Vortrag

The talk will cover work that was undertaken to quantify the thickness of thermal oxide on silicon to meet a target of 1% for the International Technology Roadmap for Semiconductors (ITRS) and also to help the development of quantitative surface analysis. The work was launched as a new initiative in surface analysis under the Consultative Committee for Amount of Substance (CCQM) and has completed a pilot study which is continuing as a Key Comparison.

Many methods are used in the semiconductor industry to measure oxides and, at the start of this work although CRMs were available at 50 nm thickness, comparisons at 4 nm showed standard deviations of 40%! Oxide samples were grown in both the US and Europe at FAB research facilities specialising in ultra-thin gate oxide development, on both (100) and (111) surfaces. Exposed to the air, these surfaces adsorb both hydrocarbons and water. Methods were developed to control these. In the pilot study, measurements were made by medium energy ion scattering, Rutherford backscattering spectrometry, secondary ion mass spectrometry, neutron reaction analysis, X-ray photoelectron spectroscopy, ellipsometry, X-ray reflectance, neutron reflectance and transmission electron microscopy. The results from these methods will be outlined. Analysis of these data show how the oxide thickness may be measured to 1% accuracy and how, using a series of developments for X-ray photoelectron spectroscopy, this may be realised in practice. Also, with development of the approach, a similar accuracy may be attained for the carbonaceous contamination.