Verification of carrier density profiles derived from spreading resistance measurements by comparing measured and calculated sheet resistance values

In the past, resistivity and carrier density profiles derived from spreading resistance measurements have been satisfactorily verified by comparing a doped layer’s measured sheet resistance, ρs, with a sheet resistance value calculated from the on-bevel resistivity profile. However, this verification process has been found to fail in the case of ultrashallow (<100 nm) source/drain implants. This paper examines all possible sources of the discrepancy between ρs-measured and ρs-calculated, and concludes that the lack of a sheet resistance edge correction in the standard multilayer SRP analysis is the dominant factor. The paper also discusses the role played by bevel surface damage, on-bevel carrier diffusion, and variations in carrier mobility in high dopant density regions.