A further calibration of the harwell series II Bi-implanted RBS standards

Abstract

An extensive investigation of the Harwell series II RBS standards, using selected samples and edge pieces from each of the six implantation runs, shows that the previously reported [1] implantation problem is much more complex than had been assumed. The following conclusions can now be drawn: o (i) Only two of the implantation runs produced Bi contents within ±1% of the series I value; most of the other runs exhibit significantly higher Bi levels. (ii) Even within a single implantation run, gradual variations as large as ±2.5% occur across the set of four wafers. However, over the dimensions of a single sample (1 × 1 cm), the uniformity was always better than ±1%. In view of the above complexities, each series 11 standard requires cross-calibration whenever an absolute accuracy of better than ± 2.5% is desired.