Impact of multiple sub-melt laser scans on the activation and diffusion of shallow Boron junctions

Abstract

Sub-melt laser annealing is a promising technique to achieve the required sheet resistance and junction depth specifications for the 32 nm technology node and beyond. In order to obtain a production worthy process with minimal sheet resistance variation at a macroscopic and microscopic level, careful process optimization is required. While macroscopic variations can easily be addressed using the proper spatial power compensation it is more difficult to completely eliminate the micro scale non-uniformity which is intimately linked to the laser beam profile, the amount of overlaps and the scan pitch. In this work, we will present micro scale sheet resistance uniformity measurements for shallow 0.5 keY B junctions and zoom in on the underlying effect of multiple subsequent laser scans. A variety of characterization techniques are used to extract the relevant junction parameters and the role of different implantation and anneal parameters will be explored. It turns out that the observed sheet resistance decrease with increasing number of laser scans is caused on one hand by a temperature dependent increase of the activation level, and on the other hand, by a non-negligible temperature and concentration dependent diffusion component.

Accurate micro Hall Effect measurements on scribe line pads

“…Thus, for a constant Hall mobility we find the activation degree of the implanted dose to be 14 % and 21 % in the pad centre and on average 18 % and 20 % at the pad edges for the wafer without and with AL, respectively. This is in reasonable good agreement with previously reported values for a non-patterned wafer [10].…”

Width‐Dependent Sheet Resistance of Nanometer‐Wide Si Fins as Measured with Micro Four‐Point Probe

“…Two separate regions of the wafer were subsequently laser‐annealed with seven laser scans, respectively at 1150 and 1250 °C. These millisecond annealing schemes activate the B dopants with minimal diffusion …”

Use of p- and n-type vapor phase doping and sub-melt laser anneal for extension junctions in sub-32 nm CMOS technology

“…For comparison, the chemical boron concentration profile of an implanted material (0.5 keV, 1 × 10 15 cm -2 ) after similar LA conditions is given in Fig. 2(c) and (d) [10]. When the number of laser scans increases, the dopant is further driven into the substrate, leading to lower Rs, as shown in Fig.…”

“…However, the exact nature of these inactive species and the microscopic mechanism of their formation have to be clarified. [10,12,13] …”