Published online by Cambridge University Press: 22 February 2011
We implanted <100> silicon 200mm wafers with 20keV 11B+ to a fluence of 5×1015 atoms/ cm2 using beam currents from 1-7mA, which produced flux of about 50-350µA/cm2. The implant temperature of all wafers rose no more than five degrees above room temperature, regardless of flux. Cross sectional TEM images (as-implanted) of the highest flux samples revealed a continuous amorphous layer from the implanted surface to a depth of about 530Å. The high flux and <30°C implantation temperature allowed amorphous layer formation even with this moderate boron fluence, as was suggested by Jones, et.al.1. We observed a strong dependence of as-implanted damage on boron flux, as previously reported by Eisen and Welch2. After 900°C, 20 sec RTA, the highest flux samples had 50% lower sheet resistance than the lowest flux samples, due to better activation, as observed in SRP. When a 1050°C, 15 sec RTA was employed, this sheet resistance and activation dependence on flux disappeared. Cross sectional TEM images revealed that the size and number of the Type II end of range defects , which were centered near the amorphous and crystalline as-implanted interface, in the highest flux samples were smaller than the Type 1 dislocation loops centered about the peak disorder in the lowest flux samples after RTA. SIMS and SRP profiles indicated that transient enhanced diffusion during the 900°C, 20 sec RTA may have been reduced in the highest flux samples. Based on these observations and on previous reports, we conclude that sufficiently high flux during room temperature boron implantation will produce a continuous amorphous layer with doses that are appropriate for p-type source/drain formation. The amorphous layer will produce improved activation and damage annealing behavior in subsequent RTA, particularly as the RTA temperature is reduced.
Full text views reflects PDF downloads, PDFs sent to Google Drive, Dropbox and Kindle and HTML full text views.