J. Micro/Nanolith. MEMS MOEMS 9共3兲, 031007 共Jul–Sep 2010兲 Three-dimensional flexible microprobe for recording the neural signal Chang-Hsiao Chen Shih-Chang Chuang Yu-Tao Lee National Tsing Hua University Institute of NanoEngineering and MicroSystems Hsinchu 30013, Taiwan Yen-Chung Chang National Tsing Hua University Institute of NanoEngineering and MicroSystems and Institute of Molecular Medicine and Department of Life Sciences Hsinchu 30013, Taiwan Abstract. We have designed, fabricated, and tested a novel threedimensional 共3-D兲 flexible microprobe used for recording the neural signals of lateral giant 共LG兲 on the escape system of American crayfish. We report an electrostatic actuation process to fold the planar probes to be the arbitrary orientations of 3-D probes for neuroscience application. The batch assembly method based on electrostatic force techniques gave more simple fabrication compared to others. A flexible probe could reduce both the chronic inflammation response and material fracture when animal breathes or moves. Furthermore, the cortex corresponds to hypothetical cortical modules with mostly vertically organized layers of neurons. Therefore, the 3-D flexible probe suits to understand how the cooperative activity for different layers of neurons. Advisedly, we present a novel fabrication for the 3-D flexible probe by using Parylene technology. The mechanical strength of the neural probe is strong enough to penetrate into a biogel. At the end, the flexible probe was used to record neural signals of the LG cell from American crayfish. © 2010 Society of PhotoOptical Instrumentation Engineers. 关DOI: 10.1117/1.3455409兴 Shih-Rung Yeh National Tsing Hua University Institute of Molecular Medicine and Department of Life Sciences Hsinchu 30013, Taiwan Subject terms: flexible; microprobe; neuron; parylene; three-dimensional. Paper 09159SSRR received Nov. 30, 2009; revised manuscript received May 1, 2010; accepted for publication May 4, 2010; published online Jul. 23, 2010. Da-Jeng Yao National Tsing Hua University Institute of NanoEngineering and MicroSystems Hsinchu 30013, Taiwan E-mail: djyao@mx.nthu.edu.tw 1 Introduction A fundamental goal of neuroscience is to develop quantitative descriptions of the functional operations and their coordination within a neuron that enable it to function as an integrated unit to process information. The communication of information among neurons, or between neuron and muscle, requires that a signal travels over a considerable distance. Through electrical stimulation applied to nerves and muscles, early electrophysiologists demonstrated that a command from the brain to a muscle to generate a movement was mediated by the flow of an electric current along nerve fibers, so-called an action potential.1 Hodgkin and Huxley2 demonstrated that axons at rest are electrically polarized, exhibiting a resting potential approximately −60 mV across a membrane. This action potential propagates along the length of the axon through local depolarization of each neighboring patch of membrane, causing that patch of membrane to also generate an action potential. Action potentials from neural cells have been measured over a wide range. To record the neuron signal, a single electrode, comprising a sharpened tungsten wire probe and a capillary glass probe, was used traditionally,3 but it is 1932-5150/2010/$25.00 © 2010 SPIE J. Micro/Nanolith. MEMS MOEMS difficult to achieve uniform geometries of such finished array probes. It is impractical to assemble, manually, individual probes into an array with a small controlled spacing, although a linear array or a two-dimensional matrix has been reported.4 The neural probes of the 3-D type could be used for neural recording in a brain system in vivo with its penetration into neural cells, with a probe of variable length. In contrast to conventional techniques, several researchers have been developing a neural probe with a microelectromechanical systems 共MEMS兲 technique. Wise made a silicon-based neural probe based on wet silicon etching.5 The individual probes are released using an ethylene diamine pyrocatechol 共EDP兲 etch. During this EDP step, the boron-diffused areas act as an etch stop.6 Neural probes combined with monolithically integrated CMOS circuitry have also been demonstrated.7 The four 64-sites probes have been assembled into a three-dimensional 共3-D兲 structure.8 The shape of neural probe from the University of Michigan formed by diffusing the boron dopant process was uncontrollable.6 The Utah neural probe consists of hundreds of silicon needles which were fabricated by using a dicing saw process followed by chemical etchant.9 However, the individual probes can support only one electrode site. Although rigid types of microelectrode have been devel- 031007-1 Downloaded from SPIE Digital Library on 19 Aug 2010 to 140.114.88.25. Terms of Use: http://spiedl.org/terms Jul–Sep 2010/Vol. 9共3兲 Chen et al.: Three-dimensional flexible microprobe for recording the neural signal 2 Design of Neural Probe Before fabricating electrode arrays with an appropriate geometry, the mechanical strength of electrodes was calculated using ANSYS. Each probe must satisfy the following design criteria: 1. The probe must be strong to minimize damage on penetrating a bio-organism. 2. The probe must be sufficiently long, ⬎1 mm, to allow for the desired depth of penetration into a brain cortex. 3. The exposed area of the surface material of an electrode must be polarized such as Au, Pt, or Ir, because the application would be used for extracellular recording. 4. The interconnect spacing from electrodes to bonding pads were kept to ⬍10 ␮m, to both limit the crosstalk capacitance 共CC兲 and facilitate fabrication using liftoff technology. 5. The use of polymer or glass as the substrate effectively eliminated the substrate capacitance 共CS兲 and the light sensitivity since the polymer, glass was nonconductive. Fig. 1 Schematic of the planar probe in the 共a兲 flat and 共b兲 folded configurations. oped for recoding, but at in vivo recording the mechanical mismatch between the stiffness probe and soft biological organization will aggravate inflammation in the implantation sites. Although silicon has a Young’s modulus of ⬃170 GPa, brain tissue has ⬃3 kPa. This large mechanical mismatch can contribute to shear-induced inflammation at the implant site. This inflammation encourages the formation of a glial sheath. In response to brain injury, glial cells such as astrocytes and microglia proliferate to form a glial sheath.10,11 The biological sheath would encapsulate the probe, which would isolate the electrode from surrounding neurons with time.12 The duration recording of chronically implanted probe of the Michigan array about four months,13 of the Utah array was about seven months.14 A flexible probe could reduce both the chronic tissue inflammation response and material fracture when the animal breathes or moves.15,16 Recently, flexible neural probes have been fabricated with polymer materials such as polyimide, Parylene, and SU-8.15,17,18 Advisedly, we present a fabrication method for a penetrating type of 3-D flexible probe by using Parylene technology. Parylene is an inert, vapor-deposited polymer. The Parylene-C is selected as the main structure material of the flexible probe, which gives its excellent flexibility, mechanical strength with Young’s modulus ⬃3 GPa, and its United States Pharmacopoeia 共USP兲 Class VI grade biocompatibility; pinhole-free coating at room temperature has low moisture permeability and has been used in the medical device industry as a biocompatible coating for implants. We report an electrostatic actuation process to fold the planar probes to be the arbitrary orientations of 3-D probes, shown in Fig. 1. Our method differs from that of Shaar et al.19 in the segments actuated by Lorentz force, and of Takeuchi et al.17 actuated by the magnetic field method for the batch assembly. Moreover, we improve the limit that toxicity of magnetic materials for long-term and in vivo recording application, which would be very useful for further development. At the in vivo recording system, the mechanical mismatch between the stiffness probe and soft bioorganization will aggravate inflammation.11 A flexible probe could reduce both the chronic inflammation response and material fracture when the animal breathes or moves. Furthermore, the cortex corresponds to hypothetical cortical modules with mostly vertically organized layers of neurons. Therefore, the 3-D flexible probe suits to understand how the cooperative activity for different layers of neurons. Advisedly, we present a novel fabrication for 3-D flexible probe by using Parylene technology. J. Micro/Nanolith. MEMS MOEMS In this paper we have designed a novel 3-D flexible microprobe used for recording the neural signals of LG of the American crayfish. We report an electrostatic actuation process to fold the planar probes to be the arbitrary orientations of 3-D probes for neuroscience application. The batch-assembly method based on electrostatic force techniques gave more simple fabrication compared to others. The advantages of using the fabrication process of the 3-D flexible neural probe as follows: the shank is longer than 3 mm, the flexibility, 3-D type, and excellent biocompatibility. Besides, the multiple electrode sites upon individual probe could be designed in the future. A neural probe of this size might be used to measure a neural signal; a long probe is susceptible to bending when it contacts 共extracellular recording兲 the neuron cell. The flexible probe insert into the cortex is difficult. Takeuchi presented a fabrication process of the Parylene probes that were filled with polyethylene glycol 共PEG兲. In order to increase the stiffness of a flexible probe during implantation, an embedded microfludic channel can be filled with a solid material that dissolves after implantation.20 The Parylene probes were coated with PEG, a biocompatible material that is solid state at room temperature, liquid state over 60 ° C and dissolves in water or body fluid. 3 Fabrication of Neural Probe MEMS technology was used to develop the fabrication of the flexible probe, the fabrication starting from Parylene_C deposition and ending with PEG coating. The fabrication process of the flexible probe was outlined in Fig. 2. The process consisted of three steps: patterning the metal layer, the structure layer, and etching the structure to release the probe. Details of the processes are as follows: 1. RCA standard clean was used to remove particles and ion contamination before the ⬃5 ␮m Parylene_C is first deposited on silicon wafer by using specialty coating system. 031007-2 Downloaded from SPIE Digital Library on 19 Aug 2010 to 140.114.88.25. Terms of Use: http://spiedl.org/terms Jul–Sep 2010/Vol. 9共3兲 Chen et al.: Three-dimensional flexible microprobe for recording the neural signal Fig. 3 SEM image of a fabricated 3-D flexible microprobe: 共a兲 The probe has one electrode on the tip. The geometry of developed probe is 3 mm long, 100 ␮m wide, and 6 ␮m thick. The close view of the probe for which the electrode was isolated with Parylene. The width of tip is 5 ␮m, and the opening size of gold electrode is 2500 ␮m2. 共b兲 The folding probe after electrostatic actuation. Fig. 2 Fabrication process of the 3-D flexible probe. 2. The process steps define patterning of electrodes by using the liftoff method. Cr/ Au were evaporated by E-beam. The metal patterns are defined in an ultrasonic bath in acetone. The trace was defined with a 5 ␮m width and 100 nm thickness. The embedded electrodes add the lateral contact area with Parylene. The O2 plasma treatment with reactive ion etching 共RIE兲 was utilized to etch part of the Parylene before depositing the gold electrode. A treated recipe of power 共100 W兲 at 60 s could improve the mechanical properties efficiently. Additionally, the rough surface added the adhesion between electrode and Parylene. 3. A second Parylene_C with a 1-␮m thickness layer was deposited on top of the entire wafer as a passivation dielectric layer. 4. The 2500 ␮m2 via was patterned in the insulation layer, which—over the ends of the Au electrodes— contact pads were opened to the Au layer using O2 plasma by RIE through a photoresist mask. This step at the same time defined the probe shape. The etch step went through the entire thickness of the Parylene device layer on the wafer, which was etched 6 ␮m, stopping on the silicon. 5. The flexible probe was peeled from the wafer by using tweezers and fixed on the glass substrate. 6. The planar probes were folded by electrostatic force. 7. The folded probes were fixed by a PEG drop with liquid state on a 60 ° C hotplate. 8. Finally, the probe was dipped with liquid-state PEG on a 60 ° C hotplate. After the cooling process, the PEG transfers to solid state, which enhances the probe’s stiffness. outermost section. The thickness of probe is targeted to be 6 ␮m, but can be varied by depositing Parylene with various thicknesses. The probe length is typically 3 mm. The size of the recording electrode sites are 2500 ␮m2. The forded probe was fixed with a PEG drop by using a pipette. The flexible probe was coated with PEG by a dipping process, as shown in Fig. 3共b兲. The thickness of PEG was ⬃5 ␮m. The dip-coating process has a conformal surface on the implant site. 4.2 Electrostatic Field Assembly The electrostatic field was used to fold the planar probe for the batch assembly, which avoids the error of manual assembly. After fabrication, the planar probe was placed in a vertical electrostatic field, which would stand up by varied electrostatic voltages. The voltage in the generator was gradually increased from 0 V to 8 KV, which folded the probe up from a horizontal position to about vertical. It was folded to near 90 deg when an external electrostatic field of 8 KV was applied, as shown in Fig. 4. 4.3 Mechanical Strength The flexible probe would be used for in vivo recording of the neural signal of the cortex in the future. The pure flexible probe was too difficult to be implant into a living organism due to its soft property. The PEG material was as a supporting structure of the Parylene probe. We had to verify that the overall neural probe structure is mechanically strong enough to penetrate test. 4 Results and Discussion 4.1 Probe Structure A scanning electron microscope 共SEM兲 image of the completed neural probe is shown in Fig. 3. In the design, the flexible probe consists of one electrode site on the probe 关shown on Fig. 3共a兲兴. The width of probe is 100 ␮m at the J. Micro/Nanolith. MEMS MOEMS Fig. 4 Electrostatic batches assembly of the probe. 031007-3 Downloaded from SPIE Digital Library on 19 Aug 2010 to 140.114.88.25. Terms of Use: http://spiedl.org/terms Jul–Sep 2010/Vol. 9共3兲 Chen et al.: Three-dimensional flexible microprobe for recording the neural signal Fig. 5 Photograph of Parylene probe coated with PEG penetrated into biogel. As shown in Fig. 5, the 3-mm Parylene probe was coated with PEG, then inserted into and pulled back from the biogel, smoothly, to simulate the hardness of the cortex and without critical buckling failure or break off. After inserting the probe into the biogel, the PEG was a biocompatible material. It would be dissolved in body fluid with time and the probe would become flexibile again. 4.4 Interface Impedance The impedance of electrode influences the ability to record minute neural signals. In order to choose the cutoff frequency of the neural signal recording system, electrochemical impedance analysis was used to measure the impedance and phase between the work electrode and reference electrode. The larger impedance of the electrode abated was, the electrical transmission would increase the noise, which influences the signal-to-noise ratio 共SNR兲 of the spike. The impedance of the microelectrode was obtained by submerging only the recording in saline solution 共NaCl 0.9%, 27 ° C兲. The impedance measurement 共HIOKI LCR meter, 3522-50兲 was built up with respect to a large reference electrode 共Ag–AgCl electrode兲. The testing signal for impedance measurement was sinusoidal 共ac voltage 10 mV, frequency 100 Hz– 10 kHz兲. For a typical electrode with a 2500 ␮m2 opening area, the average measured impedance is ⬃566.5 K⍀ at 1 kHz 共nerve working frequency兲 and the phase shift is −84 deg/ 1 kHz, tending to −90 deg representing a capacitance dominating the interface, with the electrode interface being highly polarized,21 as shown in Fig. 6. At the higher frequency, the impedance of electrode decreased gradually Fig. 6 Interface impedances of gold electrodes on flexible probe submerged in a saline buffer solution 共NaCl 0.9%兲, the average impedance of electrode is 566.5 KΩ / 1 kHz. J. Micro/Nanolith. MEMS MOEMS Fig. 7 Schematic of the electrophysiology system for recording the crayfish. Pairs of silver wires are placed on the mechanosensory primary afferent neurons for stimulation. The Parylene-based neural probe contacts on a LG neuron cell are for the measurement of the action potential. because a capacitor effect tended to a short circuit. Additionally, the long-term stability of the neural probe was tested immersing in the saline buffer solution 共NaCl 0.9%兲 by impedance measurement. The Parylene film still adhered to the probe well after one month. The adhesion remained tight with no noticeable degradation, visually, and the change of interface impedance was ⬍1% for the duration. 4.5 Neural Signal Recording To confirm the functionality of electrode on the flexible probe, we measured the neural signal of the escape circuit of a crayfish with our developed neural probe. The advantage of using the escape circuit from a crayfish is the large diameter of an axon, ⬎10 ␮m, and the ⬃1 mm length level. The neuron is large enough to contact or have implanted in it a neural probe, to minimize the error of the distance between the electrode and the measured neuron cell. The schematic view of the electrophysiology system 共shown in Fig. 7兲 could be used to record the neuron signal from microprobe. Juvenile crayfish are kept in a water tank at room temperature. The 2 – 5 cm junior crayfish was selected for our experiments. The crayfish lost the ability to move on the ice after 10 min. A crayfish was pinned in a dish filled with van Harreveld’s solution. The abdominal nerve cord of the escape circuits was exposed dorsally by removing the exoskeleton and separating phasic flexor musculature, which’s using isolated abdomen preparation. In the escape circuits of the crayfish, the mechanosensory primary afferents received the environmental excitation. Parts of the neural signal transmit directly to the LG through electrical synapses. Pairs of silver wires were placed on the mechanosensory primary afferent neurons for stimulation. The functions of the electrophysiology system were divided into two parts: one is electrical stimulation and the other one is the recording part. The electrical stimulation was produced by a 031007-4 Downloaded from SPIE Digital Library on 19 Aug 2010 to 140.114.88.25. Terms of Use: http://spiedl.org/terms Jul–Sep 2010/Vol. 9共3兲 Chen et al.: Three-dimensional flexible microprobe for recording the neural signal mechanical stiffness of developed the flexible probe, PEG was used on the surface of probe. The neural recording data were used to demonstrate that the probes are capable of recording data from an electrode using crayfish. In future work, we would develop further completed data acquisition software and measurement circuitry, including stimulating, recording, and bypassing functions with fully integrated digital components realized in an industrial CMOS process. To continue to improve the SNR of the neural signal would be important for a multipleelectrode measurement system. Future durability tests must include a toxicity test, long–term signal recording, and stimulation in neural cells. Fig. 8 Extracellular recording signals of an LG neuron were measured from an electrode of the flexible neural probe by electrical stimulation. The response amplitude of each spike was ⬃150 ␮V, and the response time ⬃1 ms. Acknowledgments This research was partially supported by the National Science Council in Taiwan through an NSC Grant No. 962627-E-007-002. References digital-to-analog converter card and amplifier circuits. The amplitude and frequency of the input voltage was controlled by software in a personal computer. With an appropriate packaging—the substrate was packaged with the biocompatible epoxy after wire bonding process 共Epoxy Technology, EPO-TEK 301-2兲 and its USP Class VI grade biocompatibility, an electrode of a flexible neural probe was used to measure the neuron signal on the LG after electrically stimulating. In the extracellular recording experiment, the flexible neural probe was contacted on the membrane of the LG to record the neural responses. The neural signals from the dorsal side of a LG nerve fiber were recorded on shocking the tailfin afferents. Figure 8 shows the recorded extracellular signals from our developed flexible neural probe. The alternating square waveform of the shock voltage was about ⫾3.7 V in a period of 0.1 ms. The responsive amplitude of the action potential was ⬃150 ␮V for 1 ms 共red curve兲. Nevertheless, the smaller shock electrical stimulation, which was about ⫾3.6 V for 0.1 ms 共blue curve兲, could not induce the action potential from LG nerve fiber because of the resting membrane potential toward to the threshold for activation is yet to surmount. Additionally, the distances between the LG neuron and the electrode varied under measurement, which would influence the amplitude of the response spike. However, most neurophysiologists are only concerned with the appearance of the spike instead of its response magnitude. The results demonstrate the capability of our developed neural probe for recording neural signals independently from a single electrode. 5 Conclusion In this paper, we designed and fabricated a 3-D flexible microprobe for recording a neural signal, which would be used for the measurement system in biomedical applications. The fabrication process of the implant flexible probe was accomplished by using MEMS technology. We report an electrostatic actuation process to fold the planar probes to be the arbitrary orientations of 3-D probes. We present a fabrication method for a penetrating type of flexible probe by using Parylene technology. Moreover, to improve the J. Micro/Nanolith. MEMS MOEMS 1. M. A. B. Brazier, A History of Neurophysiology in the 19th Century, Raven Press, New York 共1988兲. 2. A. L. Hodgkin and A. F. Huxley, “Action potentials recorded from inside a nerve fibre,” Nature (London) 144, 710–711 共1939兲. 3. G. E. Loeb, R. A. Peck, and J. Martyniuk, “Toward the ultimate metal microelectrode,” J. Neurosci. Methods 63, 175–183 共1995兲. 4. D. Jaeger, S. Gilman, and J. Wayne Aldridge, “A multiwire microelectrode for single unit recording in deep brain structures,” J. Neurosci. Methods 32, 143–148 共1990兲. 5. K. D. Wise, J. B. Angell, and A. Starr, “An integrated-circuit approach to extracellular microelectrodes,” IEEE Trans. Biomed. Eng. 17, 238–247 共1970兲. 6. K. Najafi, K. D. Wise, and T. Mochizuki, “A high-yield ICcompatible multichannel recording array,” IEEE Trans. Electron Devices 32, 1206–1211 共1985兲. 7. B. Qing and K. D. Wise, “Single-unit neural recording with active microelectrode arrays,” IEEE Trans. Biomed. Eng. 48, 911–920 共2001兲. 8. K. D. Wise, “Silicon microsystems for neuroscience and neural prostheses,” IEEE Eng. Med. Biol. Mag. 24, 22–29 共2005兲. 9. P. K. Campbell, K. E. Jones, R. J. Huber, K. W. Horch, and R. A. Normann, “A silicon-based, three-dimensional neural interface: manufacturing processes for an intracortical electrode array,” IEEE Trans. Biomed. Eng. 38, 758–768 共1991兲. 10. J. W. Fawcett and R. A. Asher, “The glial scar and central nervous system repair,” Brain Res. Bull. 49, 377–391 共1999兲. 11. V. S. Polikov, P. A. Tresco, and W. M. Reichert, “Response of brain tissue to chronically implanted neural electrodes,” J. Neurosci. Methods 148, 1–18 共2005兲. 12. D. H. Szarowski, M. D. Andersen, S. Retterer, A. J. Spence, M. Isaacson, H. G. Craighead, J. N. Turner, and W. Shain, “Brain responses to micro-machined silicon devices,” Brain Res. 983, 23–35 共2003兲. 13. R. J. Vetter, J. C. Williams, J. F. Hetke, E. A. Nunamaker, and D. R. Kipke, “Chronic neural recording using silicon-substrate microelectrode arrays implanted in cerebral cortex,” IEEE Trans. Biomed. Eng. 51, 896–904 共2004兲. 14. A. Branner, R. B. Stein, E. Fernandez, Y. Aoyagi, and R. A. Normann, “Long-term stimulation and recording with a penetrating microelectrode array in cat sciatic nerve,” IEEE Trans. Biomed. Eng. 51, 146–157 共2004兲. 15. K. C. Cheung, P. Renaud, H. Tanila, and K. Djupsund, “Flexible polyimide microelectrode array for in vivo recordings and current source density analysis,” Biosens. Bioelectron. 22, 1783–1790 共2007兲. 16. J. P. Seymour and D. R. Kipke, “Neural probe design for reduced tissue encapsulation in CNS,” Biomaterials 28, 3594–3607 共2007兲. 17. S. Takeuchi, T. Suzuki, K. Mabuchi, and H. Fujita, “3D flexible multichannel neural probe array,” J. Micromech. Microeng. 14, 104–107 共2004兲. 18. H. Lu, S. H. Cho, J. B. Lee, L. Cauller, M. R. Ortega, and G. Hughes, “SU8-based micro neural probe for enhanced chronic in vivo recording of spike signals from regenerated axons,” presented at 5th IEEE Conf. on Sensors, Daegu 共2006兲. 19. N. S. Shaar, G. Barbastathis, and C. Livermore, “Cascaded mechanical alignment for assembling 3D MEMS,” in Proc. of MEMS 2008, Tucson, pp. 1064–1068 共2008兲. 031007-5 Downloaded from SPIE Digital Library on 19 Aug 2010 to 140.114.88.25. Terms of Use: http://spiedl.org/terms Jul–Sep 2010/Vol. 9共3兲 Chen et al.: Three-dimensional flexible microprobe for recording the neural signal Shih-Rung Yeh is an associate professor at Institute of Molecular Medicine, National Tsing Hua University, Taiwan. He received his PhD from the Department of Biology, George State University at Atlanta in 1997. Being an electrophysiologist, he is interested in the the neural plasiticity of the nerve system. 20. T. Suzuki, K. Mabuchi, and S. Takeuchi, “A 3D flexible parylene probe array for multichannel neural recording,” in Proc. of 1st Int. IEEE EMBS Conf. on Neural Engineering, 2003, pp. 154–156 共2003兲. 21. H. Fricke, “The theory of electrolytic polarization,” Philos. Mag. Series 7 14, 310–318 共1932兲. Chang-Hsiao Chen is a PhD candidate at the Institute of NanoEngineering and MicroSystems, National Tsing Hua University, Taiwan, when he received his MS in SOIbased microprobe techniques in 2006. His research includes microprobe fabrication, carbon nanotube treatment, and signal process technology. Shih-Chang Chuang studied micromechanical system 共MEMS兲 at National Tsing Hua University, Taiwan, where he graduated in 2009. The subject of his master thesis was the design and fabrication of flexible neural microprobe for three-dimensional assembly. Da-Jeng Yao is an associate professor at Institute of NanoEngineering and MicroSystems 共NEMS兲, also an adjunct professor at Department of Power Mechanical Engineering and Department of Engineering System and Science, National Tsing Hua University, Taiwan. He received his PhD from the Department of Mechanical and Aerospace Engineering at University of California at Los Angeles in 2001. His research scope is to combine his strong backgrounds 共MEMS and thermal fluidics兲 in microscience research, including BioMEMS, electronic nose, MEMS packaging, thermofluidic MEMS, and thinfilm property measurement. Yen-Chung Chang is a professor at the Institute of Molecular Medicine and an adjunct professor at the Institute of NanoEngineering and MicroSystems of National Tsing Hua University, Taiwan. He received his PhD from the Department of Biochemistry and Biophysics at Iowa State University in 1985. His current research interests lie in understanding the molecular mechanisms underlying the formation and function of synapses in the mammalian central Nervous System and the proteomics of subcellular structures, such as the postsynaptic density and axon of central neurons. J. Micro/Nanolith. MEMS MOEMS 031007-6 Downloaded from SPIE Digital Library on 19 Aug 2010 to 140.114.88.25. Terms of Use: http://spiedl.org/terms Jul–Sep 2010/Vol. 9共3兲