Ultrathin SiO 2 on Si II. Issues in quantification of the oxide thickness

An analysis is made of various quantification issues concerning the analysis of ultrathin layers of SiO 2 on (100) and (111) polished Si surfaces. For analysis of the oxide thickness, a simple equation is generally used involving two parameters; the attenuation length of photoelectrons in the oxide and the ratio, R o , of the intensities of the Si 2p peak from bulk thermal SiO 2 and from pure Si. An analysis of previously reported measurements of the attenuation length gives an average value of only 6% less than the theoretical value. However, careful measurements of R o , via two routes, indicate consistently that a value of 0.88 ± 0.03 should be used rather than the calculated value of 0.53 ± 0.05. This difference may arise through systematic uncertainties in the values for the relevant inelastic mean free paths, the silicon dioxide density and the shake‐up contributions. Previously reported experimental values of R o range from 0.67 to 0.87. Uncertainties also arise from intensity variations caused by the crystal structure of the substrate. These are mapped and a position ‘A’ is found where further work is best conducted. For the (100) surface, A is 34° from the surface normal in an azimuth midway at 22.5° between the [010] and [011] azimuths. For the (111) surface at A is 25.5° from the surface normal in the [10 1 ] azimuth. Data for much of the present work are for the (100) surface at an angle of emission of 27° at position ‘B’ at 28.5° from the surface normal in the [110] azimuth, which is equivalently good but may degrade for spectrometers with high angular resolution. If the same equation is used for calculating the thickness, position B leads to a calculated thickness that is 4% less than that measured for an average orientation, whereas data acquired for normal emission lead to a value 18% lower, and those measured at A are 2% higher. Measurements of the carbonaceous contamination confirm earlier conclusions that the contamination is better described using data for an average polymer than for glassy carbon. © Crown copyright 2002. Published by John Wiley & Sons, Ltd.