No abstract

Contrary to previous beliefs, electron‐beam‐induced positive charges in insulators persist where created for at least several weeks without discernible movement. Formerly, coating with a thin metal overlay was thought to allow the charge to leak away. Coating with a conductor is shown to shield electric fields from affecting the imaging probe, but to remove no charge from the specimen. A new technique is introduced for the evaluation of the properties of electron‐beam‐induced positive charges in metal‐oxide‐semiconductor (MOS) capacitors. MOS structures were subjected to partial area exposure in a scanning electron microscope. These exposures resulted in the creation of areas of localized positive charge within the oxide, which was observed as steps in the capacitance‐voltage data. A systematic study was performed. It related the exposed area to the step height and the amount of induced charge to the voltage shift of the step. A model describing the observed phenomenon is presented, followed by a comparison of theoretical and experimental results. The progress of the charge over time was studied by performing capacitance‐voltage analysis 30 min after electron beam exposure and up to 4 weeks later.

Currently, one of the more promising silicon on insulator (SOI) technologies involves the use of ion beams to synthesize buried dielectric layers. Considerable research has been conducted in the study of buried SiO2 by oxygen implantation (separation by implanted oxygen, or SIMOX technology). The formation of SOI structures by ion implantation of nitrogen has been investigated less extensively. Still, a substantial amount of information has been accumulated (1-8). More interest has been generated after device operation was demonstrated on substrates using buried nitride (2). Earlier work has shown that a dose of 0.75 • 1018 cm -2 implanted at 160-200 keV is sufficient to produce a continuous buried layer of silicon nitride while maintaining good crystal quality in the silicon overlayer. To achieve this, a 1200~ 2h postimplantation anneal is necessary. Because of the polycrystalline nature of the resulting buried layer, leakage has been found to be a common problem (9). It has been suggested that it may be necessary to use higher doses or to use oxygen and nitrogen dual implantation (3). Nesbit et aI. (6) studied the buried