A three phase serpentine micro electrode array for AC electroosmotic flow pumping

A new three-phase electrode array with a serpentine electrode is designed and prototyped using PolyMUMPs process for micro flow pumping. Numerical model of the micropump has been developed using COMSOL Multiphysics™. Experimental testing is conducted and time-averaged flow velocities from testing and simulation agree well. Peak time-averaged flow velocity of 270 μm/s is achieved at 30 Hz using ethanol.     

A micro electromagnetic low level vibration energy harvester based on MEMS technology

This paper presents a micro electromagnetic energy harvester which can convert low level vibration energy to electrical power. It mainly consists of an electroplated copper planar spring, a permanent magnet and a copper planar coil with high aspect ratio. Mechanical simulation shows that the natural frequency of the magnet-spring system is 94.5 Hz. The resonant vibration amplitude of the magnet is 259.1 μm when the input vibration amplitude is 14 μm and the magnet-spring system is at resonance. Electromagnetic simulation shows that the linewidth and the turns of the coil influence the induced voltage greatly. The optimized electromagnetic vibration energy harvester can generate 0.7 μW of maximal output power with peak–peak voltage of 42.6 mV in an input vibration frequency of 94.5 Hz and input acceleration of 4.94 m/s² (this vibration is a kind of low level ambient vibration). A prototype (not optimized) has been fabricated using MEMS micromachining technology. The testing results show that the prototype can generate induced voltage (peak–peak) of 18 mV and output power of 0.61 μW for 14.9 m/s² external acceleration at its resonant frequency of 55 Hz (this vibration is not in a low ambient vibration level).     

Design, implementation and testing of electrostatic SOI MUMPs based microgripper

This paper presents a detail modeling, finite element analysis and testing results of MEMS based electrostatically actuated microgripper. Interdigitated lateral comb pairs have been used to actuate the microgripper. The microgripper is optimized using standard SOI-MUMPs technology in L-Edit of MEMS-Pro with dual jaws actuation at low voltages. Coupled electromechanical finite element analysis performed in COVENTOR-WARE shows total displacement of 15.5 μm at jaws tip at 50 V, which is quite comparable to experimental result of 17 μm displacement at the tip of gripper jaw for the same voltage. Micromanipulation experiments have successfully demonstrated the gripping, holding micro-objects between 53 and 70 μm in size. The simulated model is used to study detail profile of Von Mises stresses and deformations in the model. It is noted that maximum stress in microgripper is 200 MPa which is much smaller than yield stress of 7 GPa. The slight difference between finite element analysis and experimental results is because of small variations in process material parameters. The total size of gripper is 5.03 × 6.5 mm².     

High-current density DC magenetohydrodynamics micropump with bubble isolation and release system

One of the major challenges for integrated Lab-on-a-Chip (LOC) systems is the precise control of fluid flow in a micro-flow cell. Magnetohydrodynamics (MHD) micropumps which contain no moving parts and capable of generating a continuous flow in any ionic fluid offer an ideal solution for biological applications. MHD micropumping has been demonstrated by using both AC and direct current (DC) currents by a number of researchers with varying degrees of success. However, current MHD designs based on DC do not meet the flow rate requirements for fully automated LOC applications (>100 μl/min). In this research, we introduce a novel DC-based MHD micropump which effectively increases flow rate by limiting the effects of electrolysis generated bubbles at the electrode–electrolyte interface through isolation and a mechanism for their release. Gas bubbles, particularly, hydrogen generated by high current density at the electrodes are the main culprit in low experimental flow rate compared with theoretical values. These tiny bubbles coalesce in the flow channel thus obstructing the flow of fluid. Since hydrolysis is inevitable with DC excitation, compartmentalized electrode channels with bubble isolating and coalescence retarding mechanisms and bubble release systems are implemented to prevent the coalescence of these bubbles and minimize their effects on flow rate. In this novel design called bubble isolation and release system, flow rate of up to 325 μl/min is achieved using 1 M NaCl solution in DC mode with potentials of 5 V and current density of about 5,000 A/m² for a main channel of 800 μm × 800 μm cross-section and 6.4 mm length.     

Equivalent electrical network for performance characterization of piezoelectric peristaltic micropump

Utilizing an electronic–hydraulic analogy, this study develops an equivalent electrical network of a piezoelectric peristaltic micropump which has not been modeled the whole system operation completely by computational fluid dynamics (CFD) or equivalent electrical network so far due to its excessive complicated structure. The validity of the proposed model is verified by comparing the simulation results obtained using the SPICE (simulation program with integrated circuit emphasis) software package for flow rate spectrum and its maximum state of a typical micropump with the experimental observations for two working fluids, namely DI water and blood. The simulation results predict a maximum flow rate frequency and flow rate of 280 Hz and 43.23 μL/min, respectively, for water, and 210 Hz and 24.12 μL/min for blood. The corresponding experimental results are found to be 300 Hz and 41.58 μL/min for water and 250 Hz and 23.75 μL/min for blood. The relatively poorer agreement between the two sets of results when using blood as the working fluid is thought to be the result of the non-Newtonian nature of blood, which induces a more complex, non-linear flow behavior within the micropump. Having validated the proposed model, the equivalent network is used to perform a systematic analysis of the correlation between the principal micropump design parameters and operating conditions and the micropump performance. The results confirm the validity of the equivalent electrical network model as the first microfluidic modeling tool for optimizing the design of peristaltic micropumps and for predicting their performance.     

High-throughput electrokinetic bioparticle focusing based on a travelling-wave dielectrophoretic field

This study presents a sheathless and portable microfluidic chip that is capable of high-throughput focusing bioparticles based on 3D travelling-wave dielectrophoresis (twDEP). High-throughput focusing is achieved by sustaining a centralized twDEP field normal to the continuous through-flow direction. Two twDEP electrode arrays are formed from upper and lower walls of the microchannel and extend to its center, which induce twDEP forces to provide the lateral displacements in two directions for focusing the bioparticles. Bioparticles can be focused to the center of the microchannel effectively by twDEP conveyance when the characteristic time due to twDEP conveying in the y direction is shorter than the residence time of the particles within twDEP electrode array. Red blood cells can be effectively focused into a narrow particle stream (~10 μm) below a critical flow rate of 10 μl/min (linear flow velocity ~5 mm/s), when under a voltage of 14 V_p–p at a frequency of 500 kHz is applied. Approximately 90% focusing efficiency for red blood cells can be achieved within two 6-mm-long electrode arrays when the flow rate is below 12 μl/min. Blood cells and Candida cells were also focused and sorted successfully based on their different twDEP mobilities. Compared to conventional 3D-paired DEP focusing, velocity is enhanced nearly four folds of magnitude. 3D twDEP provides the lateral displacements of particles and long residence time for migrating particles in a high-speed continuous flow, which breaks through the limitation of many electrokinetic cell manipulation techniques.     

Particle streaming and separation using dielectrophoresis through discrete periodic microelectrode array

This study presents a particle manipulation and separation technique based on dielectrophoresis principle by employing an array of isosceles triangular microelectrodes on the bottom plate and a continuous electrode on the top plate. These electrodes generate non-uniform electric fields transversely across the microchannel. The particles within the flowing fluid experience a dielectrophoretic force perpendicular to the fluid flow direction due to the non-uniform electric fields. The isosceles triangular microelectrodes were designed to continuously exert a small dielectrophoretic force on the particles. Particles experiencing a larger dielectrophoretic force would move further in the perpendicular direction to the fluid flow as they traveled past each microelectrode. Polystyrene microspheres were used as the model particles, with particles of ∅20 μm employed for studying the basic characteristics of this technique. Particle separation was subsequently demonstrated on ∅10 and ∅15 μm microspheres. Using an applied sinusoidal voltage of 20 V_pp and frequency of 1 MHz, a mean separation distance of 0.765 mm between them was achieved at a flow rate of 3 μl/min (~1 mm/s), an important consideration for high throughput separation capability in a micro-scale technology device. This unique isosceles triangular microelectrodes design allows heterogeneous particle populations to be separated into multiple streams in a single continuous operation.     

Effect of dispersed phase viscosity on maximum droplet generation frequency in microchannel emulsification using asymmetric straight-through channels

Uniformly sized droplets of soybean oil, MCT (medium-chain fatty acid triglyceride) oil and n-tetradecane with a Sauter mean diameter of d _3,2 = 26–35 μm and a distribution span of 0.21–0.25 have been produced at high throughputs using a 24 × 24 mm silicon microchannel plate consisting of 23,348 asymmetric channels fabricated by photolithography and deep reactive ion etching. Each channel consisted of a 10-μm diameter straight-through micro-hole with a length of 70 μm and a 50 × 10 μm micro-slot with a depth of 30 μm at the outlet of each channel. The maximum dispersed phase flux for monodisperse emulsion generation increased with decreasing dispersed phase viscosity and ranged from over 120 L m⁻² h⁻¹ for soybean oil to 2,700 L m⁻² h⁻¹ for n-tetradecane. The droplet generation frequency showed significant channel to channel variations and increased with decreasing viscosity of the dispersed phase. For n-tetradecane, the maximum mean droplet generation frequency was 250 Hz per single active channel, corresponding to the overall throughput in the device of 3.2 million droplets per second. The proportion of active channels at high throughputs approached 100% for soybean oil and MCT oil, and 50% for n-tetradecane. The agreement between the experimental and CFD (Computational Fluid Dynamics) results was excellent for soybean oil and the poorest for n-tetradecane.     

Fabrication of three-dimensional hemispherical structures using photolithography

A photolithography technique using SU-8 and PDMS was developed to fabricate three-dimensional hemispherical structures. This technique utilized a mask-aligner and normal binary coded photomasks to generate hemispherical pits on SU-8, followed by PDMS molding to obtain an array of dome-shaped structures. Using this technique, a microfluidic device was fabricated with a patterning area that consisted of an array of 5 μm wells and dome-shaped structures with 10 μm diameter and 6 μm height. Encoded microbeads, 6 μm in size, were immobilized and patterned in the microfluidic device under flow conditions and a DNA hybridization experiment was performed to demonstrate the incorporation of encoded beads that would enable a high level of multiplexing in bioassays.     

Micro-electroforming metallic bipolar electrodes for mini-DMFC stacks

This paper describes the development of metallic bipolar plate fabrication using micro-electroforming process for mini-DMFC (direct methanol fuel cell) stacks. Ultraviolet (UV) lithography was used to define micro-fluidic channels using a photomask and exposure process. Micro-fluidic channels mold with 300 μm thick and 500 μm wide were firstly fabricated in a negative photoresist onto a stainless steel plate. Copper micro-electroforming was used to replicate the micro-fluidic channels mold. Following by sputtering silver (Ag) with 1.2 μm thick, the metallic bipolar plates were completed. The silver layer is used for corrosive resistance. The completed mini-DMFC stack is a 3.5 × 3.5 cm² fuel cell stack including a 1.5 × 1.5 cm² MEA (membrane electrode assembly). Several MEAs were assembly into mini-DMFC stacks using the completed metallic bipolar plates. All test results showed the metallic bipolar plates suitable for mini-DMFC stacks. The maximum output power density is 9.3 mW/cm² and current density is 100 mA/cm² when using 8 vol.% methanol as fuel and operated at temperature 30°C. The output power result is similar to other reports by using conventional graphite bipolar plates. However, conventional graphite bipolar plates have certain difficulty to be machined to such micro-fluidic channels. The proposed micro-electroforming metallic bipolar plates are feasible to miniaturize DMFC stacks for further portable 3C applications.     

Systematic linearisation of a microfluidic gradient network with unequal solution inlet viscosities demonstrated using glycerol

This paper presents a mathematical and experimental study of the effect of inlet concentration (and therefore viscosity) of glycerol solutions on the performance of a microfluidic network. This was achieved with analytical modelling, implemented in MATLAB, and optical measurement of the entire concentration distribution of the network. A mathematical proposal to improve the linearity of the outlet profile is also implemented and successfully verified experimentally. The concentration gradients of a two inlet–six outlet (2–6) microfluidic network device were obtained with inlet solutions of 10–40 wt% glycerol and flow rates of up to 5 μl/s per inlet. The mathematical model developed gave a good agreement with the experimental results obtained. ‘S’ shaped outlet profiles were obtained for the four glycerol cases studied and the closest results to the model were achieved at an optimised flow rate of 1μl/s for 10 wt% glycerol, 5 μl/s for both 20 and 30 wt% glycerol and 1.5 μl/s for 40 wt% glycerol. The linearity of the outlet profiles for the 20, 30 and 40 wt% inlet glycerol experiments were improved from R ² of 0.977, 0.946 and 0.966, respectively (before linearisation) to their new values of 0.997, 0.995 and 0.974, respectively (after the linearisation). This was performed by application of the mathematical model, at controlled inlet flow rate ratios of 0.77, 0.63 and 0.52 with respect to the viscous inlet, for 20, 30 and 40 wt% glycerol experiments, again with very good agreement of the outlet performance between the experimental and the mathematical results.     

Integrated SWCNT sensors in micro-wind tunnel for air-flow shear-stress measurement

We have developed SWCNT sensors for air-flow shear-stress measurement inside a polymethylmethacrylate (PMMA) “micro-wind tunnel” chip. An array of sensors is fabricated by using dielectrophoretic (DEP) technique to manipulate bundled single-walled carbon nanotubes (SWCNTs) across the gold microelectrodes on a PMMA substrate. The sensors are then integrated in a PMMA micro-wind tunnel, which is fabricated by SU-8 molding/hot-embossing technique. Since the sensors detect air flow by thermal transfer principle, we have first examined the I–V characteristics of the sensors and confirmed that self-heating effect occurs when the input voltage is above ~1 V. We then performed the flow sensing experiment on the sensors using constant temperature (CT) configuration with input power of ~230 μW. The voltage output of the sensors increases with the increasing flow rate in the micro-wind tunnel and the detectable volumetric flow is in the order of 1 × 10⁻⁵m³/s. We also found that the activation power of the sensors has a linear relation with 1/3 exponential power of the shear stress which is similar to conventional hot-wire and polysilicon types of convection-based shear-stress sensors. Moreover, measurements of sensors with different overheat ratios were compared, and results showed that sensor is more sensitive to the flow with a higher overheat ratio.     

Design, fabrication and testing of a high performance silicon piezoresistive Z-axis accelerometer with proof mass-edge-aligned-flexures

This paper presents design, fabrication and testing of a quad beam silicon piezoresistive Z-axis accelerometer with very low cross-axis sensitivity. The accelerometer device proposed in the present work consists of a thick proof mass supported by four thin beams (also called as flexures) that are connected to an outer supporting rim. Cross-axis sensitivity in piezoresistive accelerometers is an important issue particularly for high performance applications. In the present study, low cross-axis sensitivity is achieved by improving the device stability by placing the four flexures in line with the proof mass edges. Various modules of a finite element method based software called CoventorWare^™ was used for design optimization. Based on the simulation results, a flexure thickness of 30 μm and a diffused resistor doping concentration of 5 × 10¹⁸ atoms/cm³ were fixed to achieve a high prime-axis sensitivity of 122 μV/Vg, low cross-axis sensitivity of 27 ppm and a relatively higher bandwidth of 2.89 kHz. The designed accelerometer was realized by a complementary metal oxide semiconductor compatible bulk micromachining process using a dual doped tetra methyl ammonium hydroxide etching solution. The fabricated accelerometer devices were tested up to 13 g static acceleration using a rate table. Test results of fabricated devices with 30 μm flexure thickness show an average prime axis sensitivity of 111 μV/Vg with very low cross-axis sensitivities of 0.652 and 0.688 μV/Vg along X-axis and Y-axis, respectively.     

Dynamic simulation of a contact-enhanced MEMS inertial switch in Simulink?

A novel contact-enhanced design of MEMS (micro-electro-mechanical system) inertial switch was proposed and modeled in Simulink^?. The contact effect is improved by an easily realized modification on the traditional design, i.e. introducing a movable contact point between the movable electrode (proof mass) and the stationary electrode, therefore forming a dual mass-spring system. The focus of this paper is limited to a vertically driven unidirectional one for the purposes of demonstration, but this design concept and Simulink^? model is universal for various kinds of inertial micro-switches. The dynamic simulation confirmed the contact-enhancing mechanism, showing that the switch-on time can be prolonged for the dynamic shock acceleration and the bouncing effect can be reduced for the quasi-static acceleration. The threshold acceleration of the inertial switch is determined by the proof mass-spring system’s natural frequency. Since the inertial switches were fabricated by the multilayer electroplating technology, the proof mass thickness were assigned two values, 100 and 50 μm, in order to get threshold levels of 56 and 133 g respectively for the dynamic acceleration of half-sine wave with 1 ms duration. Other factors that influence the dynamic response, such as the squeeze film damping and the contact point-spring system’s natural frequency were also discussed. The fabricated devices were characterized by the drop hammer experiment, and the results were in agreement with the simulation predictions. The switch-on time was prolonged to over 50 μs from the traditional design’s 10 μs, and could reach as long as 120 μs. Finally, alternative device configurations of the contact-enhancing mechanism were presented, including a laterally driven bidirectional inertial switch and a multidirectional one.     

A PZT-driven atomizer based on a vibrating flexible membrane and a micro-machined trumpet-shaped nozzle array

This paper presents the design, and fabrication of a PZT-driven atomizer based on a flexible membrane and a micro-machined trumpet-shaped nozzle array. Tests were conducted to demonstrate that the developed atomizer can produce fine droplets. The atomizer uses a PZT bimorph plate attached to a liquid-proof HDPE membrane with a low Young’s modulus to generate a pressure wave in the liquid reservoir. The trumpet-shaped micro-nozzle array is fabricated using a surface micromachining technique and an electroplating process. The fabrication process allows the use of a low resolution photomask to fabricate a high feature-sized trumpet-nozzle array. The SMD values of the ejected droplets and the flow rate of the fabricated atomizer are measured experimentally as a function of the operating frequency and the nozzle diameter for liquids of various viscosities. The relationship between the droplet size distribution and the SMD value is also explored. The experimental results show that the atomizer is capable of generating droplets with an SMD of 4.6 μ at a flow rate of 2.5 g/min. Hence, the atomizer has the potential for use in many applications.     

Development of a solder bump technique for contacting a three-dimensional multi electrode array

 The application of a solder bump technique for contacting a three-dimensional multi electrode array is presented. Solder bumping (or C4: Controlled Collapse Chip Connections, also called Flip Chip contacting) is the most suitable contacting technique available for small dimensions and large numbers of connections. Techniques adapted from the literature could successfully be scaled down to be used for 55x55 μm pads at 120 μm heart-to-heart spacing, yielding well-conducting, reasonably strong bonds. Received: 30 October 1995/Accepted: 16 September 1996     

Use of a PDMS spring element for ink jet printing applications

Results of the design, microfabrication and testing of a proof-of-concept, diaphragm-type silicone sealing joint are presented. DRIE-etched cavities were filled with a flexible sealing element made of polydimethylsiloxane that supports a silicon piston. A series of sealing joints were produced with variable widths, and the displacement of the piston was measured after applying pressures of up to 1 bar above atmospheric pressure in 0.2 bar increments. Two masks were designed to produce several sets of silicone springs with widths of 2–10.5 μm, each consisting of a 10 μm thick silicon piston that is 2 mm long. Tests performed on the shear spring joints were found to give a displacement of 0.5 μm at 1 bar when the sealing width is 6 μm or more. The sealing joint with a 10 μm width was found to give a displacement of 0.9 μm and an elastic recovery of 88%. The results showed this type of joint in the form of an elastically-deforming seal provides sufficient displacement for propelling liquid droplets as part of a liquid propulsion system.     

Low voltage electroosmotic pump for high density integration into microfabricated fluidic systems

A low voltage electroosmotic (eo) pump suitable for high density integration into microfabricated fluidic systems has been developed. The high density integration of the eo pump required a small footprint as well as a specific on-chip design to ventilate the electrolyzed gases emerging at the platinum (Pt) electrodes. For this purpose, a novel liquid–gas (lg) separator was invented. This lg-separator separated the gas bubbles from the liquid and guided them away from the eo pump. Its operational principle was solely based on the geometry of tapered sidewalls. An eo pump sandwiched by two lg separators (microchannels in the range of 10 μm, footprint of 100 μm × 15 μm) was experimentally investigated. The lg-separator was able to reliably separate and ventilate an emerging gas flow of 2 pl s⁻¹. The eo pump achieved flow rates of 50 pl s⁻¹ at actuation voltages of 5 V.     

Gas–liquid dynamics at low Reynolds numbers in pillared rectangular micro channels

Most heterogeneously catalyzed gas–liquid reactions in micro channels are chemically/kinetically limited because of the high gas–liquid and liquid–solid mass transfer rates that can be achieved. This motivates the design of systems with a larger surface area, which can be expected to offer higher reaction rates per unit volume of reactor. This increase in surface area can be realized by using structured micro channels. In this work, rectangular micro channels containing round pillars of 3 μm in diameter and 50 μm in height are studied. The flow regimes, gas hold-up, and pressure drop are determined for pillar pitches of 7, 12, 17, and 27 μm. Flow maps are presented and compared with flow maps of rectangular and round micro channels without pillars. The Armand correlation predicts the gas hold-up in the pillared micro channel within 3% error. Three models are derived which give the single-phase and the two-phase pressure drop as a function of the gas and liquid superficial velocities and the pillar pitches. For a pillar pitch of 27 μm, the Darcy-Brinkman equation predicts the single-phase pressure drop within 2% error. For pillar pitches of 7, 12, and 17 μm, the Blake-Kozeny equation predicts the single-phase pressure drop within 20%. The two-phase pressure drop model predicts the experimental data within 30% error for channels containing pillars with a pitch of 17 μm, whereas the Lockhart–Martinelli correlation is proven to be non-applicable for the system used in this work. The open structure and the higher production rate per unit of reactor volume make the pillared micro channel an efficient system for performing heterogeneously catalyzed gas–liquid reactions.     

Effect of residual stress on RF MEMS switch

Studies have been carried out on a RF MEMS shunt switch to analyze the effect of residual stress on its electromechanical characteristics. This paper presents the simulated results as well as theoretically calculated results of a shunt switch due to the presence of residual stress gradient in respect of resonant frequency, pull down voltage and switching characteristics. The effect of introduction of holes in the beam is also studied. The calculated results, corresponding to the switch (without holes) at zero residual stress, of resonant frequency, pull-down voltage and switch on and off time are 28.14 kHz, 28.2 V, 16.35 μsec and 8.6 μsec respectively. Modal analysis of the both the structures (with and without holes) are carried out for different values of residual stress gradients. Modal analysis predicted that higher values of tensile stress gradient are not favorable for switching action. The pull-down voltages and switch on and off times are simulated at different stress gradients. With the increase in compressive stress gradient, the pull-down voltage is found to increase, whereas, switch on and off times is decreased. Corresponding to −20 MPa/μm residual stress gradient, the resonant frequency, pull-down voltage and switch on and off times are found to be 74.5 kHz, 63.5 V, 7.5 μsec and 3.36 μsec respectively. Introduction holes in the structure modified these values to 63.77 kHz, 53.1 V, 8.7 μsec, 3.92 μsec respectively.