In semiconductor plasma immersion ion implantation (PIII) applications such as the synthesis of silicon-on-insulator by hydrogen PIII and ion cut, only ions arriving at the top surface of the sample stage are important. The ions implanted into the other surfaces of the sample chuck actually not only decrease the efficiency of the power supply and plasma source but also give rise to metallic contamination. In addition, low energy ions introduced by the initial plasma sheath propagation, pulse rise time, and pulse fall time introduce a large surface hydrogen concentration that creates surface damage and affects the wafer bonding efficacy. We have theoretically demonstrated direct-current PIII (DC-PIII) which retains the x–y immersion characteristic while simultaneously reducing this low energy ion component, obviating the need for the expensive power modulator, and extending the voltage ceiling that is no longer limited by the vacuum chamber and power modulator. In this article, we describe our hydrogen DC-PIII experiments using a conducting grid placed between the wafer stage and a multipolar electron cyclotron resonance plasma source. The grounded grid stops the propagation of the plasma sheath, thereby removing the vacuum chamber size limitation. Ions are formed in the plasma sustained by an external plasma source above the grid and accelerated through the lower zone to be implanted into the wafer biased by only a dc power supply. Atomic force microscopy, hydrogen forward scattering, and secondary ion mass spectrometry analyses indicate uniform hydrogen PIII into a 100 mm silicon wafer and the surface hydrogen component is indeed reduced significantly compared to conventional pulsed PIII.

This content is only available via PDF.
You do not currently have access to this content.