Ceramic hybrid electromechanical systems (CHEMS) have been proposed to overcome some of the disadvantages while retaining most of the advantages of microelectromechanical systems (MEMS). Whereas MEMS are fabricated mostly by micromachining of silicon and have typical feature sizes of the order of microns or smaller, CHEMS could be fabricated on ceramic substrates by a wider variety of techniques and would have typical feature sizes ranging from tens of microns to millimeters. Depending on specific applications, CHEMS could serve as alternatives or complements to MEMS. CHEMS could be readily incorporated, along with integrated circuits and other microscopic components, into ceramic-based hybrid multilayer packages (e.g., multichip modules).

While the development of MEMS has been an important achievement in miniaturization, it turns out that in many practical applications, MEMS are too small to provide the required sensitivity as sensors or to provide the required forces or strokes as actuators. MEMS also suffer from sticton, squeeze-film damping, and damage induced by surface tension in liquids during processing. In addition, silicon is often not the substrate material of choice for applications in which there are requirements for electrically or thermally insulating substrates, low capacitance, resistance to corrosion, or hermetic sealing.

The proposal to develop CHEMS originated from the realization that many of the mechanical problems of MEMS could be solved more readily by fabrication of packaged microelectromechanical devices with dimensions intermediate between those of silicon-based microdevices and those of conventional macroscopic electromechanical devices. Sensors and actuators at the proposed CHEMS mesoscale could be made stronger and could be made to respond over dynamic ranges wider than those of silicon-based microdevices. Seals could be improved and strokes lengthened. Even so, CHEMS would still be small enough to fit into compact packages along with electronic integrated circuits.

In the development of CHEMS, it will be possible to take advantage of the mature technology already available for manufacture of ceramic hybrid structures in the electronics industry. There is an immense data base on ceramic materials with a wide variety of mechanical and electrical characteristics, including such sensor/actuator materials as piezoelectrics and ferroelectrics. Ceramic hybrids and multichip modules, and modern processes for manufacturing them, share many characteristics with those of silicon-based MEMS. Ceramic-hybrid technology affords the means to make laminated assemblies of ceramics, metals, and glasses that can be patterned, fired, and etched to produce three-dimensional structures. Inasmuch as silicon-based MEMS and electronic circuits are already typically integrated on ceramic substrates or headers, the fabrication of CHEMS should pose no obstacle to integration, nor should it entail additional cost. The completed systems would be of the same masses and volumes as those of packaged silicon microfabricated devices, but would have greater capabilities because of the larger sizes of the active mechanical components.

The figure illustrates an example of fabrication of a multilayer CHEMS that would include a metal cantilever over a rectangular hole plus metal layers connected to each other electrically and mechanically. Fabrication would be accomplished by use of the low-temperature cofired ceramic (LTCC) process. The starting materials for the layers would be 250-*m-thick "green" (that is, not yet fired) ceramic tapes, typically composed of 40 to 60 percent Al2O3and the balance of filler materials.

Via holes for mechanical registration and electrical contact would be stamped into the tapes by use of computer-aided design and automated cutting tools. The via holes for electrical connection would be filled with metal. The rectangular central hole would be filled with a sacrificial dielectric to support the cantilever to be formed in the next step. Metal layers would be screened onto the broad surfaces of the tapes, forming the cantilever among other metal features. The metal-patterned tapes would be stacked and aligned by use of pins through the registration holes. The stack would be laminated at a pressure of 3 kpsi (21 MPa) and temperature of 70 °C. Next, the laminated structure would be heated to 500°C to drive out volatiles. The structure would be fired at 850 °C to set the ceramic. Finally, to free the cantilever, the sacrificial dielectric would be removed from the central rectangular hole by wet and/or dry chemical etching.

This work was done by Linda Miller, Michael Hecht, Martin Buehler, Amin Mottiwala, Beverly Eyre, and Indrani Chakraborty of Caltech for NASA's Jet Propulsion Laboratory.

In accordance with Public Law 96-517, the contractor has elected to retain title to this invention. Inquiries concerning rights for its commercial use should be addressed to

Technology Reporting Office
JPL
Mail Stop 122-116
4800 Oak Grove Drive
Pasadena, CA 91109
(818) 354-2240