulation is based on momentum balance, which is preserved at all scales, ranging from cosmic to nano- and even the inter-atomic level. Although Newton posited gravitation as a universally acting force, we now know that electromagnetic forces predominate in matter at the nano or inter-atomic level. Electromagnetic forces, like frictional forces, however, can exist and persist without ever having been externally applied. Reasoning thus “by exhaustion”, Newton’s Three Laws of Motion plus the principle of universal gravitation are actually special cases of “something else”. That “something else” is far more general, viz., the universal preservation of mass-energy balance and conservation of momentum. The connecting element of this universal balance is that motion is the mode of existence of all matter. This renders time a characteristic of matter itself within the overall context of mass-energy-momentum conservation. In other words, time ceases to be mainly or only a derivative of some spatial displacement of matter. In this way, it becomes possible at last to treat time, consistently, as a true fourth dimension — and no longer as merely the independent variable. This description is consistent with Einstein’s revolutionary relativity theory, but does not rely on Maxwell’s equations as the starting point. The resulting equation is shown to be continuous in time, thereby allowing transition from mass to energy. As a result a single governing equation emerges. This equation is solved for a number of cases and is shown to be successful in discerning between various natural and artificial sources of mass and energy. With this equation, the difference between chemical and organic fertilizers, microwave and wood stove heating, and sunlight and fluorescent light can be made with unprecedented clarity. By applying this equation, a complete pathway analysis of nanomaterials is made and it is shown that engineering at nano-scale will have long-term impacts. This analysis would not be possible with conventional techniques. Finally, analysis results are shown for a number of energy- and material-related prospects.

Commentary by Dr. Valentin Fuster
IMECE2008-66561 pp. 181-185; (5 pages)

Sensitivities of microelctromechanical system (MEMS) accelerometers are typically measured and calibrated at final production test of packaged devices. This process typically requires expensive special automated test equipment (ATE) that can generate vibration stimulus. A single-axis vibration test system has been developed on an existing commercial wafer probe/trim system to calibrate sensitivities of MEMS accelerometers during wafer probe/trim process to minimize the need for such vibration test equipment. To increase signal to noise ratio but avoid damage of probe pad/pin during the test, the vibration exciter must be able to generate high frequency but small displacement vibration stimulus. The vibration exciter also needs to be small enough to fit into the existing commercial probe/trim system and requires minimum changes to the system. A high resolution but small size in-situ noncontact vibration measurement technique is needed to ensure calibration accuracy. This paper presents a unique solution to meet all these challenges. The success of this system has been validated by final product test data of a test device, a 3-axis low-g MEMS accelerometer.

Commentary by Dr. Valentin Fuster
IMECE2008-67502 pp. 187-195; (9 pages)

We present modeling and analysis for the static behavior and collapse instabilities of doubly-clamped and cantilever microbeams subjected to capillary forces. These forces can be as a result of a volume of liquid trapped underneath the microbeam during the rinsing and drying process in fabrication. The model considers the microbeam as a continuous medium, the capillary force as a nonlinear function of displacement, and accounts for the mid-plane stretching nonlinearity. The capillary force is assumed to be distributed over a specific length underneath the microbeam. The Galerkin procedure is used to derive a reduced-order model consisting of a set of nonlinear algebraic and differential equations that describe the microbeams static and dynamic behaviors. We study the collapse instability, which brings the microbeam from its unstuck configuration to touch the substrate and gets stuck in the so-called pinned configuration. We calculate the pull-in length that distinguishes the free from the pinned configurations as a function of the beam thickness and gap width for both microbeams. Comparisons are made with analytical results reported in the literature based on the Ritz method for linear and nonlinear beam models. The instability problem, which brings the microbeam from a pinned to adhered configuration is also investigated. For this case, we use a shooting technique to solve the boundary-value problem governing the deflection of the microbeams. The critical microbeam length for this second instability is also calculated.

Commentary by Dr. Valentin Fuster
IMECE2008-68066 pp. 197-207; (11 pages)

Steadily improving performance of inertial sensors necessitates significant enhancement of the methods and equipment used for their evaluation. As the nonlinearity of sensors decreases and gets close to that of the exciters, new challenges arise. One of them, addressed in this research, is a superposition of errors caused by the nonlinearity of tested devices with nonlinear distortions of excitation employed for experimental evaluation. This can lead to a cancellation, at least partial, of the effects of both imperfections and underestimation of the actual distortions of the evaluated sensors. We implement and analyze several system architectures and components of applicable motion generation systems from the viewpoint of satisfying the relevant, often conflicting requirements posed by the evaluation of high performance sensors. Robust mechanical integration of the guidance, actuation and measurement functions emerges as a key factor for achieving the needed quality of generated test patterns. We find precision air bearing stages such as ABL1500 series (Aerotech) most suitable for implementing the needed experimental setup. We propose an architecture with two reciprocating stages, implement and evaluate its core components, and illustrate its performance with experimental results.

Topics: Sensors , Accuracy
Commentary by Dr. Valentin Fuster
IMECE2008-68893 pp. 209-213; (5 pages)

Using external stimuli to control surface properties, such as surface topology or wettability, has been of great interest. This paper presents the fabrication, mechanism, and analysis of an active method that modulates surface nano-topology and wettability of polymeric films. The polymeric film possesses the unique property of electrical potential-induced wettability conversion: when its doping level is electrically altered, the interstitial ions are incorporated or released, causing the polymer network to stretch and thereby changing its surface morphology.

Commentary by Dr. Valentin Fuster
IMECE2008-67209 pp. 217-224; (8 pages)

The spinodal decomposition of a polymer-polymer-solvent ternary system was investigated in a 3-dimensional numerical model. The Cahn-Hilliard equation was employed to describe the free energy profile of the domain. A heterogeneously functionalized substrate was also implemented into the model to simulate the effect of substrate attraction. The mechanism of the morphology evolution was studied quantitatively. The well established linear relationship of the characteristic length, R(t) with t1/3 can be observed in the simulation results. The compatibility of the result pattern and the pattern on the substrate was measured by a scalar, Cs . The influence of the solvent concentration on the refinement of the result pattern was studied in this work. A critical time can be observed from the evolution of the value of Cs . The morphology evolution of a system considering the solvent evaporation was also studied.

Commentary by Dr. Valentin Fuster
IMECE2008-67312 pp. 225-232; (8 pages)

The enhancement of polymers by nano fillers can improve their mechanical properties beyond those achieved in macro, meso and micro composites with the same volume fraction, particle morphology and particle aspect ratio. The basic enhancement mechanism is analogous to traditional composite mechanics and can be partly explained by classical composite theories. However, the much higher enhancement efficiency of fillers at the nano scale, i.e. the particle size effects, can not be explained by classical composite mechanics theories alone. The interphase is a main structural feature of nano composites within which a significant surface-to-volume ratio is achieved, and it plays a crucial role in understanding the size effects and enhancement mechanisms of nanocomposites. In this investigation, a semi-empirical method of determining the interphase thickness and elastic properties is developed by a combination of finite element simulation, thermodynamic formulation and experimental calibration and applied to LBL (layer by layer) polyurethane-clay nanocomposites studied by Podsiadlo, et al. [1, 2] and Li, et al. [3]. Based on this study, the classical two-phase Mori-Tanaka model is extended into three-phases by introducing the interphase through a two-step procedure using the concept of an effective matrix. It is shown that this two-step Mori-Tanaka method can predict the brittle to ductile transition in terms of interphase overlap. Most importantly, this approach overcomes a serious drawback of the classic Mori-Tanaka model: size-independency. Particle size effects as well as shape effects and their interactions can be studied as applications of this method.

Commentary by Dr. Valentin Fuster
IMECE2008-67373 pp. 233-234; (2 pages)

We report the production of transparent, amorphous, partially exfoliated polyamide / nanoclay nanocomposites. Stiffness and yield strength increase with nanoclay content, while work-of-fracture remains constant below an inorganic content of 0.25 vol%.

Topics: Nanocomposites
Commentary by Dr. Valentin Fuster
IMECE2008-67578 pp. 235-242; (8 pages)

Poly (methyl methacrylate) (PMMA) and acrylonitrile-butadiene-styrene (ABS) – multiwall carbon nanotube (MWNT) and chopped carbon fiber (CCF) composites were prepared by a melt mixing protocol at various concentrations. Specimens were fabricated and tested using constant amplitude-of-deflection fatigue testing. The numbers of cycles to failure were recorded and analyzed using the linear version of the 2-parameter Weibull model. In the PMMA matrix, the 1.0vol% MWNT reinforced composites outperformed the neat PMMA matrix by +396% while the 1.0vol% CCF composites increased fatigue life by +198% over the control. The increase in fatigue life may be attributed to the nanoscale dimensions of the MWNTs. This enables them to directly interact with the matrix at the sub-micron scale where damage such as crazing begins, which ultimately initiates a critical crack that leads to failure of the specimen. The ABS composite specimens did not show any increase in fatigue life. The underlying reasons for the lack of fatigue improvement remain unclear.

Commentary by Dr. Valentin Fuster
IMECE2008-67693 pp. 243-247; (5 pages)

Nano-indentation is increasingly used to estimate the mechanical properties of polymeric films of nanometer-scale thickness. Hardness and modulus, as obtained on indentation are parameters that are strongly dependent upon tip geometry, elastic and inelastic material behavior, and specimen boundary conditions. The objective of this study was to analyze the mechanical response of nano-indentation loading on surfaces and interfaces of polymer films both linear and cross-linked. Polymer films on nano-indentation show confinement effect on their modulus and hardness. Reduced modulus and hardness in polyester and polycarbonate show strain softening that is associated polymer chain flexibility. The spatial constraints imposed to the plastic flow of the interface layer by the rigid indenter and substrate surfaces produce a dynamic effect, demonstrated by the loading rate dependence of the deformation response. In nano-indentation of cross-linked polymers, entanglements physical and chemical affect reduced modulus and hardness dependence on strain. Strain softening and strain hardening as well as dynamic frictional response are applied to indented polymer films consisting of surface, intermediate, and interface layers.

Commentary by Dr. Valentin Fuster
IMECE2008-68431 pp. 249-255; (7 pages)

Electrical and elastic properties of multiwalled carbon nanotubes (MWCNTs) reinforced polypropylene (PP) nanocomposites were studied experimentally and theoretically. The MWCNT-PP nanocomposites samples with a range of 0 to 12 wt% MWCNT were injection molded using different injection velocities. These nanocomposites were characterized for their electrical resistance using 2-Probe measurement and their tensile properties. Parallel to the experimental investigation, a percolation theory was applied to study the electrical conductivity of the nanocomposite system in terms of content of nanotubes and injection rate. Both Kirkpatrick [1] and McLachlan [2] models were used to determine the transition from low conductivity to high conductivity which designates as percolation threshold. Both experimental and modeling results have shown that the electrical conductivity increased suddenly as the content of MWNTs was close to percolation threshold of 3.8 wt%. The injection speed also showed an effect on electrical conductivity of the composites. In addition, several micromechanical models were applied to elucidate the elastic properties of the nanocomposites. The results indicate that the interphase between the carbon nanotubes and polymers plays an important role in determining elastic modulus of the system.

Commentary by Dr. Valentin Fuster
IMECE2008-66672 pp. 257-265; (9 pages)

Design and quality assurance of micro gear wheels and involute gear wheels involve multiple challenges regarding prediction of functionality and life cycle performance of complex and wear-resistant micromechanical systems. First of all, this is due to the fact that up to now no tolerance system for micro dimension has been defined. In second place, most measurement strategies for the dimensional characterization of involute micro gears cannot be brought forward from the macro world just as they are. There is few knowledge about the relevant quality characteristics for these micro systems, optical sensors’ precision is affected by fuzzy edges detection and no tactile scanning modes for relevant features smaller than 100 μm exist, which is the scale normally applied in macroscopic dimensions. Simulation methods for analyzing the influence on the whole system of different components’ geometrical deviations are very valuable to supplement the knowledge based on real tests. Furthermore, it could be also necessary to consider the material’s anisotropy caused by the not negligible grain structure to evaluate the stress field correctly. Therefore, a new approach for design and quality assurance needs to be developed, in order to assure the functionality and long-term performance of molded micro systems. This work uses a planetary gear train as a demonstrator. A validation of the entire product functionality test chain has to be conducted to come closer to an integrated robust design approach for mechanical micro systems — from simulation in the early design stages to test and to quality assurance of large series production. This paper outlines a threefold methodological approach integrating dimensional measurement, virtual tests based on real geometry and physical tests of real gears. The measurement of the micro systems components both in disassembled as in assembled state is conducted using multisensory coordinate measurement machines. Based on the measured contours of real gears, virtual gears are derived and meshed for its subsequent use in adequate FEA model. Simultaneously, the gears are mounted and tested on a micro gear test rig. Both simulation and test rig conduct a radial composite inspection adapted to the micro scale; results are then compared. Micro gears molded of zirconium oxide were selected as a demonstrator for the presented methodology. These 12 teeth gears, with a diameter of approximately 2.0 mm and a modulus of 169 μm, are produced within Collaborative Research Center (CRC) 499 “Development, Production and Quality Assurance of Primary-Shaped Micro Parts made of Metallic and Ceramic Materials” of the German Research Foundation (DFG).

Commentary by Dr. Valentin Fuster
IMECE2008-67593 pp. 267-273; (7 pages)

Large-scale nanostructure arrays with spatial coherence are useful for many applications. Conventional nanofabrication techniques such as electron beam lithography and focused ion beam lithography are expensive and time-consuming. In this paper, long-range ordered Au nanodisk arrays were fabricated on glass substrates using nanosphere lithography (NSL) combined with reactive ion etching (RIE) techniques. The morphology and size distribution of the Au nanodisks were examined with scanning electron microscopy (SEM) and atomic force microscopy (AFM). The sensitivity of the localized surface plasmon resonance (LSPR) of the Au nanodisk arrays to change in the surroundings’ refractive index was evaluated by integrating the Au nanodisk arrays into microfluidic channels. The measured sensitivity was supported by discrete dipole approximation (DDA) calculations. Further, we designed and fabricated an all-optical plasmonic switch based on the Au nanodisk arrays and photoresponsive liquid crystals (LCs). The high-quality optical properties and high-degree spatial uniformity of the nanodisk arrays, together with simple, low-cost fabrication and easy integration with microfluidic system, suggest tremendous potential in using these nanostructures in many other applications, including biosensing and imaging, surface-enhanced Raman spectroscopy (SERS), and plasmonic tweezers.

Topics: Manufacturing
Commentary by Dr. Valentin Fuster
IMECE2008-67711 pp. 275-279; (5 pages)

This paper reports a fabrication technique for high strength glass-metal composite micro/nanonozzles with orifice diameters ranging from 430 nm to >100 μm. Unlike the conventional methods used to build micro/nanonozzles, the fabrication technique discussed in this paper is a non-lithographic approach. It uses conventional pulled borosilicate micropipettes as a foundation and nickel as a strengthening layer to build high pressure withstanding micro/nanonozzles. Pipettes built using the pulling process offer a smooth transition to the fluid from the reservoir to the tapered part of the nozzle, providing an ideal geometry from fluid flow and stress point of view. The nozzles are tested for high pressure withstanding capacity by integrating them with a high pressure fluidic setup to drive microjets. As an example, a 1.5 μm diameter nozzle, tested with propane as the working fluid to drive a microjet is observed to withstand pressures upto 10.5 MPa. Apart from simplicity of the fabrication process, this approach also offers the ability to incorporate a wireless temperature control system for the nozzles.

Commentary by Dr. Valentin Fuster
IMECE2008-67843 pp. 281-286; (6 pages)

This paper describes the basic Direct Polymer Patterning On Substrate Technique (DPPOST) process and a modified process currently under development to provide higher robustness in the fabrication process, with the goal of achieving near 100% patterning yields. The patterning of soft-polymers and elastomers has gained interest in the last decade as the material of choice for lab-on-the chip applications; i.e., forming micro-fluidic and bio-reactor chambers. Recently, a lithographically compatible patterning method for soft-polymers and elastomers have been demonstrated by using SU-8® hard polymer resists as robust lift-off molds. This patterning technology, DPPOST, has the ability to form a wide range of structural features found in MEMS, from tens of millimeter structures to micrometer level resolutions. It has been used to embed nano-particles, such as carbon-black (∼45nm mean radius) and metal particles, and allows lithographic alignment of electrodes on micro-fluidic channels previously not possible with soft-lithography fabricated PDMS devices. The modified-DPPOST process uses conformal coating of Omnicoat™ nano-films to provide a barrier between the SU-8® and the patterned polymer, hence reducing stiction during the release process.

Topics: Elastomers , Polymers
Commentary by Dr. Valentin Fuster
IMECE2008-67861 pp. 287-294; (8 pages)

This paper describes the characterization and modeling of capacitive micromachined ultrasonic transducers (cMUTs). Computational models of the transducers were produced through the combined use of finite element analysis (FEA) and lumped element modeling. Frequency response plots were generated for both transducers in air and water environments. Through the use of laser Doppler velocimetry, transient step response and frequency sweep tests were performed on single array elements. These measurements are compared to the predicted results represented in the models. The computational results for both coupled and uncoupled arrays are compared, and show a significant increase in the array bandwidth due to coupling. Frequency sweep tests were also performed on column array elements, and results were compared between driven and adjacent, non-driven columns.

Commentary by Dr. Valentin Fuster
IMECE2008-67955 pp. 295-303; (9 pages)

Polymeric nanofibers are finding increasing number of applications and hold the potential to revolutionize diverse fields such as tissue engineering, smart textiles, sensors, and actuators. Aligning and producing long smooth, uniform and defect-free fibers with control on fiber dimensions at the submicron and nanoscale has been challenging due to fragility of polymeric materials. Besides fabrication, the other challenge lies in the ability to characterize these fibers for mechanical properties, as they are widely believed to have improved properties than bulk due to minimization of defects. In this study we present an overall strategy for fabrication and mechanical characterization of polymeric fibers with diameters ranging from sub-50 nm to sub-microns. In the proposed fabrication strategy, polymeric solution is continuously pumped through a glass micropipette which is collected in the form of aligned fiber arrays on a rotating substrate. Polymer molecular weight and polymer solution concentration play dominant roles in controlling the fiber dimensions, which can be used to deposit fibers of different diameters in the same layer or successively built up multi-layer structures. Using this approach, we demonstrate single and multi-layer architectures of several polymeric systems such as Polystyrene (PS), Poly(methyl methacrylate) (PMMA), Poly lactic acid (PLA), and poly(lactic-co-glycolic acid) (PLGA). Further, we demonstrate the ability to manufacture PMMA fixed-free boundary condition cantilevers by breaking the fixed-fixed boundary condition PMMA fibers using Atomic Force Microscope (AFM) in the lateral mode. An integrated approach for mechanical characterization of polymeric fibers is developed. In this approach, the fibers are first deposited on commercially available Transmission Electron Microscopy (TEM) grids in aligned configurations and are mapped for accurate locations under the TEM. Subsequently, the fibers are carefully placed under the AFM and mechanically characterized for flexural modulus using lateral force microscopy (LFM). Finally, accurate fiber dimensions are determined under the Scanning Electron Microscope (SEM). The unique advantage of this approach lies in the ability to deposit a large number of fibers with tunable diameters in aligned configurations with fixed-fixed boundary conditions and requires no external manipulation. Finally, we present a novel methodology to study the resonance characteristics of fixed-fixed boundary condition suspended fibers using a commercially available Laser Doppler Vibrometer (LDV) for sensor applications. The methods developed in this study will greatly aid in increasing our fundamental knowledge of polymeric materials at reduced lengthscales and allow integration of these one-dimensional building blocks in bottom-up assembly environments.

Commentary by Dr. Valentin Fuster
IMECE2008-68620 pp. 305-312; (8 pages)

A key factor for the propagation of technological applications is the miniaturization of respective components, subsystems and overall systems. To meet future requirements in such size decreasing environments the packaging and mounting technology needs new impulses. 3D-MIDs (three-dimensional molded interconnect devices) exhibit a high potential for smart packages and assemblies. A three-dimensional shaped circuit carrier allows the integration of various functional features (e.g. electrical connections, housing, thermal management, mechanical support). This combination makes a further system shrinking possible. Yet, the mounting of high-density area-array fine-pitch packaged semiconductors (BGA, CSP, MCM) or bare dies to 3D-MIDs is problematic. The lack of a three-dimensional multilayer technology makes a collision free escape routing for devices with a high I/O count difficult. Therefore a new 3D-MID multilayer process was developed and combined with an established 3D-MID metallization process. A demonstrator with three metallization layers, capable, e.g., for flip-chip mounting of area-array packages, is fabricated. The multilayer structure of the demonstrator is investigated with respect to the mechanical and electrical behavior.

Commentary by Dr. Valentin Fuster
IMECE2008-68975 pp. 313-317; (5 pages)

A thermal system used to evaluate a high throughput 96 continuous flow polymerase chain reactor (CFPCR) array was designed, fabricated, and tested. Each polymerase chain reactor (PCR) in the array required three different temperature zones to realize denaturaiton at 90°C–94°C, renaturation at 50°C–70°C, and extension at 72°C; a total of 288 temperature zones were required for the 96 CFPCR array. In an initial configuration, 18 copper strips were used to define the 288 temperature zones. Each copper strip was controlled by a PID feedback control loop. Numerical simulations were used to understand the thermal crosstalk phenomena between the micromilled copper strips, which were tightly packed since the high throughput micro-titer plate format restricted each CFPCR to a square 8 mm on a side. The lowest achievable temperature in each renaturation zone in this complicated thermal environment was also identified. Thermal crosstalk limited the minimum renaturation temperature to 61.1°C. An infrared camera was used to investigate the temperature uniformity over a 0.25 mm thick polycarbonate sheet mounted on the thermal system. The temperature distribution was not uniform due to poor contact between the copper strips and device, warm air accumulated between the packed copper strips, and greater heat transfer around the boundaries of the device. More work is required to overcome these limitations and achieve a more uniform temperature distribution for a multi well CFPCR.

Commentary by Dr. Valentin Fuster
IMECE2008-69182 pp. 319-324; (6 pages)

We present an electromechanical analysis of a novel double-sided driven carbon nanotube-based electromechanical resonator. The device comprises a cantilevered carbon nanotube actuated by two parallel-plate electrodes. Close-form analytical solutions capable of predicting the steady-state resonation of the device and its resonant pull-in conditions are derived using an energy-based method. Our close-form formulas clearly reveal the complex relationship among the device geometry, the driving voltages, and the device’s electromechanical dynamics. Our theoretical modeling shows that the stable steady-state spanning range of the resonating cantilever substantially exceeds the previously reported quasi-static pull-in limit for single-sided driven cantilevered nanotube-based NEMS, while the resonant pull-in voltage is only a small fraction of the quasi-static pull-in voltage. The unique behaviors of this novel device are expected to significantly enhance the applications of electromechanical resonators in the fields of signal processing, mass and force sensing, and chemical and molecule detection.

Topics: Carbon , Modeling , Nanotubes
Commentary by Dr. Valentin Fuster
IMECE2008-66062 pp. 325-331; (7 pages)

The aim of this paper is to present an Extended Kantorovich approach to simulate the static deflection of microplates under electrostatic voltage. The model accounts for the electric force nonlinearity of the excitation. Starting from a one term Galerkin approximation and following the Extended Kantorovich procedure, the equations governing the microplate deflection are obtained. These equations are then solved iteratively with a rapid convergence procedure to yield the desired solution. The results are validated, comparing them with other theoretical results and experimental findings, reported in the literature. It is shown that rapid convergence, high precision and independency of initial guess function makes the EKM an effective and accurate design tool for design optimization.

Commentary by Dr. Valentin Fuster
IMECE2008-66233 pp. 333-341; (9 pages)

When designing a MEMS device, both the mechanics and the control logic have to be taken into account due to their strict interaction. The considered MEMS gyroscope is made of vibrating masses suspended through micro-beams with respect to the substrate. To increase the sensitivity of the MEMS gyro (i.e. the displacement of the suspended masses), it has been decided to work at low pressure values (26Pa). At first, the mechanical modelling of the device was carried out, paying special attention to the air damping at low pressure values. Then, the electronics used for the control of the device was integrated into the model thus obtaining a complete description of the sensor allowing to optimize design parameters. In order to validate the developed model, the MEMS sensor was produced and tested.

Commentary by Dr. Valentin Fuster
IMECE2008-66234 pp. 343-350; (8 pages)

MOEMS (Micro-Opto-Electro-Mechanical System) are MEMS in which the optical part plays a dominant role. MOEMS are usually used to deflect and/or focus light from/on a given spot thus acting as optical switches. Now, imagine to have a 2D array of optical switches (e.g. a CCD) and to rotate the MOEMS so that the incoming light is focused subsequently on each of these switches. What you have designed is a 2D scanner. If you reverse the process (i.e. one light source and the MOEMS projecting on a screen) you have designed a projector. The main advantage of a MOEMS projector and/or scanner are the reduced dimensions/weight with respect to traditional projectors and scanners and the fact that the necessity of optical lenses is greatly reduced thus reducing the cost of the system. In the present paper, a 2D MOEMS projector/scanner is designed. The peculiarity of the system is that the MOEMS mirror is able to rotate around two perpendicular axes having only one excitation point instead of two thus allowing to design a system that can be produced from a planar surface. This is achieved by applying the excitation to the micro-mirror indirectly through a supporting frame that is free to rotate around an eccentric vertical axis. The micro-mirror, instead, is free to rotate around an eccentric horizontal axis with respect to such supporting frame (figure 3). To be able to excite the motion of the micro-mirror along two perpendicular axes, the actuation force has to be applied in a point that is not nodal for any of the two rigid vibration modes of the system, has to be biharmonic and, to increase displacements, has to have frequencies that correspond to the two rigid eigenfrequencies of the micro-mirror.

Commentary by Dr. Valentin Fuster
IMECE2008-66292 pp. 351-356; (6 pages)

Among many different mechanisms that are used for excitation and detection of vibration of micro-beam resonators, electrostatic comb-drives have the benefit of simplicity and large range of linear operation. The disadvantage of using comb-drives is the effect of added mass to the beam; however, the analytical model of the beam-mass system predicts that this shortcoming can be overcome by proper adjustment of the mass, rotary inertia, and location of the comb-drive. In addition, the analytical model can predict the effect of the axial force of the beam on the resonance frequencies. In this paper, the results of the experiments on two resonators are presented. These results are used to verify the validity of the analytical model and finding its parameters. Very close agreement between the theory and experiment is observed. The residual stress of the MEMS structural layer is measured using the calibrated analytical model parameters.

Commentary by Dr. Valentin Fuster
IMECE2008-66517 pp. 357-363; (7 pages)

We study the effect of bias voltage VDC on the effective nonlinearity of electrostatically clamped-clamped microbeam resonators. We identify three domains in the resonator response: hardening-type, softening-type, and near-linear behaviors. In the near linear domain we show that we can increase the power handling of the resonator without distorting its phase noise performance. We investigate the mixing of low frequency 1/f noise into the input signal. This causes phase distortion of the output signal and is quantized as its phase noise. We find that the amplitude and phase responses of the resonator’s displacement are coupled to each other through the effective non-linearity co-efficient (S), which distorts its phase response in the nonlinear regime. Finally we also present closed form expressions for resonator displacement and current in both linear and non-linear regimes.

Commentary by Dr. Valentin Fuster
IMECE2008-66647 pp. 365-370; (6 pages)

We are examining microactuation techniques by employing the electrokinetic and galvanotactic behavior of certain bacteria. We cultured selected strains of swarming Serratia marcescens which were attached to microstructures using a blotting technique that creates a bacterial monolayer carpet. These bacterial carpets naturally self-coordinate to propel the microstructures. The microstructures were placed in an open channel and a voltage was applied and polarity was switched. We have demonstrated directional control of the motion of the microstructures patterned with bacteria. This mobility is due to the patterning of bacteria on the microstructure surface and arises from a combination of electrokinetic effects and galvanotaxis.

Commentary by Dr. Valentin Fuster
IMECE2008-66728 pp. 371-384; (14 pages)

We study the feasibility of employing subharmonic resonance of order one-half to create a bandpass filter using two clamped-clamped microbeam resonators connected by a weak coupling beam. We discretize the distributed-parameter system using the Galerkin procedure to obtain a reduced-order model composed of two nonlinear coupled ODEs. It accounts for geometric and electric nonlinearities as well as the coupling between these two fields. Using the method of multiple scales, we determine four first-order nonlinear ODEs describing the amplitudes and phases of the modes. We use these equations to determine closed-form expressions for the static and dynamic deflections of the structure. The basis functions in the discretization are the linear undamped global mode shapes of the unactuated structure. We found that we can not produce a single-valued response for small excitation amplitudes. So that, it is impractical to use a single structure made of two mechanically coupled beams excited subharmonically in filtration. But we can use a pair of structures to build a bandpass filter by operating one in the softening domain and the other in the hardening domain and, more importantly, implementing processing logic and hardware schemes. However, the complications brought about by mechanically coupling of two microbeams can be avoided by using a pair of uncoupled beams. This makes the fabrication and modeling processes much easier. Using subharmonic excitation with mechanically uncoupled microbeams to realize bandpass filters is the subject of the next work.

Commentary by Dr. Valentin Fuster
IMECE2008-66794 pp. 385-393; (9 pages)

Wireless distributed micro-sensor systems have numerous applications, from equipment diagnostic and control to real time biomedical monitoring. A major obstacle in developing autonomous micro-sensor networks is the need for local, autonomous electric power supply, since using a battery is often not a viable solution. This work investigates a novel micro-power generator converting ambient vibrations to electrical energy via electrostatic transduction employing a comb-like variable capacitor, with a switchable dielectric constant. The micro-generator is designed to operate in an inplane, gap closing, and charge constrained configuration. Enhanced power levels are obtained by pulling a high dielectric constant liquid between the capacitor gaps via electrochemical force effect at maximum capacitance position. The effects of squeeze film damping and variable electrostatic force induced by the varying spatial distribution of dielectric medium are taken into account. In this configuration the microgenerator exhibits high quality factors, thus providing an output power higher of approximately 168 μW.

Commentary by Dr. Valentin Fuster
IMECE2008-66795 pp. 395-399; (5 pages)

Design, modeling, fabrication, and testing of a new type of optical communication node enabled by micromachined mirror is described. The micromirror device is capable of switching within 1μs between two states — one planar and reflective in the substrate surface normal direction and one corrugated and diffractive. It is used as one facet in a corner-cube retroreflector that can communicate by returning a temporally modulated beam to an interrogating laser transceiver. Design of a very low-cost device suitable for use as a communication node at ranges of several hundred meters is detailed. In particular, fabrication approaches are described for the micromirror that minimize the number of layers and process steps while employing scalable, high-volume thin film manufacturing. Ultimately, a process comprised of three thin film depositions, on lithographic patterning step, and one timed etch produces a robust and easily mass produced device. The devices that were tested featured up to 10:1 modulation of an incident beam in some configurations, and allow robust, secure optical switching through free space. Processes for combining the modulator with two orthogonal facets to comprise a retroreflector are described, and adequate orthogonality is demonstrated for communication over the intended range. Intrinsic stress behavior of tensile thin films are exploited to achieve nanometer scale flatness in the unactuated device, and a compact electronics driver is used to encode data and frequency modulate the active device. Electromechanical models that predict device behavior during electrostatic actuation are detailed. A functional prototype communication system using the modulated retroreflector and a telescope containing a 1.5μm wavelength laser and photodetector decoder is described, and secure communication is demonstrated over a free space path.

Commentary by Dr. Valentin Fuster
IMECE2008-67034 pp. 401-408; (8 pages)

In this paper, dynamic behavior and pull-in phenomenon of electrically actuated rectangular micro plates under the effect of squeeze-film damping and nonlinear electrostatic force is studied. Finite element method is implemented in order to drive weak formulations of linear and nonlinear micro plate equations of motion based on classical plate theory (CPT) (for thin microplates with moderate nonlinearity) and squeeze-film damping based on Reynolds nonlinear equation. Finally, an efficient reduced-order model contingent on singular value decomposition method (SVD) is used to study dynamic pull-in phenomenon. This model is constructed by the global basis functions achieved from a few runs of FEM and can reduce simulation time. Validating the macro model results with full FEM simulation shows that this model is effective.

Commentary by Dr. Valentin Fuster
IMECE2008-67379 pp. 409-420; (12 pages)

We propose and validate a new architecture for wide-band vibration-based MPGs. This architecture replaces a linear oscillator with a piecewise-linear oscillator as a vibrations harvesting element. Analytical, numerical, and experimental techniques are used to analyze a prototype of an electromagnetic MPG designed and constructed using the new architecture. The new architecture increases the bandwidth of the MPG during an up-sweep compared to a traditional MPG, while maintaining the same bandwidth in a down-sweep. Closed-form expressions for the response of the piecewise-linear MPG as well as the size of the up-sweep bandwidth are presented and validated experimentally. Simulations show that under a random-frequency base excitations new architecture collects more energy collected by the traditional architecture.

Commentary by Dr. Valentin Fuster
IMECE2008-67553 pp. 421-431; (11 pages)

The static response of an electrostatic micro-catilever beam has been obtained by using Galerkin’s method. To make the system bi-stable, a controller has been added and the static response profile is presented using a multi-mode model for the beam. The number of mode shapes leading to convergence has been studied. The softening effect of adding more mode shapes has been investigated along with the effect of changing the system parameters on the static response. Decreasing the controller gain has been found to widen the voltage range of the bi-stability region and increasing the sensor amplification factor is shown to push the upper equilibrium point away from pull-in. Properly choosing these parameters can adjust the range of voltage for bi-stability. By doing a linearization about the stable fixed points, we also found the two natural frequencies for each stable equilibrium point. Finally, we have found the dynamic response of the bistable system using one- and three-mode-models. The basins of attraction for each stable fixed point and the exchange of energy between the two potential energy wells (equilibrium points), are demonstrated.

Commentary by Dr. Valentin Fuster
IMECE2008-67559 pp. 433-442; (10 pages)

We present modeling, analysis, and experimental investigation for dynamic instabilities and bifurcations in electrostatically actuated resonators. These instabilities are induced by exciting a microstructure with a nonlinear forcing composed of a DC parallel-plate electrostatic load superimposed to an AC harmonic load. Because of the dominant effect of the electrostatic nonlinearity, several resonances and nonlinear phenomena are induced. Examples of these are the excitation of secondary-resonances, superharmonic and subharmonic, at half and twice the natural frequency of the microstructure. Also, local bifurcations, such as saddle-node and pitchfork, and global bifurcations, such as the escape phenomenon and the homoclinic tangling may occur. These lead to undesirable jumps, hysteresis, and dynamic pull-in instabilities in MEMS devices and structures. The present work represents an attempt to explore these topics in more depth. The first part of this paper is focused on analyzing and studying the nonlinear dynamics of a capacitive device both theoretically and experimentally with a focus on the case of primary-resonance excitation (near the fundamental natural frequency of the structure). The device is made up of two cantilever beams with a proof mass attached to their tips. A nonlinear spring-mass-damper model is utilized, which accounts for squeeze-film damping. Long-time integration for the equation of motion is used to compare with the obtained experimental data. Then, global dynamic analysis is conducted using a finite difference method (primary resonance) and shooting method (subharmonic resonance) combined with the Floquet theory to capture periodic orbits and analyze their stability. The domains of attraction (basins of attraction) for selected data are calculated numerically. Experimental data revealing primary and sub-harmonic resonances, dynamic pull-in, and the escape-from-a-potential-well phenomenon are shown and compared with the theoretical results.

Commentary by Dr. Valentin Fuster
IMECE2008-67695 pp. 443-448; (6 pages)

We present a technique for independently exciting two resonant modes of vibration in a single-crystal silicon bulk mode microresonator using the same electrode configuration through control of the polarity of the DC actuation voltage. Applications of this technique may include built-in temperature compensation by the simultaneous selective excitation of two closely spaced modes that may have different temperature coefficients of resonant frequency. The technique is simple and requires minimum circuit overhead for implementation. The technique is implemented on square plate resonators with quality factors as high as 3.06 × 106 .

Commentary by Dr. Valentin Fuster
IMECE2008-67763 pp. 449-455; (7 pages)

In this paper, we develop a mathematical model of an electrostatic MEMS beam undergoing impact with a stationary electrode subsequent to pull-in. We model the contact between the beam and the substrate using a nonlinear foundation of springs and dampers. The system partial differential equation (PDE) is converted into coupled nonlinear ordinary differential equations (ODEs) using the Galerkin method. A numerical solution is obtained by treating all nonlinear terms as external forces.

Commentary by Dr. Valentin Fuster
IMECE2008-67822 pp. 457-461; (5 pages)

It was found experimentally from our previous study that the operation of the piezoelectric actuator (PEA) and the friction in the piezoelectric stick-slip actuator (PE-SSA) can cause significant rise in temperature, thereby degrading the performance of the actuator. This paper presents a dynamic model for the PE-SSA by taking into account thermal effect. In particular, the dynamic model is developed by integrating the PEA model proposed by Adriaens et al. [1] and the LuGre friction model proposed by De Wit et al. [2]; the parameters involved in the models are determined using a system identification approach. Experiments are carried out to verify the effectiveness of the model. It is shown that the simulation and experimental results are in a good agreement. This study provides a new way to model thermal effect for other micro motion systems.

Commentary by Dr. Valentin Fuster
IMECE2008-68008 pp. 463-466; (4 pages)

In this paper we follow an analytical and finite element modeling approach to study the effect of anchor over/under-etch on the post-release tip displacement of MEMS cantilever beams. We show that the last release step is particularly critical in controlling the beam’s post-release displacement, which is of primary importance in a variety of applications. In this paper we isolate the effects of mean stress and imperfect anchor. We assume negligible gradient stress since its effect has been studied in detail in the literature. Beams with perfect anchors and zero under-etch are also presented for comparison. The results emphasize that through careful control of the fabrication parameters and anchor structure, the initial displacement profile and thus performance of micro-cantilevers can be accurately controlled.

Commentary by Dr. Valentin Fuster
IMECE2008-68042 pp. 467-479; (13 pages)

We study the dynamic behavior of an electrostatic MEMS resonator using a model that accounts for the system nonlinearities due to mid-plane stretching and electrostatic forcing. The partial-differential-integral equation and associated boundary conditions representing the system dynamics are discretized using the Differential Quadrature Method (DQM) and the Finite Difference Method (FDM) for the space and time derivatives, respectively. The resulting model is analyzed to determine the periodic orbits of the resonator and their stability. Simultaneous resonances are identified for large orbits. Finally, we develop a first-order approximation of the microbeam dynamic response, which reveals an erosion of the basin of attraction of the stable orbits that depends heavily on the amplitude and frequency of the AC excitation. Simulations show that the smoothness of the boundary of the basin of attraction can be lost to be replaced by fractal tongues, which increase the sensitivity of the microbeam response to initial conditions. As a result, the locations of the stable and unstable fixed points are likely to be disturbed.

Commentary by Dr. Valentin Fuster
IMECE2008-68061 pp. 481-488; (8 pages)

This paper presents two fuzzy-based controllers designed to scan non-contact atomic force microscopes (AFM) over a specimen surface. Firstly, we develop a conventional fuzzy controller to achieve asymptotic probe tip tracking for bounded tip trajectories. Secondly, a hybrid PD-fuzzy controller is designed for the same purpose where the PD gains are tuned online by means of fuzzy logic. Finally, we compare the efficacy and requirements of the two controllers and compare their results to those of other recently proposed controllers for the same purpose. We show improved performance using both of our controllers; the most significant advantage being increased controller bandwidth and thus faster AFM scan rates. However, the PD-fuzzy controller is found to impose more realistic actuation demands on the plant than the Fuzzy controller.

Commentary by Dr. Valentin Fuster
IMECE2008-68081 pp. 489-493; (5 pages)

This research presents a novel linearly tunable MEMS capacitor with flexible electrode and modified structural stiffness. The capacitor is designed for PolyMUMPs as a standard three-structural-layer fabrication process. The moving electrode is divided into segments interconnected through torsional springs. Under each connecting point (node) two flexible and rigid steps are located. The flexible steps are cantilever beams, and as the bias voltage increases, they touch their corresponding nodes and consequently their stiffnesses are added to the total structural stiffness. This is the core idea of the proposed design to linearize the capacitance-voltage (C-V) response. An analytical model is developed to investigate the behavior of the new capacitor. In this model, the governing equations of the capacitor are numerically solved to obtain the system’s C-V response. An optimization problem with different design variables, such as dimensions of the segments or the beams stiffness coefficients, is solved to maximize the linearity of the C-V curves. The numerical results demonstrate drastic improvement in capacitors performance, where a highly linear C-V response and a maximum tunability of 94% is reached.

Commentary by Dr. Valentin Fuster
IMECE2008-68090 pp. 495-500; (6 pages)

This paper presents a novel geometry and modified structural stiffness for electrostatically actuated MEMS tunable capacitors. The design is based on parallel-plate configuration and four triangular plates are put together to form a butterfly shape flexible moving electrode. Each triangle is suspended by three uneven supporting beams. The capacitor is also equipped with extra beams, called here the “middle beams”, located under the triangles’ corners (nodes). An analytical model is developed to solve the governing equations of a triangular-plate electrode with uneven sides and supporting beams, where the stiffness of the middle beams is gradually added to the system as actuation voltage increases. The numerical simulations reveal that each triangle can be individually tuned up to 150% and the capacitance-voltage (C-V) response is broken into small sections due to added middle beams. Using the model developed in this paper and by design optimization, a linear C-V response is obtained, where the tunability in linear region reaches 100%. The simplicity of the proposed design allows the device to be fabricated using a three-structural-layer process such as PolyMUMPs and could therefore be monolithically integrated with other RF devices and ICs. Moreover, adding additional insulator layer on top of the fixed electrode increases the tunability to over 200% displaying a smooth and low sensitive response.

Commentary by Dr. Valentin Fuster
IMECE2008-68600 pp. 501-510; (10 pages)

Instabilities in a vibratory MEMS gyroscope that is subject to stochastic fluctuations in input angular rates are investigated. The vibratory-type gyroscope considered in the present study is of the spring-mass type. For the purpose of acquiring stability conditions, when the angular rate input is subject to small intensity stochastic fluctuations, dynamic behaviour of stochastically perturbed linear gyroscopic systems is studied in detail. An asymptotic approach based on the method of stochastic averaging has been employed for this purpose, and closed-form conditions for mean square stability of dynamic response are obtained for the case of exponentially correlated noise. Results are shown to depend only on those values of the excitation spectral density near twice the natural frequencies and the combination frequencies of the system. The presented results remain valid if the stochastic parametric excitation has a small correlation time compared with the system relaxation time. Stability predictions have been illustrated via stability diagrams in the power-spectral–density-damping-ratio space. Further, to illustrate the applicability of the results in practice, conditions for varying input angular rates are mapped. Although, the above conditions are predicted for the spring-mass type gyroscope, the predictions can be easily translated to other vibratory gyroscope designs.

Commentary by Dr. Valentin Fuster
IMECE2008-68795 pp. 511-519; (9 pages)

The dynamic behavior of an atomic force microscope cantilever probe is studied for dual frequency excitation. By using the Euler-Bernoulli beam equation with a multi-mode approximation, the system is modeled with base excitation and tip-sample interaction forces obtain from molecular dynamics simulations. The dynamic response of the cantilever probe is simulated for a range of separation distance values and analyzed using Poincaré sections, bifurcation diagrams, and spectral analysis. The response of the cantilever probe is found to display a qualitative change when influenced by surface forces. The frequency component at half of the fundamental frequency provides an effective way to monitor the amount of force that the probe is applying to the surface of the sample. With this frequency component, an amplitude modulation operation mode is proposed in order to maintain near-grazing behavior during imaging.

Commentary by Dr. Valentin Fuster
IMECE2008-68798 pp. 521-527; (7 pages)

Photo-thermally actuated polymer (SU-8) microgrippers were designed, simulated and characterized. The microgrippers were actuated by thermal expansion of compliant polymer parts. The required heat was transferred to the devices by laser absorption. The microgrippers were made of a single SU-8 layer and dyed before the experiments for enhanced laser absorption. Finite element simulation results were used to predict the working range of the grippers. It has been demonstrated that using a simple design 22 μm of deflection can be achieved for a micro-gripper of approximately 900 μm long. The gripping experiments have demonstrated the successful operation of polymeric photo-thermal microgrippers.

Commentary by Dr. Valentin Fuster
IMECE2008-68939 pp. 529-533; (5 pages)

A general method of modeling and simulating a micro-machined inertial sensor under closed-loop control by a sigma-delta modulator is presented. The model of the mechanical subsystem captures all 6 mechanical degrees of freedom and the complex electrostatic fields in electrostatic comb drives. A technique is presented that makes transient simulations of the sophisticated mechanical model in conjunction with a modulator more tractable. The benefits of the modeling and simulation approach are demonstrated on an example consisting of a single-axis accelerometer and a second-order sigma-delta modulator. The example shows that transient simulations of the complete system can be performed in relatively short time while capturing cross coupling between mechanical modes of the sensor.

Commentary by Dr. Valentin Fuster
IMECE2008-68965 pp. 535-542; (8 pages)

This work is concerned with the modeling, nonlinear dynamic analysis and control design of an electrostatically actuated clamped-clamped microbeam filter. The model accounts for the mid-plane stretching and nonlinear form of the electrostatic force actuated along the microbeam span. A reduced-order model is constructed, using the method of multiple scales, to examine the microsystem static and dynamics behaviors. To improve the microbeam behavior, a nonlinear feedback controller is proposed. The main control objective is to make it behave like commonly known one-degree-of-freedom self-excited oscillators, such as the van der Pol and Rayleigh oscillators, which depict attractive filtering features. We present a novel control design that regulates the pass band of the fixed-fixed microbeam and derive analytical expressions that approximate the nonlinear resonance frequencies and amplitudes of the periodic solutions when the microbeam is subjected to one-point and fully-distributed feedback forces.

Commentary by Dr. Valentin Fuster
IMECE2008-69018 pp. 543-546; (4 pages)

The problem associated with beams subject to large elastic deflection finds application in micro switches and other MEMS. Perhaps the most challenging task in finding approximate closed form equations is the choice of the function form. While systematic least squares fit of polynomial or power functions are means of finding such approximations, usually the error in the approximations rises with increase in the range of deflection. An interactive/optimization based scheme is employed in this paper to derive the closed form approximate equations for elastic beams experiencing large angular and transverse deflections as a result of application of relatively large loads. It is shown that the approximate equations provide prediction of beam tip angular deflection in the range 0–82 degrees, with error less than 0.2 degrees. The approximate equations are employed to define a pseudo-rigid body link of a compliant MEMS.

Topics: Deflection , Equations
Commentary by Dr. Valentin Fuster
IMECE2008-69238 pp. 547-556; (10 pages)

Micromachined Scanning Grating Interferometer (μSGI) array offers a viable solution to the high resolution, large bandwidth, non-contact and high throughput metrology. Parallel active control of μSGIs is necessary to reduce the effect of positioning errors and ambient vibration noise. To achieve individual control of the μSGIs, the gratings in the μSGI are micromachined on Silicon membranes, which can be electrostatically actuated. These tunable gratings are designed to have sufficient range of motion (∼400nm) and sufficient bandwidth (∼50kHz) for effective noise reduction. The tunable gratings are fabricated successfully using Silicon on Insulator wafers with a two mask process. A novel recurrent calibration based control algorithm is designed to actively control the tunable gratings. The novel algorithm is implemented digitally using FPGA on an array of μSGIs simultaneously. The algorithm compensates for the non-linearities of the actuator and problem due to limited range of motion. A system model is built to design and analyze the control algorithm and is verified by experimental results. Experimental results show 100 times noise reduction at low frequencies and 6.5kHz noise reduction cutoff frequency. A resolution of 1×10−4 nmrms /√Hz is achieved by implementation of this algorithm on μSGI.

Commentary by Dr. Valentin Fuster
IMECE2008-67127 pp. 557-566; (10 pages)

Poiseuille number, the product of friction factor and Reynolds number (f · Re) for quasi-fully developed concentric micro annular tube flow was obtained for both no-slip and slip boundary conditions. The numerical methodology is based on the Arbitrary-Lagrangian-Eulerian (ALE) method. The compressible momentum and energy equations were solved for a wide range of Reynolds and Mach numbers for both isothermal flow and no heat conduction flow conditions. The detail of the incompressible slip Poiseuille number is kindly documented and its value defined as a function of r* and Kn is represented. The outer tube radius ranges from 50 to 150μm with the radius ratios of 0.2, 0.5 and 0.8 and selected tube length is 0.02m. The stagnation pressure, pstg is chosen in such away that the exit Mach number ranges from 0.1 to 0.7. The outlet pressure is fixed at the atmospheric pressure. In the case of fast flow, the value of f · Re is higher than that of incompressible slip flow theory due to the compressibility effect. However in the case of slow flow the value of f · Re is slightly lower than that of incompressible slip flow due to the rarefaction effect, even the flow is accelerated. The value of f · Re obtained for no-slip boundary conditions is compared with that of obtained for slip boundary conditions. The values of f · Re obtained for slip boundary conditions are predicted by f · Re correlations obtained for no-slip boundary conditions since rarefaction effect is relatively small for the fast flow.

Commentary by Dr. Valentin Fuster
IMECE2008-67483 pp. 567-571; (5 pages)

We report our study of an interfacial gas phase formed between a smooth hydrophobic surface and water using pool boiling experiments. Nanoscopically smooth hydrophobic islands of 10 ∼ 100 μm in width are fabricated on a hydrophilic silicon substrate using photolithography. We observe sustained bubble nucleation at superheat as low as 9 °C. The amount of dissolved gas in water does not have a significant effect on bubble nucleation. By comparing the measured minimum superheat for the onset of bubble nucleation with theoretical prediction, we provide indirect evidence for the presence of an interfacial gas phase with widths on the order of 1 μm and contact angle >160°. This high contact angle, which has been postulated to be the reason for the long lifetime of nanobubbles, drastically decreases the nucleation barrier, even on nanoscopically smooth hydrophobic surfaces.

Commentary by Dr. Valentin Fuster
IMECE2008-68710 pp. 573-583; (11 pages)

Examination of metastable states of fluids provides important information pertinent to cavitation and homogeneous nucleation. Homogeneous nucleation, in particular, is an important topic of research. Molecular Dynamics simulation is a well-endorsed method to simulate metastabilitites, as they are limited to mesoscopic scales of length and time and this life-time is essentially zero on a laboratory time scale. In the present study, a molecular dynamics code has been used in conjunction with MOLDY to investigate phase change in a Lennard-Jones liquid. The Lennard-Jones atoms were subjected to different temperatures at various number densities and the pressure was recorded for each case. The appearance of a change of phase is characterized by the formation of clusters or formation of voids as described by the radial distribution function.

Commentary by Dr. Valentin Fuster
IMECE2008-68898 pp. 585-592; (8 pages)

Two-phase microchannels promise an efficient method to dissipate heat from high performance electronic systems by utilizing the latent heat of vaporization during the phase-change process. However, phase-change in microchannel heat sinks leads to challenges that are not present in macroscale systems due to the increasing importance of surface tension and viscous forces. In particular, flow instabilities often occur during the boiling process, which lead to liquid dry-out in the microchannels and severely limits the heat removal capabilities of the system. We propose a microscale breather device consisting of an array of hydrophobic breather ports which allow vapor bubbles to escape from the microchannels to improve flow stability. In this study, we use the combination of microfabricated structures and surface chemistry to separate vapor from the liquid flow. We designed test devices that allow for cross-sectional optical visualization to better understand the governing parameters of a breather design with high vapor removal efficiencies and minimal liquid leakage. We examined breather devices with average liquid velocities ranging from 0.5 cm/s to 4 cm/s and breather vacuum levels between 1 kPa and 9 kPa on the maximum gas removal rate through the breather. We demonstrated successful breather performance. In addition, a model was developed that offers design guidelines for future integrated breathers in microchannel heat sinks. The breathers also have significant promise for other microscale systems, such as micro-fuel cells, where liquid-vapor separation can significantly enhance system performance.

Commentary by Dr. Valentin Fuster
IMECE2008-66319 pp. 593-594; (2 pages)

Metallic thin films have been extensively used as coatings, interconnections, sensors and as part of micro and nano-electromechanical devices (MEMS and NEMS). The conventional substrates utilized to deposit those films are normally rigid, such as silicon. However, for applications where the substrate is subjected to significant mechanical strain (e.g. automotive coatings, electronic textiles, bioengineering, etc.) the film-substrate system needs to be flexible and conformable. Compliant polymeric substrates are ideal candidates for such a task. Some interesting mechanical properties not achieved with conventional rigid substrates can be transmitted to the film by the use of polymeric substrates. In this work, mechanical properties of 50 to 300 nm gold films deposited by thermal deposition over two thermoplastic substrates are investigated. A commercial thermoplastic, Polysulfone (“PSF”), and a home-synthesized isophthalic polyester based on the reaction of 4, 4′ -(1-hydroxyphenylidene) phenol and isophthaloyl dichloride (“BAP”) [1] were used as raw materials for substrate production. Substrates were selected based on their good mechanical properties and flexibility. The use of two different substrates allows us to investigate the influence of the substrate mechanical properties in the bimaterial response. Substrates of 80 μm thickness were prepared by solution casting and cut to rectangular shapes of nominal dimensions of 30 mm × 5 mm. High purity (99.999%) commercial gold splatters were used for film deposition. Gold films with thickness of 50, 100, 200, and 300 nm were deposited onto PSF substrates by thermal evaporation inside a vacuum chamber at 3×10−5 Torr. Au films with 100 nm thickness were also deposited over BAP substrates. Four replicates of each type were deposited (at the same time) and used for tensile testing. Tensile testing of Au/PSF (film thickness 50–300 nm) and Au/BAP (film thickness 100 nm) specimens was conducted. Tests of the neat PSF and BAP substrates (6 replicates) were also conducted as a baseline. Tensile testing was conducted in a small universal testing machine with a load cell of 200 N and a cross head speed of 0.05 mm/min. The film mechanical properties were extracted from the tensile response of the film/substrate system, considered as a bimaterial. Based on sum of forces and strain compatibility, the film modulus (Ef ) and stress (σf ) can be extracted from characteristics of the bimaterial (EBim , σBim ) and substrate (Es , σs ), to generate a stress-strain curve for the film, see e.g. [2],

Ef = 1Af[ABimEBim − AsEs]
  = 1 + tstfEBim − tstfEs    (1a)
σf = 1Af[P − Ps]
  = 1 + tstfσBim − tstfσs    (1b)
where P is the applied load, A = wt is the cross sectional area and sub-index “Bim” corresponds to the film-substrate bimaterial (ABim = w(ts +tf )). Figure 1 shows film stress (σ)-strain (ε) representative curves for Au films with different thicknesses extracted from the Au/PSF bimaterials. The film behavior presents only a small region of plasticity close to the ultimate strain. Thus, the numerical value of the maximum stress (strength) is close to its yield strength. The large plasticity of the substrate may hinder the plasticity of gold when acting as a bimaterial. As observed from this figure, the film modulus, strength and ultimate strain increase as the film thickness decreases, evidencing a “thickness-effect” not observed in bulk materials. Slightly different properties were obtained for the Au films deposited over the BAP substrate, which evidences some substrate-dependency of the film properties.

Commentary by Dr. Valentin Fuster
IMECE2008-67282 pp. 595-601; (7 pages)

The solidly mounted resonator (SMR) is one of the major focuses in filter research because it can be used in the frequency above GHz range. The reflection structure composed of periodic layers is vital to the performance of this type of resonator due to its capability in confining acoustic energy in the piezoelectric layer. Therefore the design of reflection layers is a key issue in the development of SMRs. The performance of reflection layers is revealed by the attenuation coefficient that governs the energy distribution in the periodic structures. The behavior of waves propagate in the finite periodic structures are solved by transfer matrix method while the Hill’s method is employed to find the exact solutions in the corresponding phononic crystal. By comparing their displacement fields, it is observed that the attenuation coefficients of infinite and finite periodic structures are almost identical provided the number of layers is adequate. Therefore referring the design of reflection layers to the band structures of the corresponding phononic crystals is reasonable although the attenuation coefficient of a finite periodic structure can not be calculated directly. For one dimensional phononic crystals, the attenuation coefficient becomes larger as the first band gap gets wider. Moreover, the characteristic impedance ratio and density ratio between two interlaced materials increase simultaneously; the first band gap width also increases. This character can be adopted as a guideline in the design of solidly mounted resonators. Based on this guideline, Al and W are chosen as materials for the reflection structure. By calculating its electric impedance, the resonant frequency is found to be the same as the center frequency of first band gap of the corresponding phononic crystal. It shows that employing this stop band character to design the reflection structure of SMR is adequate and efficient.

Topics: Reflection
Commentary by Dr. Valentin Fuster
IMECE2008-68100 pp. 603-609; (7 pages)

Evaluating state of damage in a ductile material as it experiences mechanical fatigue and cyclic loading poses much complexity and has been the subject of many researches. This study revisits the anisotropic damage model developed by Lemaitre (1992) and proposes to use his model combined with a micro-mechanics and mechanism based damage evolution model (Energy Partitioning Damage Evolution (EPDE)) and also a Unified Creep Plasticity-based model to predict the state of damage. The model is examined for pure shear and is applied to Pb-free solder materials. New anisotropic damage model exponents are generated using experimental data for Pb-free solder for both EPDE and UPC-based models and are compared with exponents generated previously under the assumption of isotropic and homogenous damage evolution.

Commentary by Dr. Valentin Fuster
IMECE2008-66624 pp. 611-620; (10 pages)

The Lattice-Boltzmann Method (LBM) has been used for investigating flow behavior and characteristics in mini, micro and nano channels with the objective of describing the transition among different length scales. In particular, we have used the LBM to describe the air bearing lubrication problem at very small scales. For doing this, first we simulate and characterize the Poiseuille flow through different length scale and compare the LBM numerical results to existing experimental and numerical results. We put special attention on the application of the slip boundary condition on the channel wall for very small length scales. Our numerical results for the Poiseuille flow show an acceptable agreement with the Fukui & Kaneko numerical solution for continuous and slip-velocity regimes. For both, the rarified flow regime and the free molecular flow regime our solutions do not show an acceptable agreement with the Fukui & Kaneko Model. Then, we focus on the Couette flow characterization at very small length scales. The pressure distribution on both walls for different Knudsen numbers is obtained and compared to existing numerical results. Last, we concentrate in the air bearing problem. We have looked at the best simulation parameters for successfully describing this device flow dynamics, and particularly, for determining the pressure distribution and the net force with a good accuracy.

Commentary by Dr. Valentin Fuster
IMECE2008-66993 pp. 621-630; (10 pages)

A novel microscale device based on phase change technology is proposed for localized cooling of very high heat flux electronic substrates. Device physics and salient design features are elucidated and a numerical model is developed to predict device performance and optimize the design parameters. The Micro-columnated Loop Heat Pipe (MLHP) consists of evaporator and condenser sections connected by liquid and vapor microchannels, all fabricated on a silicon wafer using MEMS microfabrication techniques. A novel micro-columnated wick is designed to prevent wick dryout to enhance reliability. The MLHP operates using capillary pumping thus requiring no external power and provides enhanced localized cooling capabilities by interfacing directly with a heat producing chip. Recent studies on microscale fluid flow and heat transfer physics in microchannels are implemented in a numerical model to predict the performance of the MLHP. A detailed design optimization study is performed to maximize the device heat flux carrying capacity, which at 1400 W/cm2 is a considerable improvent on all of the existing high heat flux cooling technologies.

Topics: Cooling , Heat flux
Commentary by Dr. Valentin Fuster
IMECE2008-67283 pp. 631-636; (6 pages)

A device able to pump a fluid with no moving mechanical parts represents a very encouraging alternative since such device would be practically maintenance free. A magnetocaloric pump could achieve this purpose by providing a magnetic pressure gradient to a ferrofluid placed inside a magnetic field while experiencing a temperature change. If the temperature change is produced by extracting heat out of an element that needs refrigeration, coupling this generated heat with the magnetocaloric pump will result in a passive cooling system. For applications near ambient temperature the ferrofluid must have specific characteristics such as low “Curie temperature”, high pyromagnetic coefficient, high thermal conductivity and low viscosity. This work presents an analysis of the ferrohydrodynamic governing equations, emphasizing the importance of the Kelvin force in the magnetocaloric pump analysis. The general equations are simplified and scaled to show which parameters are important in the generation of the magnetic pressure gradient. Based on the scaling analysis, a variable magnetic field and a higher saturation magnetization is needed to generate a higher magnetic pressure gradient. The working fluid used is an aqueous Mn0.5 Zn0.5 Fe2 O4 ferrite ferrofluid synthesized by the co-precipitation technique. This ferrite shows lower “Curie temperature” than commercially available magnetite. Important issues in the design of a magnetocaloric pump prototype with a variable magnetic field source are also discussed.

Commentary by Dr. Valentin Fuster
IMECE2008-67808 pp. 637-645; (9 pages)

We study the wetting behavior of water droplets on superhydrophobic arrays of lithographically fabricated square posts. To determine the droplet wetting state, we measure static contact angles and compare the results to predictions for equilibrium Cassie and Wenzel states. Surprisingly, we find that roll-off angles are minimized on surfaces expected to induce Wenzel-like wetting in equilibrium. We argue that droplets on these surfaces are metastable Cassie droplets whose internal Laplace pressure is insufficient to overcome the energy barrier required to completely wet the posts. These metastable Cassie droplets show superior roll-off properties because the effective length of the contact line that is pinned to the surface is reduced. We develop a model that can predict the transition between the metastable Cassie and Wenzel regimes by comparing the Laplace pressure of the drop to the capillary pressure associated with the wetting energy barrier of the textured surface. In the case of impacting droplets the water hammer and Bernoulli pressures must be compared with the capillary pressure. Experiments with impacting droplets show very good agreement with this simple pressure-balance model. Together these models can be used to optimize texture design for droplet-shedding and droplet-impact resistant surfaces.

Commentary by Dr. Valentin Fuster
IMECE2008-67932 pp. 647-653; (7 pages)

This paper presents a study of fluid flow through microchannel. Based on the work of Senta and Nnanna, [23], a trapezoidal-shaped manifold is used to ensure uniform flow distribution in the microchannel. Analysis further shows that flow uniformity among the channels largely depends on shape of the manifolds, length and location of inlet and outlets, and the inlet flow rate. The test setup consists of one hundred twenty-six 14.5μm-width channels, flow loop, heat source, thermal sensors and pressure transducers. Flow of fluid through the channels is regulated using a peristaltic pump. Experiments were conducted from various flow rates and heat loads. According to experimental data, microchannel has significant impact in the heat transfer rate for all the flow rates considered. This enhancement could be attributed to laminar flow in the microchannels, conduction heat transfer through the walls of the channel, fluid-channel wall interaction, and microconvection within the channel. Results show raises some concerns on the use of empirical correlations for flow between two parallel plates to predict heat transfer behavior in microchannels. In the absence of experimental data, f ≈ −2(dp/dx)dh um 2 provides a reasonable estimate of friction factor in microchannel.

Commentary by Dr. Valentin Fuster
IMECE2008-68111 pp. 655-660; (6 pages)

A precise control of orientation and position of micro- and nano-particles is critical for their applications in micro- and nano-scale systems. This research is focused on a numerical investigation of controlling the cylindrical particle/road with transverse electric field in a pressure driven microchannel flow with Finite Element Method (FEM). The particle/rod is initially located at the channel centerline with different initial angles. The electrophoretic force and hydrodynamic force are combined to control particle/rod movement. It was found that pressure, applied voltage and initial angles have different impact on particles dynamic behavior.

Commentary by Dr. Valentin Fuster
IMECE2008-68170 pp. 661-668; (8 pages)

Based on the rarefied flow phenomenon of thermal creep (or thermal transpiration), the Knudsen Compressor is an unconventional micro/meso-scale compressor or pump. Optimization studies have shown that a Knudsen Compressor operates most efficiently when its membrane’s flow channels are at the transitional flow regime, between continuum and molecular flows; simultaneously it provides a desired mass flow and pressure ratio. At higher pressures (> 1 atm), to maintain membrane channel Knudsen numbers in the transitional regime (Kn ∼ 1), the corresponding membrane channel size needs to be less than about 50 nm. More specifically, at 10 atm, the membrane channel size should be as small as 5 nm to provide the most efficient Knudsen Compressor operation. Prior to this work, there has been no documented experimental investigation of thermal creep measurements through channels less than 5 nm. Phenomena that could be associated with such flows are briefly discussed, and possible selection criteria for thermal creep membranes are included in this study. Apparatus design is discussed. Experimental results are provided for thermal creep flows, within a single stage Knudsen Compressor with 4 nm diameter membrane channels. The maximum pressure increases across the Knudsen Compressor’s thermal creep membrane were measured, over a range of operating pressures from 1 atm to 1.1 atm with Helium or Argon as the working gas. Results showed apparent thermal creep effects across the porous glass membrane, and possibly significant force field effects within the nano-scale channels.

Commentary by Dr. Valentin Fuster
IMECE2008-68851 pp. 669-675; (7 pages)

A number of microfluidic applications require precise thermal control where in-channel temperature measurements are necessary during the prototyping stage. Rhodamine B based laser induced fluorescence is a common technique used to obtain high resolution measurement of the fluid temperature field. However, PDMS has a tendency to absorb small hydrophobic dyes, such as Rhodamine B, which results in a steady increase in the overall fluorescent signal. This increase in light intensity causes a significant problem that must be overcome to obtain reliable temperature measurements. In this work a simple technique is described to remove the fluorescent signal originating from absorbed Rhodamine B dye particles that does not require surface modification or any significant alterations to the experimental setup. Instead a high power light source is used to photobleach the particles prior to taking images for thermometry analysis. Herein we demonstrate the technique with a conventional fluorescence microscope and a 100W mercury arc lamp and study the temperature field at the intersection of a Y-channel PDMS/glass chip where hot and cold streams merge.

Commentary by Dr. Valentin Fuster
IMECE2008-69183 pp. 677-686; (10 pages)

This theoretical investigation analyzed the three-dimensional momentum and mass transfer characteristics arising from the multiple inlets and single outlet in micro chamber which consists of a right square prism, an octagonal prism or a cylinder. The effects of various geometric parameters, inlet velocities, and the types of lamination on the mixing characteristics were investigated, and the results were presented in terms of flow fields, concentration profiles, and mixing index. Numerical results indicated that vortex flow and numbers of inlets dominate the mixing index. At larger Taylor number, more inertia caused the powerful vortex flow in the chamber, and the damping effect on diffusion was diminished, which then increased the mixing performance. Furthermore, passive micromixers utilizing hybrid laminations showed better mixing results than those with parallel laminations.

Topics: Vortices , Laminations
Commentary by Dr. Valentin Fuster
IMECE2008-66441 pp. 687-696; (10 pages)

Thermal cycling of chalcogenide materials (Ge2 Sb2 Te5 or GeSb for example) causes switching between two electrical resistance levels in the materials. This is the basis of PC memory, and offers the possibility of use for programmable signal switching in electronic systems as well. Here we propose a design for connection topology, using dual tip AFM-type probes. The design subdivides a single phase change via into a parallel array of three-terminal sub-vias which are well-suited to addressing with probes. This sub-division reduces required power and current to acceptable levels. Experimental inputs to the model were extracted from two sources. First, current levels were limited to levels that have previously been shown possible to deliver with AFM tips. Secondly, measurements of PC resistance as a function of cooling time were used to determine required heat sinking of the sub-via structures.

Topics: Design , Probes
Commentary by Dr. Valentin Fuster
IMECE2008-66520 pp. 697-701; (5 pages)

The release of microstructures from a Si substrate depends directly on the underetching (isotropic) characteristics of the etchant used. For the purpose of this study, XeF2 gas was selected as the etchant medium. Etching by XeF2 is primarily a function of pressure, which determines the rate of interaction between the Si surface and the etchant gas. However, other factors play a large role in XeF2 etching characteristics. Testing was conducted to determine the etch rate and profile of XeF2 etching when various parameters of the structure design are changed (Si exposure area, size and dimension of structures, spacing of structures). The gas was introduced at a pressure cycle of approximately 4.0 Torr for 180 s. Uniformity of the etched surface is improved by increasing the number of etch cycles, or by increasing the gas pressure of the surrounding XeF2 .

Commentary by Dr. Valentin Fuster
IMECE2008-66867 pp. 703-711; (9 pages)

If we define the slowness curve by the normal vector of the material point of the crystal surface divided by the growth (or etching) rate of that point. Frank’s theorem states that, in the course of crystal’s growing or etching, each point of the boundary surface moves along its own straight line (or the characteristic line) which is parallel to the normal of the slowness curve. Since he didn’t find out the speed of the crystal surface point advancing along this characteristic, the etched crystal shape can not be determined quantitatively. In this paper we derive the equation in explicit form expressing the surface moving speed along the characteristics in terms of the microscopic parameters (in atomic size) and variables such as step density, step flux, and step height. Not all the microscopic variables are measurable, so we have to drive certain equations relating the microscopic variables to the macroscopic ones (such as the orientations of the crystal boundary lattice plane, slopes of the characteristic lines, etching rates, and so on) through the kinematics theory of particles. One measurement of the macroscopic variables at any particular instant is enough to determine not only the etching (or growth) rates of the crystal but also the etched crystal shape of any subsequent instant.

Topics: Crystals , Etching
Commentary by Dr. Valentin Fuster
IMECE2008-66870 pp. 713-721; (9 pages)

Based on the maximum CHF (critical heat flux) criterion, an optimal heat transfer criterion, which is called H criterion, was proposed. Experimental apparatuses were conducted. Distilled water was used as the working fluid. Three different DANFOSS nozzles with cone angles being 54°, 50° and 54° respectively were used. A 30×30mm2 square copper surface was used as the heated surface. Experimental results indicated that the volumetric fluxes were proportioned to P0.5 , where P is the pressure drop across the nozzles. The optimal distance between the nozzles and the heated surface were derived. The results indicated that the optimal heat transfer appeared while the outside of the impellent thin spray film inscribed in the square heated surface. Based on the H criterion aforementioned, two DANFOSS nozzles of the three, with cone angles being 54° and 50° respectively, were used to study the temperature distribution of the heated surface while there were spray inclination angles during spray cooling experiments. Distilled water was also used impacting on the 30×30mm2 square copper surface aforementioned and a circular heated copper surface with diameters being 30mm respectively. The heat flux of the surface was kept in constant (about 26–35W/cm2 ). The inclination angles were 0°, 10°, 20°, 30°, 40° and 50° respectively. Three thermocouples imbedded in the heated surface were used to predict the grads of the temperature of the surface. Experimental results indicated that the temperature and the grads of the temperature of the surface increases first and then decreases with the increase of the inclination angle.

Topics: Cooling , Boiling , Sprays
Commentary by Dr. Valentin Fuster
IMECE2008-66909 pp. 723-730; (8 pages)

In flex-tensional piezoactuators, due to the low displacement of piezostacks, a compliant mechanism is used to amplify displacement of piezostack. In this paper, optimization of a compliant mechanism with corner-filleted flexure hinges is carried out using real-coded genetic algorithms (GAs) to avoid trapping in local optimums. The objective functions are displacement amplification and stiffness of mechanism and design variables are cross-sectional size and material used. The constraints which are applied on mechanism are based on piezostack dimensions and manufacturing limits. Displacement amplification and stiffness are calculated using strain energy and Castigliano’s displacement theorem.

Commentary by Dr. Valentin Fuster
IMECE2008-66946 pp. 731-740; (10 pages)

An optical micro-coupling system of whispering-gallery mode usually consists of a resonator (e.g. a sphere) and a coupler (e.g. a taper). In this report, silica microspheres of 50–500 μm in diameter are fabricated by hydrogen flame fusing of an end of a single mode fiber or fiber taper. Fiber tapers are fabricated by the method of heating and pulling that meets an adiabatic condition. Taper’s waist diameter can routinely be made less than 1 μm and almost zero transmission loss in a taper is achieved which allows an effective and phase-matched coupling for a wide range sizes of microspheres. Both resonators and couplers’ surface microstructure and shapes are examined by scanning electronic microscopy. Three regimes of coupling are achieved, enabling a good flexibility to control Q value and coupling efficiency of a micro-coupling system. Whispering gallery mode shift is used to demonstrate a novel temperature micro-sensor. Its sensitivity determined from actual experimental results agrees well with the theoretical value. A concept of using the photon’s cavity ring down (CRD) in the microsphere to make a novel high-sensitivity trace gas micro-sensor is proposed. The CRD time constant when ammonia is chosen as the analyte gas is predicted using the simulated absorption lines.

Commentary by Dr. Valentin Fuster
IMECE2008-67031 pp. 741-747; (7 pages)

This research focuses on designing, fabricating, and testing a strain sensor for different soil applications. Using finite element modeling and analysis, the initial dimensions of the diaphragm were designed, and a diaphragm diameter of 1.35 cm with a thickness of 300 μm was chosen. The fabrication process of the sensor prototype is discussed in this paper along with the design for specific test benches to test the electrical and mechanical characteristics at different stages of fabrication. Displacement tests on the sensor diaphragm were performed and the corresponding voltages produced were tabulated. A maximum displacement of 250 μm was achieved producing a maximum voltage of 7.3 mV. The voltage produced by the sensor was recorded using LabVIEW, and its values were tabulated and plotted against corresponding displacement and strain magnitudes.

Topics: Design , Soil , Strain sensors
Commentary by Dr. Valentin Fuster
IMECE2008-67245 pp. 749-753; (5 pages)

This paper describes a novel method of nanowire assembly using a superhydrophobic surface as a template. Well-defined superhydrophobic structures on a template surface direct the site-specific self-assembly of nanowires due to interfacial tension in evaporation, enabling simple but highly-efficient and ordered assembly of nanowires. High-aspect-ratio (HAR) microstructures with tapered tips are fabricated by deep reactive ion etch (DRIE) and are coated with a self-assembled monolayer (SAM) of octadecyltrichlorosilane (OTS) for hydrophobicity. Nickel nanowires are fabricated in a porous alumina membrane by electrodeposition. A uniformly-dispersed nanowire suspension is dispensed and evaporated on the superhydrophobic template surface. Due to surface superhydrophobicity, a three-phase (i.e., liquid-solid-gas) interface is created on the surface structures, enabling the nanowires to reside only over the interface. After complete evaporation, the nanowires are mostly left on the structural tips, driven by the interfacial forces constituted at the three-phase boundary. Although the alignment yield rate of the nanowires to the surface pattern has not reached 100%, current experimental results demonstrate that the idea of using interfacial tension on superhydrophobic surfaces can serve as a novel nano-assembly technique with high throughput and high rate. The key parameters affecting the yield of self-assembly and alignment will continue to be studied for further improvement.

Commentary by Dr. Valentin Fuster
IMECE2008-67299 pp. 755-759; (5 pages)

A new droplet microgripper that can pick up and release micro objects by electrowetting is presented. Electrowetting is utilized to dynamically change the contact angle between the liquid bridge and the gripper surface to control the capillary lifting forces. A coplanar interdigitated gold electrode pair was employed to simplify the microfabrication. The lifting force generated by the microgripper was experimentally characterized. The pick and release routine were successfully demonstrated for various micro glass beads. The design, microfabrication and testing of the microgripper are presented.

Commentary by Dr. Valentin Fuster
IMECE2008-67526 pp. 761-764; (4 pages)

Multi-wall carbon nanotube (MWNT)-based bimorph nanoactuators were synthesized and characterized. A thin metal film was deposited on the sidewall of MWNTs using a pulsed laser deposition method to form a bimorph nanostructure. The preliminary actuation was demonstrated using a fabricated bimorph MWNT. Thermal bending performance of the nanoactuator on a thermal stage will be further investigated by using a scanning electron microscope.

Commentary by Dr. Valentin Fuster
IMECE2008-67594 pp. 765-769; (5 pages)

We propose novel biomimetic polymer microactuators. The actuation mechanism is inspired by nastic movement of the moving plant, Mimosa pudica, which folds its leaves upon external stimulus by regulating turgor pressure of cells in specific location. Photo-cured poly(ethylene glycol) diacrylate (PEGDA) microactuator is fabricated using projection micro-stereolithography (PμSL) capable of complex 3D micro fabrication. The swelling effect of PEG in water and organic solvent is exploited as an actuation mechanism of the device. Stress relaxation in the structure due to solvent absorption is controlled locally by delivering solvent through microfluidic channels embedded in the actuator, thereby generating a net movement in the device. Timescale of the motion derived from analytical swelling model suggests that actuation speed can be effectively increased by scaling down the actuator because the characteristic swelling time depends on the length as L2 , which is verified experimentally.

Commentary by Dr. Valentin Fuster
IMECE2008-67844 pp. 771-779; (9 pages)

The design, fabrication, and characterization of a surface micromachined, front-vented, 64 channel (8×8), capacitively sensed pressure sensor array is described. The array was fabricated using the MEMSCAP PolyMUMPs® process, a three layer polysilicon surface micromachining process. An acoustic lumped element circuit model was used to design the system. The results of our computations for the design, including mechanical components, environmental loading, fluid damping, and other acoustic elements are detailed. Theory predicts single element sensitivity of 1 mV/Pa at the gain stage output in the 400–40,000 Hz band. A laser Doppler velocimetry (LDV) system has been used to map the spatial motion of the elements in response to electrostatic excitation. A strong resonance appears at 480 kHz for electrostatic excitation, in good agreement with mathematical models. Static stiffness measured electrostatically using an interferometer is 0.1 nm/V2 , similar to the expected stiffness. Preliminary acoustic sensitivity studies show single element acoustic sensitivity (as a function of frequency) increasing from 0.01 mV/Pa at 200 Hz to 0.16 mV/Pa at 2 kHz. A more in depth analysis of acoustic sensitivity is ongoing.

Commentary by Dr. Valentin Fuster
IMECE2008-67865 pp. 781-784; (4 pages)

The controlled assembly of nanowires is essential for nanoscale processes. The dielectrophoretic (DEP) assembly process enables a very simple and efficient assembly; however, controlling the number and dimension of nanowires to bridge electrodes is extremely intricate. The micromachined nanowire diluter presented in this paper can automatically dilute and sort nanowires in solution without requiring conventional centrifuge equipment. The device consists of a glass substrate with an array of gold electrode pairs and a PDMS microchannel. Nickel nanowires (30 μm-long) were fabricated by a template-directed electrodeposition process using nanoporous alumina templates. A liquid solution containing nanowires was injected into an inlet of the diluter. Pulsed voltages were applied to 16 pairs of electrodes. The nanowires were subsequently trapped or released in the microchannel at specific pulsed electric fields. As a result, the number of nanowires at the outlet of the channel was dramatically reduced, implying that the device presented here can effectively dilute nanowire suspensions for controlled assembly.

Commentary by Dr. Valentin Fuster
IMECE2008-67880 pp. 785-790; (6 pages)

Ratcheted structures with superhydrophobicity have potential to drive motion of liquid drops in microfluidic devices. We report development of a new process to integrate microscale pillars with varied periods of ratchets from sub-micrometer to 1.5 mm in SU-8. For the fabrication, an hierarchical mixed process of thermal imprint lithography and optical UV-lithography was employed. Use of Ni ratchets as stamps and a hydrophobic silane coating on the stamp surfaces significantly improved imprinting performance into SU-8. We have successfully demonstrated fabrication of ratcheted micropillars as well as submicrometer ratchets integrated into micropillars. In addition, we will discuss optimization of the SU-8 imprinting process and the air gap compensation for photolithography of ratcheted SU-8 layers, which are essential for the fabrication.

Commentary by Dr. Valentin Fuster
IMECE2008-67938 pp. 791-798; (8 pages)

This paper reports the findings of an investigation on application of secondary thermoelectric (TE) module as a heat exchanger for the primary TE module. The experimental system consists of two commercially available thermoelectric modules arranged thermally in series with a heat sink and an integrated circuit (IC) chip. Heat produced from the IC chip is transferred to the heat sink via the TE modules. A total of nine experimental setups were analyzed using measured temperature data to assess the efficacy of the setups. Experimental evidence shows that the secondary TE module provides additional cooling advantage. The cooling capacity for a system with secondary TE module is 10.95W compared to 3.5W for systems where secondary modules are non-existent. The respective coefficient of performance, COP = Qc /Qp are 2.43 and 0.78. The use of a secondary TE module as a heat exchanger for the primary thermoelectric module is ineffective when compared with liquid-cooled heat exchanger. Results further showed that during early stages of heating and cooling processes, there exists lag in response time between the integrated circuits chip. This could result in over-heating or under-cooling the IC chip.

Commentary by Dr. Valentin Fuster
IMECE2008-68078 pp. 799-802; (4 pages)

In this letter, based on the beam theory and the thermal analysis of a bi-material cantilever, we demonstrate that the effective thermal conductance of the cantilever and the temperature at the tip of the cantilever can be determined by measuring the bending of the cantilever in response to two different thermal inputs: power absorbed at the tip and ambient temperature. The bi-material cantilevers were first introduced as a calorimeter to measure the heat generated in chemical reactions. [1] The same device was demonstrated to be sensitive enough to measure power as small as 100 pW or energy of 150 fJ in photothermal measurements. [2] They were also used as IR detectors [3, 4, 5] or as scanning thermal imaging probes. [6] Although the bi-material cantilevers are often used as temperature or heat flux sensors based on the beam bending due to the unequal thermal expansion of the two materials, the exact temperature at the tip of the cantilever is usually unknown. Directly measuring the temperature is difficult due to the small geometry of the cantilever structure. To find out the temperature of the cantilever, one should obtain the thermal conductance of the cantilever. However, since the thermal properties of two layers of the cantilever are dependent on their thickness, one cannot rely on theoretical calculation. In this letter, we develop a technique to determine the thermal conductance of the cantilever by measuring the bending of the cantilever in response to the variations of the absorbed power at the tip and the ambient temperature. A triangular silicon nitride cantilever coated with 70 nm gold film is used in the current experiment. As shown in Fig.1 (a), a semiconductor laser beam is focused on the tip of the cantilever and reflected onto a position sensing detector (PSD). The deflection of the reflected laser beam spot on the PSD is used as a measure of the deflection of the cantilever. A part of the laser power is absorbed by the cantilever and thus creates a temperature rise at the end of the cantilever. The output of the PSD is converted into an X or Y signal corresponding to the position of the laser spot on the PSD and a sum signal proportional to the incident laser power.

Commentary by Dr. Valentin Fuster
IMECE2008-68143 pp. 803-809; (7 pages)

Validation of a tolerance analysis for the assembly of modular, polymer microfluidic devices was performed using simulations and experiments. A set of three v-groove and hemisphere-tipped post joints was adopted as a model assembly features. An assembly function with assembly feature dimensions and locations was modeled kinematically. Monte Carlo methods were applied to the assembly function to simulate variation of the assembly. Assembly accuracy was evaluated assuming that the variations of the assembly features were randomly distributed. The estimated mismatches were 118 ± 30 μm and 19 ± 13 μm along the X- and Y-axes, respectively. The estimated vertical gap between the modules at the alignment standards along the X- and Y-axes 312 ± 37 μm and 313 ± 37 μm, compared to the designed value of 287 μm. To validate the tolerance model, two micromilled br