On the analysis of the activation mechanisms of sub-melt laser anneals

Abstract

In order to fabricate carrier profiles with a junction depth (∼15 nm) and sheet resistance value suited for sub-32 nm Si-CMOS technology, the usage of sub-melt laser anneal is a promising route to explore. As laser annealed junctions seem to outperform standard anneal approaches, a detailed assessment of the basics of laser induced activation seem appropriate. In this work the electrical activation is studied from a comparison between the dopant profiles as measured by Secondary Ion Mass Spectrometry, and the electrically active fraction as extracted from sheet resistance and mobility measurements. The latter is based on a large variety of techniques. For the sheet resistance we use conventional Four-Point Probe (FPP), Variable Probe Spacing (VPS), contactless junction photo voltage (JPV), Micro Four-Point Probe (M4PP) and an optical technique, namely Model Based Infra-red spectroscopic Reflectrometry (MBIR). For the sheet carrier density and sheet mobility extraction we use conventional Hall (without cloverleaf van der Pauw patterning, to reflect the need for fast turn-round sheet measurements), MBIR, and a recently developed new Hall-like capability using M4PP. By recognizing the interaction between the various parameters as they are not completely independent, it is possible to test the consistency of the various methods and to identify potential short comings. This concept is applied to the activation behavior of low and high implanted Boron doses and indicates that the obtained electrically active concentration level as well as the concurrent mobility is dependent on the dopant concentration level. This implies that the activation of B through the laser anneal process in the explored temperature–time space is governed by kinetic processes (i.e. the dissolution of B–I pairs) and not by the (temperature related) solid solubility.