452 IEEE TRANSACTIONS ON SEMICONDUCTOR MANUFACTURING, VOL. 16, NO. 3, AUGUST 2003

Photoresist Deposition Without SpinningGökhan Perçin, Member, IEEEand Butrus T. Khuri-Yakub, Fellow, IEEE

Abstract—A technique for resist deposition using a novel fluidejection method is presented in this paper. An ejector has been de-veloped to deposit photoresist on silicon wafers without spinning.Drop-on-demand coating of the wafer reduces waste and the costof coating wafers. Shipley 1400-21, 1400-27, 1805, and 1813 resistswere used to coat sample 3- and 4-in wafers. Later, these waferswere exposed and developed. The deposited resist film was 3.5mthick and had a surface roughness of about 0.2 m. The ultimategoal is to deposit resist films with a thickness of the order of 0.5 mand a surface roughness of the order of 30�A, which is currentlyachieved for 200-mm silicon wafers by using a spinning method.Such goals can be attained by using micromachined multiple ejec-tors or with better control over the deposition environment. In themicromachined configuration, thousands of ejectors are made intoa silicon die, as presented by Perçinet al. (2002), and thus allowfor a full coating of a wafer in a few seconds. Coating in a cleanenvironment will allow the lithography of circuits for microelec-tronic applications. Other potential applications for the technologyin the semiconductor manufacturing are in deposition of low- ma-terials, wafer cleaning, manufacturing of organic LEDs and or-ganic FETs, direct lithography, nanolithography, and coating forhard-disk drives.

Index Terms—Coating, direct lithography, ink-jet, microelec-tromechanical system (MEMS), photoresist, resist deposition,resist dispenser.

I. INTRODUCTION

T HERE IS a continuing need for alternative depositiontechniques of organic polymers in precision droplet-based

manufacturing and material synthesis [1], [2], such as thedeposition of doped organic polymers for organic light emittingdevices of flat panel displays [3]–[7]. There is also need fordeposition of photoresist without spinning on large or oddlyshaped substrates. In addition, small particle ejectors arenecessary for the study of heating and combustion behavior ofsmall solid particles, such as coal and metals [8]. To date, therehas been no report of a drop-on-demand solid particle ejectorthat can eject fine solid particles with spatial control, although acontinuous mode pneumatically operated solid particle ejectorhas been reported [8].

Of all the needs for a liquid or solid particle ejector, the depo-sition of organic polymers used in semiconductor manufacturing

Manuscript received March 15, 2002; revised February 16, 2003. This workwas supported in part by the Defense Advanced Research Projects Agency, De-partment of Defense, and was monitored by the Air Force Office of ScientificResearch under Grant F49620-95-1-0525. This work made use of the NationalNanofabrication Users Network facilities funded by the National Science Foun-dation under Award ECS-9731294. This work was conducted in part during G.Perçin’s Ph.D. study at Stanford University.

G. Perçin is with ADEPTIENT, Palo Alto, CA 94306-2021 USA (e-mail:percin@adeptient.com).

B. T. Khuri-Yakub is with Edward L. Ginzton Laboratory, Stanford Univer-sity, Stanford, CA 94305-4085 USA (e-mail: khuri-yakub@stanford.edu).

Digital Object Identifier 10.1109/TSM.2003.815197

and microelectromechanical systems (MEMS) is worth the mostattention. Lithography is the most expensive step in microelec-tronics technology.Photoresistcoating is theoneof theexpensivesteps in the lithography process. For common applications, thespin coating of photoresist on an appropriate substrate, in mostcases a round silicon wafer without any significant topography,is conventionally used in microelectronics IC fabrication. One ofthe most critical roles of spin coating systems is to properly cast athin film of photoresist on the surface of a silicon wafer. The pho-toresist film, after application to the substrate, must have uniformthickness and must be chemically isotropic so that its responseto exposure and development is uniform. The film must be ex-tremely uniform in thickness 6 , and wafer-to-wafer meanthickness control must be better than 30, total indicated rangeover extended periods of time. In the literature, several types ofphotoresist coating methods are reported [9], [10]: spin coating,spray coating, dip coating, meniscus coating, plasma-depositedphotoresist, electrodeposited (electrophoretic) photoresist, androller, curtain and extrusion coating. A meniscus coating processhas been established for flat panel display substrates. The ad-vantage of this technology is the savings of the coating material(only 5%–10% loss) and the avoidance of edge build-up, unlikespin coating. Techniques which involve a direct contact betweenthe substrate surface and the photoresist bulk source, such asmeniscus coating or silk screen printing techniques, lead to aproper coverage depending on the adhesion of the resist on thesubstrate surface. Another interesting method is the depositionof polymer from gas phase, i.e., plasma-deposited photoresist.This method requires a monomeric coating material, which maybe evaporated at ambient temperatures and forms its polymericform after deposition on the substrate. Electrodeposited (elec-trophoretic) photoresist has been proposed for the coating ofsubstrates with extreme topography. This process needs metalplating of substrates and has some associated process complex-ities such as the requirement for wafer electrical biasing duringthe resist coating process.

Spin coating remains the method of choice of the microelec-tronic industry. In present applications, over 95% of the pho-toresist is wasted and has to be disposed of as a toxic material.Radial thickness variations associated with the application ofphotoresist by spinning must be avoided for 300 mm or largersilicon wafers. The origin of the potential variations in physicalproperties of the photoresist film lies in the spin coating tech-nique. A large amount of extra free volume is trapped in the filmduring the spin coating.

The device we developed uses a drop-on-demand methodto dispense photoresist on wafers. Drop-on-demand coatingof wafers reduces the waste and the cost of coating thesewafers. DUV photoresists currently available on the marketcost $3000–$6000 per gallon, and 0.7–1.0 cc photoresist is

0894-6507/03$17.00 © 2003 IEEE

PERÇIN AND KHURI-YAKUB: PHOTORESIST DEPOSITION WITHOUT SPINNING 453

(a)

(b)

Fig. 1. (a) General schematic of the ejector. (b) Picture of the ejector.

used with the current tracks to cover a 200-mm silicon wafer.The developed deposition method saves up to $10 per 200-mmmemory wafer in production. This is a significant saving giventhat approximately 0.5 millions wafers per year are processedper track, and each semiconductor foundry has 30–50 tracks.In addition to reducing waste, this method can be used to coatoddly shaped substrates (i.e., flat panel displays), to planarizethe resist profile on the wafers (i.e., putting more resist on someparts of the wafers), to finely control its spatial distributionin real-time, to possibly reduce the use of HMDS, and to dodirect write for MEMS where critical dimensions are in theorder of several microns. The piezoelectric fluid ejector canfind applications in integrated circuit manufacturing not onlyfor photoresist coating but also for dispensing chemicals todesired regions. The ejector is harmless to sensitive fluids.The method also opens new possibilities, e.g., intentionallyvarying the resist thickness across the wafer. The adaptation ofthe piezoelectric fluid ejector technology to the semiconductorindustry will meet the future demands of the microelectronicand MEMS industries.

II. EJECTOR

A schematic and picture of the ejector are shown in Fig. 1.A thin shim with a small orifice is bonded to a piezoelectricannular disk. A cylinder is attached to the shim to serve both asa fluid reservoir and to clamp the ends of the compound plateformed of the shim and piezoelectric. The reservoir is open, andthe fluid is at atmospheric pressure. A silicon micromachinedversion of the device is also presented in Perçinet al. [1], [2].

We use brass shims, steel shims, and silicon membranes as acarrier plate and Murata Surface Wave Material [11], MotorolaPZT 3203HD [12], and lithium niobateLiNbO as a piezo-electric material in our designs. Table I shows the physical con-

stants of the piezoelectric materials used. The piezoelectric ma-terials were chosen because they retain their properties whenpolished to a thickness of 25m, whereas brass and steel shimswere chosen because they are easily available. On the otherhand, we use silicon membranes in order not to contaminate thephotoresist. We designed the droplet ejector to have maximumdisplacement at the center of the flexurally vibrating circularplate at its resonant frequency when loaded with fluid. Severaliterations were run to maximize the displacement of the plate asa function of the dimensions of the piezoelectric annular disk byusing the analytical model developed in Perçinet al.[13]. Max-imum displacement is obtained when the piezoelectric annulardisk (Murata Surface Wave Material in this case) has an innerdiameter of 2 mm and outer diameter of 7 mm for a brass shimwith a diameter of 9 mm.

Large-scale model devices shown in Fig. 1 were fabricatedusing the optimum configuration obtained by using the modeldeveloped. The reservoir was made of brass with a height of 8mm. A 25- m-thin shim was bonded to the 25-m-thick piezo-electric annular disk. The inner and outer diameters of the an-nular disk were 2 and 7 mm, respectively. The orifice rangedin diameter from 50 to 200m and was made using a drill in asmall lathe or chemical etching. The measured first resonant fre-quency of the device was 2.5 kHz in air and 1.2 kHz with fluidloading on one side. Both resonant frequencies are in agreementwith the model prediction.

A. Vibration of Compound Plate

The harmonic transverse axisymmetric displacement,, of the compound plate shown in

Fig. 1 satisfies the following differential equation for bothsingle-layered (only metal shim) and double-layered (metalshim and piezoelectric) regions:

(1)

where is the angular frequency of vibrations,is the pressure,is the built-in tensile force per unit length of the thickness of

the plate, is the flexural rigidity of the compound plate, whichis different for each region, and

(2)

where is the mass density of the plate.The general axisymmetric solution to (1) is in the following

form:

(3)

where

(4)

454 IEEE TRANSACTIONS ON SEMICONDUCTOR MANUFACTURING, VOL. 16, NO. 3, AUGUST 2003

TABLE IPROPERTIES OFPIEZOELECTRICMATERIALS USED

One should apply the boundary conditions for bending momentsper unit length, displacements, and inplane forces per unit lengthto obtain the the coefficients, , , and in (3) for single-and double-layered regions separately. The details of the sim-ilar analysis, where built-in tensile force is zero , ispresented in Perçinet al. [13].

B. Drop Formation

The vibrating plate sets up capillary waves at the liquid-air in-terface and raises the pressure in the liquid above atmospheric(as high as 1.5 MPa) during part of a cycle, and if this pres-sure rise stays above atmospheric pressure long enough duringa cycle, and this is high enough to overcome inertia and surfacetension restoring forces, drops are ejected through the orifice.If the plate displacement amplitude is too small, the meniscusin the orifice simply oscillates up and down. If the frequency istoo high, the pressure in the fluid does not remain above atmo-spheric long enough to eject a drop.

The motion of a drop of inviscid liquid in a gas of negligibledensity (such as air) can be computed by a boundary integralmethod, which has the advantage of requiring a numerical gridonly on the surface of the drop. The velocity potential of anirrotational flow can be expressed as a surface distribution ofdipoles

(5)

where

(6)

is the velocity potential at a pointdue to an infititesimal sourcelocated at , a point on the surface of the drop, is thederivative in the direction of the outer normal to the surfaceat , and is the dipole density per unit area. The readershould note that the dipole density equals the jump in the ve-locity potential, , across the surface of the drop.

Since the velocity of the surface of the drop can be computed,the surface can be evolved in time with the normal velocity,which can be found from

(7)

(a)

(b)

Fig. 2. Dry photoresist thickness simulations.

Assuming that gravitional forces are negligible, the pressuredifference across the surface of the drop must be balanced byonly surface tension yielding the following pressure boundarycondition:

(8)

where is the pressure inside the drop,is the surface tension,and is the outward unit normal.

PERÇIN AND KHURI-YAKUB: PHOTORESIST DEPOSITION WITHOUT SPINNING 455

Fig. 3. S1813 photoresist ejection simulations and picture at 7.15 kHz through 110-�m diameter orifice.

(a)

(b)

Fig. 4. Ejected photoresist droplets through 110-�m diameter orifice by usinglarge-scale device. Note that the camera is tilted.

The pressure can be determined from Bernoulli’s equation as

(9)

Then, using (8)

(10)

(a)

(b)

Fig. 5. Photoresist covered 3-in wafers.

456 IEEE TRANSACTIONS ON SEMICONDUCTOR MANUFACTURING, VOL. 16, NO. 3, AUGUST 2003

Fig. 6. Top, scan of individual photoresist drops; bottom left, scan at the edge of the photoresist pattern; bottom right, scan inside the photoresistpattern. Verticalaxis is ink�A, and the horizontal axis is in�m.

These equations can be put in dimensionless form by using acharacteristic length, the radius of the orifice, and time ,the period of plate oscillations. The previous equations thenhave identical form, except in (10), where is replaced bythe dimensionless surface tension parameter

(11)

where is the radius of the orifice, is the frequency of oscilla-tions, and and are surface tension and density of the liquid.This provides a scaling law for drop ejection. All other thingsbeing equal, such as amplitude and plate mode shape, this saysthat droplet size and shape are only a function of this single di-mensionless parameter. For instance, ifis made larger theneither the orifice radius should be smaller ormade larger.

To solve for the shape of the drop, one must solve (7) and(10) simultaneously. A computational model which simulatesdroplet ejection has been also developed using a boundaryintegral method in Perçinet al. [14]. Singular potential flowdipole solutions were distributed along the liquid/air interfacein an axially symmetric configuration, as in the previous work[15], and singular source solutions were distributed on the solidmembrane surface. Enforcing boundary conditions on thesesurfaces, pressure-surface tension balance on the liquid inter-face and specifying the velocity on the solid membrane (alsomode shape) using (3) gives integral equations to determinethe dipole density and the source density functions. The useof both dipoles and sources in this manner results in coupledFredholm equations of the second kind which can be solvedsimultaneously by an iterative process. It should be emphasizedthat viscous effects have been neglected in this analysis giventhat the ejection velocity scales with . A Reynolds number

(12)

which describes the ratio of inertial forces to viscous forces,should be sufficiently large for the analysis to be valid, whereis kinematic viscosity of the liquid.

C. Drop Spreading

Another crucial point in the theory is the spreading of pho-toresist drops on the wafers. There is a wide range of literatureon this topic and can be found in Baueret al.[16], Shikhmurzaev[17], Yarin et al. [18], Mundoet al. [19], Daviset al. [20], andPasandideh-Fardet al. [21]. In Mundoet al. [19], the relationbetween the ejected drop radius and the spot radius on the sub-strate is given by

(13)where is the radius of the ejected drop, is the radius ofthe spot on the substrate, is Reynolds number,

is Weber number, is the contact angle,is the ejected drop velocity, andis dynamic viscosity of

the liquid. In Fig. 2, the simulations for dry photoresist thick-ness on a wafer are given as a function of droplet ejector ori-fice diameter. In the simulations, one drop is ejected to everyspot; however, by ejecting multiple drops to each location, samephotoresist can be used to obtain different film thicknesses. Onthe other hand, the conventional spin coating relies on the dif-ferent photoresist formulations with different viscosities to ob-tain different film thicknesses. For the film thickness variation,the orifice diameter is assumed to vary by 1%. In the compu-tations, we have used N/m, mPa.sec,

g/cm , , and m/s.

III. EXPERIMENTS AND RESULTS

Shipley Microposit S1400-21, S1400-27, S1805, and S1813photoresists were ejected using the ejector under the conditions

PERÇIN AND KHURI-YAKUB: PHOTORESIST DEPOSITION WITHOUT SPINNING 457

(a)

(b)

Fig. 7. Novel inkjet deposited, exposed, and developed photoresist. Lines andspaces are 10�m wide. Pattern at the right is located at 150-�m deep silicontrench.

described in Section II, with a 200-V peak-to-peak voltage.Table II shows the physical properties of the photoresistsejected. The ejected photoresist drop size is 85% of the orificesize as shown in Figs. 3 and 4, where orifice size is 110m.Fig. 3, where the simulated droplet ejections are shown withactual dimensions at every interval, also shows thatthe simulated droplet ejection closely resembles the actualphotoresist droplet ejection picture with a similar elongated tail.The orifice diameter is 110m, the ejection frequency is 7.15kHz (period 139.86 s), and the ejected fluid is Shipley S1813photoresist. The final velocity is about 1.79 m/s for the firstmode and 0.98 m/s for the second mode. In the computation,the drop pinchoff time was 123.0s and the surface tensionparameter . We have used N/m and

g/cm for calculating these parameters. Fig. 5 showssquare coating of 3-in silicon wafers with zero waste. Thespacing between the wafer and the ejector was around 2 mm.The resist is 3.5 m thick and has a surface roughness of about0.2 m. Fig. 6 shows profilometer (alpha-step) scans of thedeposited photoresist pattern. The roughness at the edge of thephotoresist pattern is caused by overlapping photoresist dropson the wafer. Since there is no overlapping photoresist dropat the edge of the pattern, the surface tension restoring forcescannot smooth out the thickness profile. The resist coating

(a)

(b)

Fig. 8. Photoresist coverage of 150-�m deep anisotropically (orientationdependent) etched silicon trenches. Photoresist coverage of these trenches isnot possible with spinning method.

Fig. 9. Drop-on-demand direct write with photoresist. Lines are 350�m wide.

was done without HMDS treatment in a dry laboratory thatcontains dust particles, and nonuniformity is due to the quickevaporation of the solvent in the photoresist. This drying results

458 IEEE TRANSACTIONS ON SEMICONDUCTOR MANUFACTURING, VOL. 16, NO. 3, AUGUST 2003

TABLE IIPROPERTIES OFSHIPLEY MICROPOSITPHOTORESISTSUSED IN THE EXPERIMENTS

Fig. 10. Configuration of the micromachined droplet ejector. Spacing betweenthe adjacent array elements ranges from 150 to 400�m. Orifice diameter rangesfrom 4 to 10�m. Vibrating plate diameter ranges from 90 to 500�m.

in nonuniformity as the wafer is coated. Using a chamber witha solvent saturated environment will alleviate both problems ofdirt incorporation and nonuniformity [22].

Fig. 7 shows photoresist patterns that were exposed and de-veloped into the wafer. The lines and spaces are 10m wide,and, on the right, the same pattern is located on a 150-m deepsilicon trench. Fig. 7 also shows that the developed depositionmethod does not harm the ejected photoresist chemically orphysically. The uniform coverage of sidewalls of etched cavitiesis a challenge in fabrication of MEMS devices. Micromachinedcomponents are fabricated using various etching technologies(e.g., wet anisotropic, dry plasma etching and deep reactive ionetching) on substrates which may then expose different crys-tallographic planes. The combination of these parameters leadsto several different slopes in the sidewalls of etched trenches.Fig. 8 shows SEM image of photoresist coverage in a deep sil-icon trench. Although the application of common spin coatingleads to a flat resist surface of highly uniform thickness, thistechnology fails dramatically in the coverage of deeply etchedtrenches and grooves. As shown in Fig. 8, substrates with rect-angular shapes and MEMS devices with extreme topographyin height and size can be covered with a uniform resist layerby using the fluid ejection method developed. Fig. 9 demon-strates the ability to deposit lines of photoresist (without HMDStreatment) that are 350m wide. Narrower lines can be de-posited with smaller drops for microelectromechanical systems(MEMS) where critical dimensions are in the order of severalmicrons. Coating in a clean environment will allow the lithog-raphy of circuits for microelectronic applications.

IV. CONCLUSION

In summary, we have developed a photoresist coating tech-nique by using inkjet technology. To do so, we have also de-veloped an ejector which is silicon micromachined into two-di-mensional arrays. The micromachined modified version of thedevice shown in Fig. 10 is presented in Perçinet al. [23], wheretwo-dimesional array is actuated with bulk piezoelectric mate-rial bonded above the fluid reservoirs rather than with individualpiezoelectric elements deposited on each vibrating plate as pre-sented in Perçinet al. [1], [2]. The ejector is based on usinga variation of a flextensional transducer and the transducer de-sign was optimized by using the model developed in Perçinet al.[13]. The developed photoresist deposition method needs morework to improve the uniformity and the speed of the coverage;however, it is demonstrated that it has practically zero waste.

REFERENCES

[1] G. Perçin and B. T. Khuri-Yakub, “Piezoelectrically actuated flex-tensional micromachined ultrasound droplet ejectors,”IEEE Trans.Ultrason., Ferroelect., Freq. Contr., vol. 49, pp. 756–766, June 2002.

[2] , “Micromachined droplet ejector arrays for controlled ink-jetprinting and deposition,”Rev. Sci. Instrum., vol. 73, pp. 2193–2196,May 2002.

[3] W. S. Wong, S. Ready, R. Matusiak, S. D. White, J. P. Lu, J. Ho, and R.A. Street, “Amorphous silicon thin-film transistors and arrays fabricatedby jet printing,”Appl. Phys. Lett., vol. 80, pp. 610–612, Jan. 2002.

[4] P. Calvert, “Inkjet printing for materials and devices,”Chem. Mater., vol.13, pp. 3299–3305, Oct. 2001.

[5] V. Bulovic and S. R. Forrest, “Polymeric and molecular organic lightemitting devices: A comparison,”Semiconduct. Semimetals, vol. 65, pp.1–26, 2000.

[6] Y. Yang, S. C. Chang, J. Bharathan, and J. Liu, “Organic/polymeric elec-troluminescent devices processed by hybrid ink-jet printing,”J. Mater.Sci.—Mater. Electron., vol. 11, pp. 89–96, Mar. 2000.

[7] T. R. Hebner and J. C. Sturm, “Local tuning of organic light-emittingdiode color by dye droplet application,”Appl. Phys. Lett., vol. 73, pp.1775–1777, Sept. 1998.

[8] C. C. Hwang, “Small solid-particle ejector,”Rev. Sci. Instrum., vol. 51,pp. 581–584, May 1980.

[9] Personal communication Advanced Process Concepts, S. Bagen. (1997).[Online]. Available: BAGEN@aol.com

[10] B. Bednár, J. Králícek, and J. Zachoval,Resists in Microlithography andPrinting. Amsterdam, The Netherlands: Elsevier Science, 1993, pp.77–82.

[11] Murata Electronics North America, Inc., Smyrna, GA.[12] Motorola, Ceramic Products Division, Albuquerque, NM.[13] G. Perçin and B. T. Khuri-Yakub, “Piezoelectrically actuated flexten-

sional micromachined ultrasound transducers—I: Theory,”IEEE Trans.Ultrason., Ferroelect., Freq. Contr., vol. 49, pp. 573–584, May 2002.

[14] G. Perçin, T. S. Lundgren, and B. T. Khuri-Yakub, “Controlled ink-jetprinting and deposition of organic polymers and solid-particles,”Appl.Phys. Lett., vol. 73, pp. 2375–2377, Oct. 1998.

PERÇIN AND KHURI-YAKUB: PHOTORESIST DEPOSITION WITHOUT SPINNING 459

[15] T. S. Lundgren and N. N. Mansour, “Oscillations of drops in zero gravitywith weak viscous effects,”J. Fluid Mechanics, vol. 194, pp. 479–510,Sept. 1988.

[16] J. Bauer, G. Drescher, and M. Illig, “Surface tension, adhesion and wet-ting of materials for photolithographic process,”J. Vacuum Sci. Technol.,pt. B, vol. 14, pp. 2485–2492, July–Aug. 1996.

[17] Y. D. Shikhmurzaev, “Spreading of drops on solid surfaces in a qua-sistatic regime,”Phys. Fluids, vol. 9, pp. 266–275, Feb. 1997.

[18] A. L. Yarin and D. A. Weiss, “Impact of drops on solid surfaces: Self-similar capillary waves, and splashing as a new type of kinematic dis-continuity,” J. Fluid Mechanics, vol. 283, pp. 141–173, Jan. 1995.

[19] C. H. R. Mundo, M. Sommerfeld, and C. Tropea, “Droplet-wall colli-sions: Experimental studies of the deformation and breakup process,”Int. J. Multiphase Flow, vol. 21, pp. 151–173, Apr. 1995.

[20] S. H. Davis and D. M. Anderson, “Spreading and evaporation of liquiddrops on solids,”ASME AMD Surface-Tension-Driven Flows, vol. 170,pp. 1–7, 1993.

[21] M. Pasandideh-Fard, Y. M. Qiao, S. Chandra, and J. Mostaghimi, “Cap-illary effects during droplet impact on a solid surface,”Phys. Fluids, vol.8, pp. 650–659, Mar. 1996.

[22] R. P. Mandal, J. C. Grambow, T. C. Bettes, D. R. Sauer, E. Gurer, andE. R. Ward, “Method of uniformly coating a substrate,” U.S. Patent no.5 670 210, Sept. 23, 1997.

[23] G. Perçin, G. G. Yaralioglu, and B. T. Khuri-Yakub, “Micromachineddroplet ejector arrays,”Rev. Sci. Instrum., vol. 73, pp. 4385–4389, Dec.2002.

Gökhan Perçin (S’92–M’00) was born in Ezine,Çanakkale, Turkey. He received the B.S. degree inelectrical and electronics engineering from BilkentUniversity, Ankara, Turkey in 1994. He received theM.S. degree in electrical engineering in 1996, andthe Ph.D. degree in electrical engineering, in 2002both from Stanford University, Stanford, CA.

He worked at Numerical Technologies, Inc., SanJose, CA, as a Senior Engineer to develop optical mi-crolithography software for rule-based optical prox-imity correction and photoresist modeling. He also

worked in Microbar Inc., Sunnyvale, CA, as the director of MEMS technologyand applications. He is Co-Founder and Board Director of ADEPTIENT, PaloAlto, CA. He has years of experience in the areas of fluid ejection, microflu-idics, micromachining, micromachined electromechanical systems, ultrasonicactuators, ultrasound transducers, biomedical imaging, software development,sensors, and actuators. He has authored over ten scientific publications and hasbeen principal inventor of three issued patents.

Dr. Perçin is a member of IEEE Ultrasonics, Ferroelectrics, and FrequencyControl Society, IEEE Electron Devices Society, The Electrochemical Society,and the American Vacuum Society.

Butrus T. Khuri-Yakub (S’70–S’73–M’76–SM’87–F’95) was born in Beirut, Lebanon. Hereceived the B.S. degree in 1970 from the AmericanUniversity of Beirut, the M.S. degree in 1972 fromDartmouth College, and the Ph.D. degree in 1975from Stanford University, Stanford, CA, all inelectrical engineering.

He joined the research staff at the E. L. GinztonLaboratory, Stanford University in 1976 as aResearch Associate. He was promoted to a SeniorResearch Associate in 1978 and to a Professor of

Electrical Engineering (Research) in 1982. He has served on many universitycommittees in the School of Engineering and the Department of Electrical En-gineering. Presently, he is the Deputy Director of the E. L. Ginzton Laboratory.He has been teaching both at the graduate and undergraduate levels for over15 years, and his current research interests includein situ acoustic sensors(temperature, film thickness, resist cure, etc.) for monitoring and control ofintegrated circuits manufacturing processes, micromachining silicon to makeacoustic materials and devices such as airborne and water immersion ultrasonictransducers and arrays, fluid ejectors, and ultrasonic nondestructive evaluationand acoustic imaging and microscopy. He has authored over 300 publicationsand has been principal inventor or co-inventor of 54 issued patents.

Dr. Khuri-Yakub is a Senior Member of the Acoustical Society of America,and a Member of Tau Beta Pi. He is an Associate Editor ofResearch inNondestructive Evaluation, a journal of the American Society for Nonde-structive Testing. He received the Stanford University School of EngineeringDistinguished Advisor Award, June 1987, and the Medal of the City ofBordeaux for contributions to NDE, 1983.