Ramp-rate effects in transient enhanced diffusion: a quasi-analytical examination

Summary form only given. Some experimental evidence has accumulated in recent years to support the use of "spike anneal" temperature trajectories with very fast heating and cooling rates for making ultrashallow junctions by ion implantation. Improved device properties have been claimed using heating rates of 400°C/s or more. This procedure supposedly optimizes junction depth and sheet resistance by reducing transient-enhanced diffusion (TED) of the dopant. However, rigorous theoretical justification for using such fast ramps has been spotty in the literature. Since the design and use of fast-ramp annealing tools will require substantial investments by equipment manufacturers and IC manufacturers alike, it is important to obtain a clear picture for the occurrence and potential magnitude of such effects. We employ a quasi-analytical approach to the complex, interrelated phenomena that govern TED, confirmed by numerical simulations for boron based on FLOOPS. Our approach employs a set of rate equations and associated parameters (especially for kick-in and kick-out) that has much firmer kinetic grounding than is generally seen in the literature. We find that increasing the ramp rate β narrows the diffused profile according to β^-12/ as long as TED runs out before the top of the spike. Improvement decreases progressively as TED runs out after the top. High ramp rate also affects dopant activation indirectly (and in a complicated way) through interstitial clustering kinetics. The optimum value of β depends sensitively on the temperature trajectory near the peak and the details of the clustering model used. The details of the initial temperature stabilization step can influence subsequent profile evolution in a significant way under certain circumstances.