Fabrication of SiCN ceramic MEMS using injectable polymer-precursor technique

Download Fabrication of SiCN ceramic MEMS using injectable polymer-precursor technique

Post on 04-Jul-2016

222 views

Category:

Documents

3 download

Embed Size (px)

TRANSCRIPT

<ul><li><p>Fabrication of SiCN ceramic MEMS using injectablepolymer-precursor technique</p><p>Li-Anne Liew*, Wenge Zhang, Victor M. Bright, Linan An,Martin L. Dunn, Rishi Raj</p><p>Department of Mechanical Engineering, NSF Center for Advanced Manufacturing and Packaging of Microwave,</p><p>Optical, and Digital Electronics (CAMPMODE), University of Colorado, Boulder, CO 80309-427, USA</p><p>Abstract</p><p>In this paper, a novel and cost-effective technology for the fabrication of high-temperature MEMS based on injectable polymer-derived</p><p>ceramics is described. Micro-molds are fabricated out of SU-8 photoresist using standard UV-photolithographic processes. Liquid-phase</p><p>polymers are then cast into the molds and converted into monolithic, fully-dense ceramics by thermal decomposition. The resultant ceramic,</p><p>based on the amorphous alloys of silicon, carbon and nitrogen, possess excellent mechanical and physical properties for high-temperature</p><p>applications. This capability for micro-casting is demonstrated in the fabrication of simple single-layered, high aspect ratio SiCN</p><p>microstructures. A polymer-based bonding technique for creating more complex three-dimensional structures is also presented.</p><p># 2001 Published by Elsevier Science B.V.</p><p>Keywords: MEMS; High-temperature; Polymer; Ceramic</p><p>1. Introduction</p><p>MEMS for extreme temperature environments have</p><p>attracted much attention due to their many potential appli-</p><p>cations, such as optical MEMS for high power laser applica-</p><p>tions [1] and microcombustors [2] for MEMS power</p><p>sources. However, the high temperature environments</p><p>remain a significant challenge to current MEMS technology.</p><p>The fabrication of MEMS for high-temperature applications</p><p>is a two-fold problem: selecting suitable refractory materials</p><p>and developing appropriate microfabrication techniques.</p><p>Traditional microfabrication processes, such as surface</p><p>micromachining and LIGA, rely on polysilicon and plas-</p><p>tic/nickel as structural materials, respectively. These mate-</p><p>rials cannot operate at high temperatures for extended</p><p>periods of time. At the same time, state-of-the-art ceramics</p><p>that are designed for high-temperature environments cannot</p><p>be easily processed using existing microfabrication techni-</p><p>ques. Moreover, CVD of SiC [3], a technique currently</p><p>under development for high-temperature MEMS, is time-</p><p>consuming and expensive. Also, the planar nature of</p><p>CVD prevents the fabrication of complex three-dimensional</p><p>structures. Therefore, the development of new materials and</p><p>appropriate microfabrication techniques for high-tempera-</p><p>ture MEMS is of both scientific and practical interest to the</p><p>MEMS community. Fig. 1 compares the aspect ratios</p><p>achieved and maximum operating temperature of tradition-</p><p>ally-microfabricated MEMS with that of the polymer-</p><p>derived SiCN MEMS that are described in this paper.</p><p>2. Injectable polymer derived ceramics</p><p>A novel fabrication technique for high-temperature</p><p>MEMS using an injectable polymer-derived SiCN ceramic</p><p>has been developed. This new technology is based on the</p><p>recently developed polymer-derived ceramics, which are</p><p>bulk ceramics fabricated by the thermal decomposition of</p><p>compacted crosslinked polymer powders [4]. The polymer-</p><p>derived ceramics are amorphous alloys of silicon, carbon,</p><p>and nitrogen (SiCN) which remain thermally stable up to</p><p>15008C. The compositions of the new ceramics can bevaried through the use of different polymer precursors,</p><p>and can be tailored to produce SiCN with excellent thermal</p><p>and mechanical properties. Table 1 compares the physical</p><p>properties of SiCN with those of Si and SiC. Youngs</p><p>modulus, Poissons ratio, and density are in the same range</p><p>as those of SiC and Si. The creep resistance of SiCN is</p><p>Sensors and Actuators A 89 (2001) 6470</p><p>* Corresponding author. Tel.: 1-303-492-3842;fax: 1-303-492-3498; URL: http://mems.colorado.edu.E-mail address: li.liew@colorado.edu (L.-A. Liew).</p><p>0924-4247/01/$ see front matter # 2001 Published by Elsevier Science B.V.PII: S 0 9 2 4 - 4 2 4 7 ( 0 0 ) 0 0 5 4 5 - 8</p></li><li><p>comparable to that of SiC and Si3N4 [5,6], while its oxida-</p><p>tion resistance exceeds that of the same materials [7]. In</p><p>addition, the thermal shock resistance of SiCN appears very</p><p>promising for high-temperature applications.</p><p>SiCN may be obtained from liquid- or powder-based</p><p>polymer precursors. However, the SiCN obtained from</p><p>the powder-route shows relatively low strength and hardness</p><p>due to the high porosity of powder-derived ceramics in</p><p>general (typically 10 vol.% porosity). In addition, pow-der-processing cannot be easily integrated into existing</p><p>microfabrication techniques. Therefore, liquid polymer pre-</p><p>cursors were used to develop a novel micro-casting techni-</p><p>que for the fabrication of SiCN MEMS structures. Fig. 2</p><p>compares the microstructure, strength, and hardness of SiCN</p><p>samples obtained through powder- and casting-routes.</p><p>Fig. 1. Diagram of microstructure aspect ratio against maximum use</p><p>temperature for different MEMS materials and processes.</p><p>Table 1</p><p>Comparison of physical properties of SiCN, Si, and SiC</p><p>SiCN Si SiC</p><p>Density (g/cm3) 2.20 2.33 3.17</p><p>E modulus (GPa) 158 163 405</p><p>Poissons ratio 0.18 0.22 0.14</p><p>CTE 106 (K) 0.5 2.5 3.8Hardness (GPa) 15 11.2 30</p><p>Strength (MPa) 250 175 418</p><p>Toughness (MPa m1/2) 3.5 0.9 46</p><p>Fig. 2. (a) Scanning electron micrographs showing the microstructure of polymer-derived SiCN from powder-route and (b) from casting-route. This figure</p><p>shows that the material resulting from casting is fully dense in that no defects can be seen despite a higher magnification than that in (a). (c) Comparison of</p><p>strength and hardness (which are not directly related to each other) of samples obtained from powder- and casting-routes.</p><p>L.-A. Liew et al. / Sensors and Actuators A 89 (2001) 6470 65</p></li><li><p>3. Microfabrication of SiCN MEMS</p><p>Fig. 3 outlines the process flow in the fabrication of SiCN</p><p>MEMS. First, a mold is fabricated using standard photo-</p><p>lithographic techniques. The liquid polymer precursor is</p><p>then cast into the mold, and the mold and polymer precursor</p><p>are then heated or thermal-set at 2508C to solidify thepolymer. After thermal-setting, the polymer becomes a</p><p>transparent solid, and may be separated from the mold if</p><p>suitable techniques are used. After thermal-setting, the</p><p>polymer part is crosslinked by heating to 4008C underisostatic pressure. After crosslinking, the polymer becomes</p><p>infusible, remaining transparent. In the final stage (pyroly-</p><p>sis), the crosslinked polymer part is heat-treated at10008Cto convert it to a monolithic ceramic part.</p><p>The commercially-available Ceraset (CerasetTM, Lanxide</p><p>Company, USA) was used as the starting polymer. A catalyst</p><p>(Dicumyl peroxide, Aldrich Chemical, Milwaukee WI,</p><p>USA) is added to lower the thermosetting temperature to</p><p>1408C, so that it is compatible with photoresist processing.The mold was made using SU-8 photoresist [8] from Micro-</p><p>Chem Corp. SU-8 is a well-known negative, epoxy-type</p><p>photoresist based on EPON SU-8 resin. Its key features are</p><p>its ability to produce high-aspect ratio structures and low</p><p>optical absorption in the near-UV range, which leads to</p><p>straight vertical sidewalls. Other advantages of using SU-8</p><p>as the mold are: (1) standard UV-photolithography can</p><p>produce high-aspect ratio structures; (2) batch fabrication;</p><p>(3) the SU-8 photoresist decomposes during pyrolysis, enabling the release of the structures using the lost mold</p><p>technique.</p><p>Fig. 4 outlines the micro-casting technique in more detail.</p><p>First, the photoresist is spun onto a substrate (a). Then, the</p><p>photoresist is patterned using standard UV-lithography and</p><p>developed, producing cavities of desired shapes (b). The</p><p>liquid precursor (Ceraset) is then cast into the cavities by</p><p>spinning, and the wafer is thermal-set in an oven for 20 min.</p><p>At the end of this step, the Ceraset is solid and there is a thin</p><p>layer of it covering the entire wafer (c). This thin top layer of</p><p>Ceraset is polished-off (d). The wafer is then crosslinked</p><p>under isostatic pressure (e), and during pyrolysis the SU-8</p><p>decomposes and the SiCN part no longer adheres to the</p><p>substrate (f). The final result is a free-standing high aspect-</p><p>ratio microstructure (g).</p><p>Of course, this schematic shows only one structure being</p><p>fabricated; in reality, multiple structures are fabricated at</p><p>once on a wafer, which is one of the attractions to micro-</p><p>casting using photoresist molds.</p><p>Fig. 5(a) shows an SU-8 mold for a micro-gear. Fig. 5(b)</p><p>shows the same mold when filled with the polymer pre-</p><p>cursor. For small cavities (less than 20 mm wide), often theliquid does not fully fill the cavity due to the surface tension</p><p>of the liquid and/or air bubbles in the mold cavity. This can</p><p>be solved by placing the filled mold in vacuum to remove</p><p>most of the air bubbles, and then pressing a thin film of</p><p>Teflon over the wafer to force the liquid into the cavities.</p><p>This film is left on the mold during thermosetting, and alsoFig. 3. General processing steps in the fabrication of injectable polymer-</p><p>derived ceramics.</p><p>Fig. 4. Fabrication process for SiCN MEMS.</p><p>66 L.-A. Liew et al. / Sensors and Actuators A 89 (2001) 6470</p></li><li><p>serves to minimize the thickness of the extra top layer of</p><p>Ceraset that needs to be polished-off.</p><p>4. Crosslinking and pyrolysis</p><p>The key to the success of the casting-route is the applica-</p><p>tion of isostatic pressure during crosslinking. The heat-</p><p>treatment during crosslinking generates gaseous by-pro-</p><p>ducts. This out-gassing may cause the formation of micro-</p><p>cracks which disintegrate the sample [4]. In the casting</p><p>approach, the applied isostatic pressure provides an obstacle</p><p>to the nucleation of bubbles and microcracks. The gaseous</p><p>by-products would need to build up sufficient pressure to</p><p>overcome the applied pressure before bubbles and micro-</p><p>cracks can nucleate. Due to the open structure of the</p><p>polymer, the gases will diffuse out of the sample before</p><p>they can build up a pressure that is high enough to overcome</p><p>the applied pressure. Thus, transparent and defect-free cross-</p><p>linked polymers can be obtained by the application of</p><p>isostatic pressure during crosslinking. The fact that no</p><p>defects can be seen in the final ceramics indicates that</p><p>the infusible polymer network has enough strength to with-</p><p>stand the further heat-treatment during pyrolysis.</p><p>The crosslinked samples are then pyrolyzed at 10008C for8 h to convert the polymer into a ceramic. Very low heating/</p><p>cooling rates are used: the heating rate is about 18C/min to4008C and 258C/h to 10008C; the cooling rate is 18C/min to</p><p>room temperature. Higher heating rates result in the produc-</p><p>tion of most of the gaseous by-products at a narrow tem-</p><p>perature range (H2 at 6007008C, and CH4 at 6008C), whichcould generate defects in the pyrolyzed sample and degrade</p><p>the mechanical strength of the structures. Examples of</p><p>microstructures fabricated from this casting process are</p><p>shown in Fig. 6.</p><p>5. Demolding</p><p>Initially during pyrolysis, the SU-8 photoresist mostly</p><p>decomposed but left a thick, hard residue on the walls of the</p><p>samples, as can be seen in Fig. 6 and also in Fig. 7(a). These</p><p>SU-8 by-products, being chemically resistant to etchants and</p><p>strongly attached to the SiCN, could not be easily removed</p><p>through either chemical or physical means. Attempts to</p><p>burn-off the residue at 10008C in an oxygen atmospherewere also not successful. The by-products are the result of</p><p>reactions between the liquid Ceraset and the polymer mold</p><p>at the interface, producing unknown compounds. Many</p><p>attempts were made to de-mold the parts before pyrolysis</p><p>in order to prevent the reactions.</p><p>One drawback of using SU-8 is that it is not easily</p><p>removed, even when using the solvent provided by the</p><p>manufacturer. It was found that the SU-8 could be stripped</p><p>if the solvent was heated to about 908C. However, beforepyrolysis, the Ceraset is highly reactive and chemically</p><p>Fig. 5. (a) Photoresist mold (200 mm diameter and 75 mm thick); (b) mold filled with Ceraset.</p><p>Fig. 6. (a) Gear (200 mm diameter and 45 mm thick) made from the mold shown in Fig. 8; (b) cantilever beam (137 mm thick); (c) tensile test sample (142 mmthick).</p><p>L.-A. Liew et al. / Sensors and Actuators A 89 (2001) 6470 67</p></li><li><p>reacts with many polymers. Various photoresists that could</p><p>be easily stripped were experimented with as alternatives to</p><p>SU-8, but the highly reactive nature of Ceraset still proved to</p><p>be the limiting factor. This problem was finally solved</p><p>through a combination of an extended bake and extended</p><p>UV-exposure of the SU-8, following the mold fabrication</p><p>but prior to casting. This ensured that the SU-8, being a</p><p>branched polymer, was completely crosslinked before the</p><p>Ceraset was cast into it. Results were encouraging. SU-8</p><p>thus remained the material of choice for mold fabrication,</p><p>and the lost mold technique was pursued following this</p><p>post-fabrication treatment of the mold prior to casting.</p><p>Fig. 7(b) shows the marked improvement in surface quality</p><p>following this treatment.</p><p>Despite the success of the post-fabrication treatment of</p><p>the mold in producing SiCN parts that were clean and</p><p>residue-free, it is still advantageous to de-mold prior to</p><p>pyrolysis, especially if polymer-based bonding (decribed</p><p>later in this paper) is to be used to fabricate multi-layer</p><p>structures.</p><p>6. Size limitations</p><p>We have studied the size range of samples that can be</p><p>successfully cast. It has been found that the samples will</p><p>crack at the crosslinking and pyrolysis stages if the length</p><p>exceeds a certain critical value, which is a function of the</p><p>sample thickness, being larger for thicker samples. Ideally,</p><p>there should not be any limit on the sample size, which is</p><p>proven by casting in macro-sized Teflon molds that have</p><p>been conventionally machined. In this case, the thermally-</p><p>set samples are released from the Teflon mold and cross-</p><p>linked and pyrolyzed as free-standing samples. The diameter</p><p>of these samples can be up to 10 mm (which is limited by the</p><p>size of our furnace). The reason for the observed scale limit</p><p>for the micro-cast samples is explained with the aid of Fig. 8.</p><p>The Ceraset experiences volume shrinkages of 5 and 25%</p><p>during crosslinking and pyrolysis, respectively. However, at</p><p>the same time the Ceraset adhesion to the silicon substrate</p><p>results in a tensile stress, s, in Ceraset structure. This tensilestress in turn induces a shear stress, t, at the silicon/Ceraset</p><p>Fig. 7. (a) SiCN cantilever beam encased in SU-8 residue; (b) SiCN gear tooth (75 m thick), obtained from SU-8 mold that had undergone post-fabricationtreatment to prevent residue formation.</p><p>Fig. 8. Induced stresses during heat-treatment of Ceraset on silicon.</p><p>68 L.-A. Liew et al. / Sensors and Actuators A 89 (2001) 6470</p></li><li><p>interface. Assuming a disk-shaped sample, the tensile and</p><p>shear stresses are related by</p><p>s r2ht</p><p>where s increases with the progress of heat-treatment. Let tcand sc be the critical values for which the sample will peel-off from the substrate and fracture, respectively. Then, if the</p><p>ratio r/h is too large (which corresponds to thin samples),</p><p>then sc will be reached first and sample will crack. Ifthe sample is thick,...</p></li></ul>