JOURNAL OF MICROELECTROMECHANICAL SYSTEMS, VOL. 12, NO. 6, DECEMBER 2003 761 Etch Rates for Micromachining Processing—Part II Kirt R. Williams, Senior Member, IEEE, Kishan Gupta, Student Member, IEEE, and Matthew Wasilik Abstract—Samples of 53 materials that are used or potentially can be used or in the fabrication of microelectromechanical systems and integrated circuits were prepared: single-crystal silicon with two doping levels, polycrystalline silicon with two doping levels, polycrystalline germanium, polycrystalline SiGe, graphite, fused quartz, Pyrex 7740, nine other preparations of silicon dioxide, four preparations of silicon nitride, sapphire, two preparations of aluminum oxide, aluminum, Al/2%Si, tita- nium, vanadium, niobium, two preparations of tantalum, two preparations of chromium, Cr on Au, molybdenum, tungsten, nickel, palladium, platinum, copper, silver, gold, 10 Ti/90 W, 80 Ni/20 Cr, TiN, four types of photoresist, resist pen, Parylene-C, and spin-on polyimide. Selected samples were etched in 35 different etches: isotropic silicon etchant, potassium hydroxide, 10:1 HF, 5:1 BHF, Pad Etch 4, hot phosphoric acid, Aluminum Etchant Type A, titanium wet etchant, CR-7 chromium etchant, CR-14 chromium etchant, molybdenum etchant, warm hydrogen peroxide, Copper Etchant Type CE-200, Copper Etchant APS 100, dilute aqua regia, AU-5 gold etchant, Nichrome Etchant TFN, hot sulfuric phosphoric acids, Piranha, Microstrip 2001, acetone, methanol, isopropanol, xenon difluoride, HF HO vapor, oxygen plasma, two deep reactive ion etch recipes with two different types of wafer clamping, SF plasma, SF O plasma, CF plasma, CF O plasma, and argon ion milling. The etch rates of 620 combinations of these were measured. The etch rates of thermal oxide in different dilutions of HF and BHF are also reported. Sample preparation and information about the etches is given. [1070] Index Terms—Chemical vapor deposition (CVD), etching, evap- oration, fabrication, materials processing, micromachining. I. INTRODUCTION W HEN designing a microfabrication process, the etch rate of each material to be etched must be known. Knowing the etch rates of other materials that will be exposed to the etch, such as masking films and underlying layers, enables an etch process to be chosen for good selectivity (high ratio of etch rate of the target material to etch rate of the other material)—if one exists. While several large literature-review compilations of etches that target specific materials have been made [1], [2], these only report etch rates in some cases, and rarely have corre- sponding selectivity information. This paper provides such in- formation, expanding on an earlier paper [3] to give 620 etch rates of 53 materials in 35 etches that have been used or may Manuscript received June 3, 2003; revised October 1, 2003. Subject Editor A. J. Ricco. K. R. Williams was with Agilent Laboratories, Agilent Technologies, Palo Alto, 94303 CA USA. He is currently a private consultant at 185 Willowbrook Dr., Portola Valley, CA 94028 USA (e-mail: kirt_williams@ieee.org). K. Gupta was with with Agilent Laboratories, Agilent Technologies, Palo Alto, CA 94303 USA. He is now at 804 Gregory Ct., Fremont, CA 94359 USA (e-mail: kishang@ieee.org). M. Wasilik is with the Berkeley Sensor & Actuator Center, University of Cal- ifornia at Berkeley, Berkeley, CA 94720-1770 USA. Digital Object Identifier 10.1109/JMEMS.2003.820936 be used in future fabrication of microelectromechanical systems (MEMS) and integrated circuits (ICs) (approximately 50 etch rates measured in the earlier paper have been included in this one). These data allow the selection of new combinations of structural material, underlying material, and etchant for micro- machining. Table I summarizes the etches tested, abbreviated names for the etches, and the target materials for each. Table II lists etch rates of Si,Ge, SiGe, and C in the SI units of nm/min (not /min as in the earlier tables) [3]. Table III covers films and wafers that are primarily silicon dioxide, produced under many different conditions. Table IV is on silicon nitride and aluminum oxide. Table V covers the metals Al, Ti, V, Nb, Ta, and Cr. Table VI continues with the metals Mo, W, Ni, Pd, Pt, Cu, Ag, Au, alloys 10 Ti/90 W, 80 Ni/20 Cr, and compound TiN. Finally, Table VII gives etch rates of organics: photoresists, a resist pen, and a spin-on polyimide. Section II of this paper lists the materials etched, their prepa- ration, and some uses or potential uses in MEMS and ICs. Sec- tion III describes the preparation and applications of the wet and dry etches that were studied, as well as some key experimental results. Section IV describes etch-rate measurement techniques, and Section V discusses the results. II. SAMPLE PREPARATION The preparation of the samples in the etch-rate tables is de- scribed below, listed by the labels (in italics) used across the tops of the tables. All coated materials were deposited on 100-mm- diameter silicon wafers. For the isotropic silicon etchant, potas- sium hydroxide, and a few other etches, the wafers were first coated with LPCVD silicon nitride so that etches would not pen- etrate into the silicon or attack the back side of the wafer. In several cases, similar materials were prepared using dif- ferent methods (e.g., wafer form, PECVD, LPCVD, and ion- milled silicon dioxide; annealed and unannealed films) to study and emphasize the effect on their etching characteristics. Existing or potential MEMS applications are given for the materials. Many of the materials were discussed in more detail previously [3]. A. Silicon, Germanium, SiGe, and Carbon (100) Si Low-Doped Wafer: Single-crystal silicon, (100) ori- entation, phosphorus-doped n-type, resistivity of 3–40 -cm, grown with the Czochralski (CZ method). Single-crystal silicon is the standard starting material for bulk micromachining. Float-Zone Si Wafer: Single-crystal silicon, (100) orienta- tion, undoped, grown with the float-zone (FZ) method for a high resistivity of -cm. Float-zone wafers have been used as substrates in RF MEMS application to reduce eddy-current loss. 1057-7157/03$17.00 © 2003 IEEE Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE. This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.