Search

Developing doping technologies for the channel region of CMOS devices for 180 nm technology and beyond requires accurate profiles of electrically active dopants. We have recently improved the accuracy of spreading resistance profiles on sub-80 nm source/drain boron implants. Previous SRP analyses ignored the fact that, in ultra-shallow source/drain structures, the sheet resistance dominates measured spreading resistance values. Because these SRP measurements are made at the edge of a doped layer, this implies that an additional sheet resistance edge correction should be applied to measured spreading resistance data. In this paper, we extend these improved spreading resistance corrections to the analysis of a variety of channel doping technologies. We examine sub-80 nm source/drains made with boron implants and study the implications of the new analysis. We also compare channel doping (V/sub T/) profiles with profiles obtained by a highly repeatable MOS-CV method.

The spreading resistance profiling (SRP) method has undergone significant improvements over the last five years, enhancing the techniques’ capability for profiling sub-100 nm junctions. Newly developed procedures for conditioning probes, preparing beveled surfaces, and processing the raw data now yield a high degree of success. We believe that the sample preparation technique determines the quality of the bevel surface and thus plays a key role in the repeatability and sensitivity of SRP measurements. Although some previous work has suggested that the diamond ground beveled surfaces used for these measurements involve significant surface damage and hence, a high surface state density, little work has been done to quantify bevel surface quality. In this article, Hg gate capacitance–voltage (C–V) and current–voltage (I–V) have been used for the first time to determine the electrical properties of the diamond-beveled surfaces. Diamond particle sizes ranging from 0.05 to 0.25 μm were used to bevel p silicon s...

Si (100) wafers were boron implanted using an xRLEAP ion implanter at energies of 0.2, 1.0, and 3.0 keV all at a dose of 1×1014 cm−2. Advanced spreading resistance (SR) measurements were performed on the as-implanted samples to obtain high-resolution profiles of the damaged layer produced during implantation. The measured spreading resistance was highest at, or near, the surface on all samples. The 1.0 and 3.0 keV implants also showed minima in the SR near the substrate. The as-implanted samples were examined with secondary ion mass spectrometry to determine the dopant profile. Forward and reverse Schottky I–V measurements were also done on the same as-implanted samples with a mercury probe. The forward Schottky I–V curves for the 0.2 and 1.0 keV samples showed nonideal Schottky diode behavior. The 3.0 keV sample was similar to that of a good quality Schottky diode and had an ideality factor of 1.28. The reverse Schottky I–V data were processed to obtain the effective dielectric constants for each sample;...