I. Introduction

Microfabrication of electronic and mechanical structure at the submillimeter scale is typically a time-consuming and expensive process. Lithographic techniques for silicon micromachining, used to fabricate integrated circuits and MEMS, typically take several weeks to go from drawings to completed chips, and require expensive facilities and extreme processing conditions. An alternate approach in which multiple small volumes of metallic, semiconducting, or insulating material are deposited at computer-defined positions could enable the all-additive fabrication of such devices on a much faster and less expensive basis. Techniques that involve expelling small droplets of molten metal onto a substrate [1]– [5], however, have met with mixed success, primarily because of the difficulty of adhering droplets to previously solidified layers [2]. Other problems include oxidation of the liquid metal [1] and the difficulty of fabricating a droplet-expulsion mechanism compatible with the melting temperatures of most high-quality metals beyond low-temperature solders [1], [2]. Other approaches for droplet deposition, called ink-jet printing, as a route to silicon-like device fabrication have included printing metallo–organic decomposition inks [6], [7], dry powders [8], organic light-emitting materials [9], [10], photonics and solders [1], and printing resin binders into successive layers of loose powder [11]. However, to date, such processes have been limited in terms of electrical conductivity, feature complexity and thickness, resolution, or material quality, and none have been able to fabricate active MEMS devices.