Development and characterization of ... - Semantic Scholar

536

JOURNAL OF MICROELECTROMECHANICAL SYSTEMS, VOL. 11, NO. 5, OCTOBER 2002

Development and Characterization of Micromachined Hollow Cathode Plasma Display Devices Jack Chen, Sung-Jin Park, Associate Member, IEEE, Zhifang Fan, J. Gary Eden, Fellow, IEEE, and Chang Liu

Abstract—In this paper, we report the development of hollowcathode micro plasma devices made using micromachining techniques. Compared with larger discharge devices, micromachined discharge devices operate at much higher pressures, up to 1 atm. Linear current-voltage relationships are obtained, potentially simplifying the control electronics for such devices. The size of micromachined discharge units is reduced and the distribution of sizes and light intensity in an array is more uniform relative to previous devices. [721] Index Terms—Hollow cathode, microdischarge, micromachine.

I. INTRODUCTION

T

HE hollow cathode effect was discovered by Paschen [1] in 1916 and subsequently developed for its use in a spectroscopic light source. Since then, its application has been extended to lasers, pseudosparks [2], ion thrusters [3] and display panels [4]. A hollow cathode discharge (HCD) is a subclass of glow discharge and offers characteristics quite unique from that of a planar cathode [5]. In a discharge between two planar electrodes, there are three main regions to consider. In the cathode fall region, electrons are accelerated. At the edge of the cathode fall, where the plasma potential is essentially constant, maximum electron kinetic energy occurs. Beyond that is the negative glow region, where electrons are slowed. The length of the negative glow region is proportional to the square of the voltage applied. Therefore, the power density of a planar glow discharge is limited because the glow region is spread over a long distance. The third region is the positive column, which makes up the remaining length of the cathode-anode distance. In HCD, the cathode is a hollow cavity and the negative glow region of the glow discharge is effectively folded within the interior of the cavity. Under the influence of a radial electrical field, electrons are accelerated from the cathode wall toward the center of the cathode and decelerated as they move toward the opposite cathode wall. This has the effect of increasing the average electron energy and multiplying the intensity of the glow. The key features of the HCD over a planar discharge can be summarized as follows. First, the HCD is associated with an increased current density and discharge efficiency. Second, for given dimensions and voltage, the discharge current is only Manuscript received July 18, 2001; revised February 7, 2002. Subject Editor T. Kenny. The authors are with the Electrical and Computer Engineering Department, University of Illinois at Urbana-Champaign, Urbana, IL 61801 USA (e-mail: [email protected]). Digital Object Identifier 10.1109/JMEMS.2002.802907.

weakly influenced by gas pressure [6]. Finally, the HCD requires a lower operating voltage, thereby simplifying the control electronics. It is important to reduce the size of the cavity of HCD devices. For display applications, for example, the reduced sizes of each discharge device translate into increased display resolution. Another important motivation for reducing sizes is to increase the operating pressure. Operation near atmospheric pressure greatly simplifies device packaging by eliminating the needs to maintain a large differential pressure over a long period of time. In order for an HCD to operate properly the pressure of the discharge gas must be below an upper pressure limit. The ) for HCD operation theoretical upper limit of pressure ( represents the transition to a normal glow discharge. Early experiments by White [7] were conducted at a product of 7.5 torr-cm in 100 torr of neon, in which is the diameter of the cavity. When the pressure is above this limit, the negative glow length will drop to below the cathode wall spacing and the oscillatory effect characteristic of the HCD will not be sustained. Therefore, reduction of the cavity dimensions allows a device to operate as a hollow cathode at higher pressure. A conventional hollow cathode discharge with cathode spacing in the centimeter range will typically operate at pressures up to a few torr. The decrease in the diameter of a cylindrical cathode down to the micron level, for example, will permit an HCD to operate at pressures beyond one atmosphere. In recent years, researchers began to investigate methods for miniaturizing discharge devices. Microdischarge devices have been made in molybdenum [8], silicon [9] and polymer/metal film structures [10]. The microcavities were generally made in these devices by drilling: mechanically, with ultrasound, or with a laser. However, the feature sizes that were made by using these methods are on the order of 75 m [5]. Devices made by such methods suffer from a number of shortcomings. First, the size of the cathode opening is large and its cross-section profile is not uniform. The wear and tear of the drill bits, along with normal mechanical inaccuracies, prevents achieving accurate dimensional control and repeatability. Second, due to the serial nature of the processes, it is expensive and time-consuming to realize a large array of such discharge devices. The objective of this work is to develop and characterize micromachined HCD devices and their operation in a large array. Micromachining technology promises to further reduce the sizes of HCD as well as their costs. Because of the batch-nature of micromachining, not only are the performance characteristics of the device improved, but the cost of fabricating large arrays is also reduced. The ability to successfully fabricate large arrays with precise tolerances and high density

1057-7157/02$17.00 © 2002 IEEE

CHEN et al.: MICROMACHINED HOLLOW CATHODE PLASMA DISPLAY DEVICES

537

Fig. 1. Schematic diagrams of three types of microdischarge devices.

(resolution) makes these devices feasible for display applications. Other potential applications include gas lasers, miniature wavelength calibration light sources and any application where a high-resolution lighting source is required. II. DESIGN AND FABRICATION We have designed and fabricated three types of devices (see Fig. 1). A Type-1 device is a reference planar microdischarge device that does not have a cathode cavity. A Type-2 device is a micro-HCD device with an inverted square pyramidal cavity. A Type-3 HCD device includes a cylindrical cathode cavity made using deep reactive ion etching (DRIE). The structures of these three types of devices, along with their fabrication methods, are discussed in detail. Two silicon bulk-micromachining techniques are available to create the hollow cathode cavity with two distinct geometric profiles in the silicon substrate: anisotropic crystalline wet etching and DRIE [11]. Using the wet etching method, an inverted pyramid will be generated with the surfaces defined by {111} crystal planes. The overall depth of such cathodes is exactly determined by the width of the cavity opening. The DRIE process creates a hole with relatively vertical sidewalls. The depth of the holes is, in this case, independent of the cathode diameter. A. Planar Microdischarge Devices (Type 1) The planar microdischarge device was fabricated to establish a reference for evaluating the performance of HCD devices (i.e., Type-2 and Type-3 devices). Each Type-1 device consists of a 1500- -thick nickel thin film on top of an 8- m-thick polyimide dielectric layer, which is, in turn, deposited onto a {100} Si surface. The fabrication procedure is as follows. First, an 8- m-thick polyimide film (PI 2611) is spun onto a lightly doped silicon substrate and cured at 350 C for 2 h. Second, a 1500- -thick nickel film is evaporated and patterned on top of the polyimide dielectric. The nickel film serves as a mask against subsequent plasma etch and as the discharge anode. The rate of metal deposition is controlled such that the substrate is not overheated and the polyimide film is not deformed or damaged by heat generated in the metal evaporation process. The nickel film is patterned in a diluted HCl: HNO solution, which does not attack the polyimide. Third, the polyimide is patterned using O reactive ion etching (RIE).

A number of dielectric materials have been evaluated, including thermally grown silicon dioxide and polyimide. Thermally grown silicon dioxide film has good dielectric breakdown characteristics, with the nominal breakdown field being on the order of 10 V/cm [12]. However, the thickness of such an oxide is limited. In order to increase the maximum allowed voltage across the dielectric, we chose to use polyimide, which has a slightly lower dielectric breakdown field (2 10 V/cm) compared to that of silicon dioxide, but can achieve greater thickness with spin coating. A thicker polyimide layer can, of course, withstand greater voltages. B. Microdischarge Device With Inverted Pyramidal Cathode (Type 2) The fabrication process for a Type-2 device is illustrated in Fig. 2. The inverted cavity is formed before the polyimide dielectric is spun on because the polyimide material (PI 2611) can be attacked by the silicon anisotropic wet etchant and therefore cannot function as a mask. The cavity is formed using KOH etchant at 65 C with SiO as the masking layer. Etching is self-limited, stopping after four {111} walls are completely developed. The SiO mask is subsequently removed. Before the polyimide is deposited, a 100:1 polyimide primer (VM651) diluted with water is first spun on to prevent peeling of the polyimide during the subsequent curing process. The wafer is baked in a convection oven to eliminate the solution-phase primer in the cavity. To achieve an 8- m polyimide thickness, the polyimide is spun-on twice with a 20-min soft bake in between. The final curing is performed in a 350 C furnace with flowing N gas for 2 hrs. Subsequently, a 1500- -thick nickel film is evaporated onto the cured polyimide surface and patterned using photolithography. The nickel layer serves as both the anode and the masking layer for the polyimide during the subsequent etch step. O RIE etching is then performed to fully remove the polyimide material within the cavity. Two different views of fabricated Type-2 device, are given by the scanning electron micrographs of Fig. 3. C. Microdischarge Device With a Vertical Cavity Having a Rectangular Cross-Section (Type 3) A Type-3 device with vertical sidewalls rather than sloped {111} surfaces is created using a DRIE etching process that has recently become commercially available. The fabrication

538


Fig. 3. SEMs of a Type-2 discharge device. (a) SEM showing a top view of the device. (b) SEM of the cross section of a single device. Particulates in the figure result from dicing.

[7], such a device will accommodate discharges at pressures up to 1500 torr. D. Setup of Experimental Characterization

Fig. 2. Schematic diagram of the fabrication process for a Type-2 device. (a) The process starts with a silicon substrate with approximately 4000 A thermal SiO on both sides. (b) An inverted pyramidal cavity is formed using KOH etching. (c) The SiO layer is stripped and an 8 m polyimide is spun onto the wafer. (d) 1500 A thick Nickel anode is evaporated and patterned. (e) The polyimide film within the cavity is etched in an O plasma.

process is shown in Fig. 4. The processing of the nickel and the polyimide layer is similar to the planar discharge device, but they are also used as a mask during the etching of the silicon in an inductively coupled plasma (ICP) system. The silicon exposed by the polymer opening is etched down to a desired depth. SEMs of the device and its cross section are shown in Fig. 5, parts (a) and (b). The device has a 50- m-square cavity cross section. If we accept the value from White

We tested the fabricated devices in a stainless-steel, turbomolecular-pumped vacuum station. Contact to the anode is made with gold wires and a copper base provides electrical contact to the silicon substrate, the cathode. The chamber is pumped down to 10 torr and backfilled with research-grade neon gas to the desired pressure. The voltage applied between the cathode and anode is gradually increased until breakdown occurs. In certain cases, external ballast is used during device operation to stabilize discharge performance and protect the device. Measurement of the optical power produced by the device in the 300–800 nm spectral region is made with a calibrated pin diode and spectra were acquired in first order with a spectrograph and diode array. III. RESULTS AND DISCUSSION The current-voltage characteristics of the discharge devices were collected from all three types of devices. Type-1 devices exhibit relatively low power output and low light emission, a result that is similar to those reported in by Frame and Eden. For such devices, we did not make arrays or investigate their spectroscopic property owing to low emission intensity. Several iterations of Type-2 devices were fabricated. Spectral data as well as I–V characteristics of arrays of these devices were obtained (see Fig. 6).


539

Fig. 5. Microdischarge device with a trench formed by DRIE etching. (a) SEM showing a top view of the device. (b) SEM of the cross section of a single device.

Fig. 6. Schematic diagram of the setup used for testing the I–V characteristics of the discharge devices. Fig. 4. Schematic diagram of the fabrication process for a Type-3 device. (a) Starting Si wafer. (b) 8 m-thick polyimide thin film is spun-on. (c) 1500 m-thick Nickel anode is evaporated and patterned. (d) The polyimide film is patterned in an O plasma. (e) The silicon substrate is etched by DRIE.

A. Comparison Between Type-1 and Type-2 Devices In order to compare the discharge performance, we present the current–voltage (I–V) characteristics for two singular microdischarge devices, one of which is Type-1 (see Fig. 7) and another is of Type-2 (see Fig. 8). Both devices have surface areas (at the cathode) of 50 m and the same dielectric and anode thickness. As illustrated in Figs. 7 and 8, the I–V relationships for the Type-1 and Type-2 devices were investigated at Ne gas pressure in the ranges of 400–800 torr and 300–800 torr, respectively.

For all of the pressures investigated, the I–V curves for the planar cathode (Type I) show two regions with distinct degrees of slope. The first region is associated with a larger I–V gradient, i.e., differential resistivity. In the second region, the current increases quickly with little or no change in voltage. For a Type-2 device, in which the square pyramid serves as the cathode, the linear regime covers a much wider current range than its Type-1 counterpart. Moreover, the I–V characteristics are very similar over the entire range of gas pressures (300–800 torr). Under a fixed device voltage, the current varies by less than 20% as the Ne pressure is increased above one atmosphere. When the bias is 300 V, the current changed from 1.0 A to 1.15 A. In contrast, the current changes drastically when the pressure is varied for a Type-1 device. This insensitivity to pressure is consistent with characteristics of HCD.

540


Fig. 7. Measured I–V characteristics for a Type-1 microdischarge device operated in neon gas. 50-m-wide cavity opening, 8-m-thick polyimide, 1500- A-thick nickel electrode.

Fig. 8. I–V characteristics for a Type-2 microdischarge device operated in neon gas. 50-m-wide cavity opening, 8-m-thick polyimide, 1500- A-thick nickel electrode.

Fig. 9. I–V characteristics of a Type-3 microdischarge device operated in neon. 50-m-wide cavity opening, 8-m-thick polyimide, 1500- A-thick nickel electrode, approx. 200-m-deep trench.

Fig. 10. Emission spectrum in the near-UV (330–370 nm) for (50 m) Type-2 device with a composite dielectric (SiO =Si N /polyimide) operated at a voltage of 270 V. Peaks labeled by asterisks ( ) result from singly charged Ne ions. The discharge current is also indicated.

B. Comparison Between Type-1 and Type-3 Devices Next, we compare the measured I–V characteristics for a Type-1 and Type-3 device (see Fig. 9). Once again, both devices have the same cross-sectional area [ 50 m ] at the cathode-dielectric interface. Also, Ne gas pressures ranging from 400 to 800 torr were used for the Type-3 device. The primary difference between the measured characteristics of these two devices is that the operation voltage of the Type-3 device is significantly lower than that of the Type-1 (or Type-2) device. Operating voltages as low as 145 V were observed when the Ne pressure is 600 torr, for example. We also note that the differential resistivity for the Type-3 device increases slightly with rising pressure. Finally, at low operating voltages ( 230 V), the Type-3 device operates at currents at least an order of magnitude greater than those available with the planar cathode.

Fig. 11. I–V characteristics for arrays of Type-2 discharge devices. 50-m-wide cavity opening, 8-m-thick polyimide, 1500- A-thick nickel electrode, 50-m cavity spacing.


541

2

Fig. 12. Photograph of a 3 3 array of microdischarge devices having cross-sectional areas of (50 m) . (a) CCD image of the array before ignition. (b) CDD image of the array during operation in Ne. (c) CDD image with a neutral density filter between the camera and device, illustrating uniform light intensity. The operating voltage is 433 V and the current is 21.4 A.

C. Comparison Between Type-2 and Type-3 Devices The differential resistance recorded in the Type-2 device ( 1 ) is approximately four orders of magnitude larger than those recorded for devices having a cylindrical cavity extending as much as several hundred microns into the Si [9]. This can be partially attributed to the shallow depth of the Type-2 device cathode, which is only 35 m deep in the present device. Also, the bulk resistivity of the silicon wafer itself is a factor. Therefore, in an effort to further increase the effective cathode area, a vertical cavity 200 m deep was produced in the Type-3 device. Another group of devices has been made with a more shallow depth of only 50 m. The I–V characteristics of a 200- m-deep Type-3 device are shown in Fig. 9. Similar to the curve for the Type-2 devices with a pyramidal cavity, the I–V slope is positive and linear over most of the current range investigated. The working voltage for the Type-3 is approximately 100 V lower than that for a Type-2 device having a square pyramidal cathode and the current is an order of magnitude larger. The differential resistance of , which is about one order of the Type-3 device is 10 magnitude smaller than that of the Type-2 device. Of course, the lower operating voltages of the Type-3 devices simplifies the driving circuits. Preliminary measurements suggest that the I–V characteristic of a Type-3 device with a 50- m-deep cavity is similar to that for a Type-2 device. D. Hollow Cathode Effect at Small Scale In order to verify that the fabricated Type-2 devices operate within the HCD regime, we investigated the spectroscopic signature of these discharges. Emission spectra serve as a reliable indicator of key discharge properties and the relative electron temperature. Although more than 90% of total emission produced by a glow Ne discharge lies in the red, emission in the ultraviolet region is most sensitive to the electron energy distribution function and, hence, to hollow cathode operation [9]. In the spectra of Fig. 10, the transition from hollow cathode operation to a normal glow is clearly observed as pressure inside the discharge column increases. In the figure, we present the output spectra of a Type-2 device operating at pressures between 300 and 900 torr. The output spectrum of a pen lamp, a commercial positive column discharge device, is shown as a reference.

The pen lamp and the Type-2 device both exhibit several emission peaks between 330 to 380 nm, many of which are assignable to neutral Ne. However, several spectral peaks (marked by an asterisk) arise from the Ne species that are not observed in the pen lamp. A few of the Ne ion transitions originate from excited states lying more than 55 eV above the 2 ground state for the neutral species. As expected, the intensity of the Ne lines diminishes quickly with increasing pressure and is quite weak at one atmosphere. E. Array Operation of Micro Discharge Devices The ability to fabricate and integrate microdischarge devices using micromachining techniques allows for large arrays of devices with nearly identical geometries to be realized. To date, arrays as large as 10 10 have been made to ignite uniformly. The I–V characteristics of arrays ranging from 2 2 to 6 6 are presented in Fig. 11. All of the individual devices have a width and interdevice spacing of 50 m and the data shown are for a pressure of 700 torr Ne gas with a 56 k ballast. The ballast is not necessary for operation but is used to facilitate current measurement. For a given bias voltage, experiments show that the output current increases with the size of the array. The voltage necessary to ignite these arrays does not vary significantly with the array size. m m devices are Three photos of a 3 3 array of shown in Fig. 12. The center and right micrograms show the array in operation and when the emission is attenuated with a neutral density filter (Fig. 12(c)). It can be seen that the emission produced by the array is spatially uniform. The operating voltage and current are 433 V and 21.4 A, respectively. A micrograph of a 10 10 array, the largest array tested to date, is shown in Fig. 13, operating at 1200 torr. Arrays larger than 10 10 have also been made. Limited tests were conducted and it was found that not all devices within the array ignite simultaneously and emissions are not uniform from device to device. F. Long Term Testing and Device Failure Preliminary lifetime tests of fabricated devices, consisting of measuring the wavelength integrated output power of an array or device, have also been conducted. Type-2 devices with at least 20 h of continuous working lifetime have been demonstrated for DC excitation. Time-intensity measurement of 8 sets of 3 3 arrays operating simultaneously is shown in Fig. 14.

542


and electrons. An SEM of a Type-2 device after failure is shown in Fig. 15. Our spectroscopic observations of the microdischarge plasmas show no signs of Si lines; therefore we expect that much greater lifetimes can be attained if doped polysilicon is used for the electrode instead of metals. The lifetime of Type-3 devices is expected to be at least on the same order of magnitude as Type-2 devices due to the similarity of dielectric and electrode materials. More detailed lifetime tests of these devices, including optical and electrical measurements, are in progress. IV. CONCLUSION

2

Fig. 13. Optical micrograph of a 10 10 array of Type-2 devices during operation. 50-m-wide cavity opening, 8-m-thick polyimide, 1500- A-thick nickel electrode and 50-m cavity spacing. Operational parameters: V = 244 V, I 20 mA, P = 1200 torr.

Three types of microdischarge device have been fabricated in silicon using surface and bulk-micromachining techniques. Micromachined HCD devices (Type-2 and Type-3) exhibit a number of important advantages: (1) the dimensional control afforded by micromachining techniques allows single microdischarge devices smaller than 50 m to be fabricated and be operated at pressure above 1 atm, (2) large arrays producing spatially uniform emissions are possible and (3) the fabrication costs for a large array are reduced. We have demonstrated that the Type-2 microdischarge device, with anisotropic etched (tapered) cavities, can operate over a wide range of pressure, from 400 torr to well above atmospheric pressure. The cavity allows the discharge to operate in the hollow cathode mode, thereby producing a ballistic component to the electron energy distribution function. Type-3 devices with vertical cavities and larger effective cathode areas exhibit even greater current output capability, greater light output intensity and lower operating voltage. Larger arrays of Type-2 and Type-3 devices are currently under development. ACKNOWLEDGMENT

2

Fig. 14. Long term relative intensity of eight 3 3 discharge array at a fixed voltage measured with acalibrated Si head power meter positioned 5 cm away from the microplasma.

The authors wish to thank J. Hughes and R. Blaney for their assistance with microfabrication laboratory support and staff members of the Micro-Miniature Systems Lab for the help with ICP etching. REFERENCES

Fig. 15. An SEM of an array of damaged Type-2 devices, illustrating the erosion of the dielectric and anode materials.

We found that failures are commonly caused by the erosion of materials inside the discharge devices induced by energetic ions

[1] F. Paschen, Ann. Physik, vol. 50, p. 901, 1916. [2] G. Schaefer and K. H. Schoenbach, 1990 Physics and Applications of Pseudosparks, ser. Nato ASI Series B 219, M. A. Gundersen and G. Schaefer, Eds.. [3] D. G. Fearn et al., “The operation of ion thruster hollow cathodes using rare gas propellants,” in 21st Internationl Electric Propulsion Conference, July 1990. [4] K. C. Choi, “A new DC plasma display panel using microbridge structure and hollow cathode discharge,” IEEE Trans. Electron Dev., vol. 46, p. 2256, 1999. [5] J. W. Frame and J. G. Eden, “Planar microdischarge arrays,” Electron. Lett., vol. 34, no. 15, pp. 1529–1531, July 1998. [6] R. R. Arslanbekov, A. A. Kudryavtsev, and R. C. Tobin, “On the hollowcathode effect: Conventional and modified geometry,” Plasma Sources Sci. Technol., vol. 7, no. 3, pp. 310–322, Aug. 1998. [7] A. D. White, J. Appl. Phys, vol. 30, p. 711, 1959. [8] K. H. Schoenbach, R. Verhappen, T. Tessnow, F. E. Peterkin, and W. W. Byszewski, “Microhollow cathode discharges,” App. Phys. Lett., vol. 68, no. 1, pp. 13–15, Jan. 1996. [9] J. W. Frame, D. J. Wheeler, T. A. DeTemple, and J. G. Eden, “Microdischarge devices fabricated in silicon,” App. Phys. Lett., vol. 71, no. 9, pp. 1165–1167, 1997. [10] S.-J. Park, C. J. Wagner, C. M. Herring, and J. G. Eden, “Flexible microdischarge arrays: Metal/polymer devices,” App. Phys. Lett., vol. 77, no. 2, pp. 199–201, 2000.


[11] E. Klassen, K. Peterson, J. Noworolski, J. Logan, N. Maluf, J. Brown, C. Strorment, W. McCulley, and G. Kovacs, “Silicon fusion bonding and deep reactive ion etching: A new technology for microstructures,” Sens. Actuators, Phys. A, vol. 52, no. 1–3, pp. 132–139, 1996. [12] A. C. Adams, “Dielectric and polysilicon film deposition,” in VLSI Technology, S. M. Sze, Ed. New York: McGraw-Hill, 1983.

Jack Chen received two B.S. degrees in mechanical engineering and electrical engineering from the University of Illinois at Urbana-Champaign, in May 2000. As an undergraduate, his research focused on fabricating microdischarge devices in silicon. He is currently pursuing a Ph.D. degree in electrical engineering at the Micro and Nanotechnology Laboratory at the University of Illinois. His current research focuses on micromachined flow sensors and microdischarge device fabrication.

Sung-Jin Park (A’01) received the M.S. and Ph.D. degrees in chemistry from the Myongji University, Korea, in 1995 and 1999, respectively. He worked on high-resolution spectroscopy and various organic materials for the application in optoelectronic devices. He then joined Laboratory for Optical Physics and Engineering in Department of Electrical and Communication Engineering, University of Illinois, Urbana-Champaign. His current research involves study of microdischarge devices and arrays, their application in photonics and development of electronic materials. Dr. Park is a Member of IEEE Laser and Electro-Optics Society, Society for Information Display (SID), and Directed Energy Professional Society (DEPS).

Zhifang Fan recieved the B.S. and M.S. degrees from the Electrical Engineering Department of Tsinghua University, Beijing, China, in 1955 and 1961. From 1976, he was the Chief Research Engineer and Professor at the Institute of High Energy Physics, Academia Sinica, People’s Republic of China; there, he worked on semiconductor oriented research in semiconductor materials and engaged in nuclear electronics research of High Energy Accelerator. From 1990, he was a Research Engineer in the Coordinated Science Laboratory, University of Illinois at Urbana-Champaign. He was involved in research in compound semiconductor thin film growth by molecular beam epitaxy, device design, fabrication of high-speed device of GaN and GaAs fabrication and characterization. He also performed research in heterojunction bipolar transistors, field-effect transistors, power transistors, laser diodes and photodetectors. From 1998, he was Senior Research Engineer work in Microelectronics laboratory at University of Illinois at Urbana-Champaign, where he was involved in research in compound semiconductor, get GaN blue light emit diode (LED) in Thin film Group of Electrical Communication Engineering (ECE) and research and fabrication the high-speed IC of GaAs in High-Speed Integrate Circuit Group of ECE. Now he is a research engineer with the Micro Actuators, Sensors and Systems Group (MASS) at the Micro and Nanotechnology Laboratory in the University of Illinois. His research involves the design and fabrication of tactile Sensor and Flow Sensor with Integrated Circuits control circuit.

543

J. Gary Eden (S’75–M’76–SM’82–F’88) was born in Washington, DC, on October 11, 1950. He received the B.S. degree in electrical engineering (high honors) from the University of Maryland, College Park, in 1972 and the M.S. and Ph.D. degrees from the University of Illinois at Urbana-Champaign, in 1973 and 1976, respectively. In 1975, he was awarded a National Research Council Postdoctoral Research Associateship at the U.S. Naval Research Laboratory (NRL). In November of 1976, he joined the staff of the Laser Physics Branch (Optical Sciences Division) of NRL. During his tenure at NRL, he made several contributions to the area of visible and ultraviolet lasers and gas laser spectroscopy, including the codiscovery of the KrCl laser at 222 nm and the first long-pulse (> 1 sec) excimer laser. He received a Research Publication Award (1979) for his work at NRL in which he co-discovered the proton beam pumped laser (Ar-N , XeF). Since joining the faculty of the University of Illinois in 1979, he has been engaged in research in atomic and molecular spectroscopy, the development of visible and ultraviolet lasers and ultrafast optical physics. He and his students have demonstrated several new lasers and amplifiers in both the gas phase and in glass fibers, including the first ultraviolet and violet fiber lasers. Recent accomplishments of his research group include the generation of coherent UV by atomic wavepackets and the development of microdischarge devices in silicon. At the University of Illinois, he has served as Assistant Dean in the College of Engineering and Associate Dean of the Graduate College and is currently Associate Vice-Chancellor for Research, Professor in the Department of Electrical and Computer Engineering and Director of the Laboratory for Optical Physics and Engineering. He has over 180 publications and holds 15 patents. Dr. Eden is a Fellow of the Optical Society of America (OSA) and the American Physical Society. From 1996 through 1999, he was the James F. Towey University Scholar at the University of Illinois. He is Editor-in-Chief of the IEEE JOURNAL OF QUANTUM ELECTRONICS and, in 1998, served as President of the IEEE Lasers and Electro-Optics Society (LEOS). Previously, he served as a Member of the LEOS Board of Governors and the Vice-President for Technical Affairs. Recently, He received the LEOS Distinguished Service Award and was awarded the IEEE Third Millennium medal.

Chang Liu pursued his undergraduate studies at Tsinghua University, Beijing, China. He received the M.S. and Ph.D. degrees from the California Institute of Technology, Pasadena, in 1991 and 1996, respectively. He is currently an Assistant Professor at the University of Illinois at Urbana-Champaign, where he directs the Micro Actuators, Sensors, and Systems (MASS) Research Group. His group is concentrating on the following research areas: bioinspired, low-cost integrated sensors, microfluidic systems for bio/chemical analysis and high throughput nanolithography systems. A summary of research topics being conducted at the group can be accessed from the following html web site: http://mass.micro.uiuc.edu