(446a) Defect Engineering For Ultrashallow Junctions Using Surfaces | AIChE

(446a) Defect Engineering For Ultrashallow Junctions Using Surfaces

Authors 

Kwok, C. - Presenter, University of Illinois
Vaidyanathan, R. - Presenter, University of Illinois, Urbana - Champaign
Yeong, S. - Presenter, Department of Chemical and Biomolecular Engineering, National University of Singapore
Srinivasan, M. P. - Presenter, National University of Singapore
Colombeau, B. - Presenter, Chartered Semiconductor Manufacturing Ltd, Singapore
Chan, L. - Presenter, Chartered Semiconductor Manufacturing Ltd, Singapore


Formation of extremely shallow pn junctions with very low electrical resistance is a major stumbling block to the continued down scaling of microelectronic devices. Recent work in our laboratory has shown that the behavior of defects within silicon can be changed significantly by controlling the chemical state at the surface. Certain chemical treatments of the surface induce it to act as an active ?sink? for point defects that removes Si interstitials selectively over impurity interstitials, leading to less dopant diffusion and better electrical activation. The present work demonstrates such effects experimentally for several dopants such as boron, arsenic and phosphorous in both crystalline and Ge pre-amorphized silicon wafers. Moreover, such active surfaces dramatically reduce the number of end-of-range defects observed after annealing. In the case of boron, a continuum model for boron diffusion and activation has been developed to quantify the surface effects under a wide range of annealing protocols ranging from hours to milliseconds in duration. Two-dimensional simulations based on this model indicate that the beneficial effects of active surfaces in the source-drain region extend laterally to the surface toward the channel region of a device as well as perpendicularly to the surface into the bulk. In a separate but parallel mechanism, fixed charges created at surfaces or interfaces can interact with charged defects in the bulk. Simulations suggest that this electrostatic mechanism results in a deeper junction and dopant pile-up near the surface/ interface. The work also discusses possibilities for creating such active surfaces in real RTP conditions.