Answers are grounded in this paper and cite their sources.
An analysis is made of the amount of carbonaceous contamination remaining on silicon wafers with a thin oxide overlayer. Wafers are either left “as-received” or deliberately contaminated with a thumbprint and then subjected to one of 42 different cleaning procedures with the remaining carbon level monitored by x-ray photoelectron spectroscopy (XPS). The XPS measurement of the carbon level in terms of a thickness in nm, using the traditional equation, depends sensitively on the assumed form of the carbon; the thickness for an average polymerlike contamination being computed as 1.67 times the thickness for a glassy carbon layer for a given spectrum. However, if the measurement is defined in terms of carbon atoms/m2, this sensitivity is removed. Alternatively, if absolute measurements are used, the sensitivity in terms of the thickness in nm is also reduced and the data are consistent with the results for the average polymer. Of the cleaning methods, ultraviolet radiation with ozone is best but this affects the substrate by increasing the oxide thickness by approximately 0.2 nm and so is not recommended for work where the oxide thickness should remain unaltered. The best group of methods that leave the oxide unaltered are those involving an overnight soak in a high purity solvent such as isopropyl alcohol (IPA), chloroform, or heptane, followed by an ultrasonic agitation for 1 min in fresh solvent, a rinse in further fresh solvent, and drying in an argon jet. This reduces the final carbon level to thicknesses, in terms of an average polymer, of around 0.16 nm, with IPA giving the best result for the selected solvents. This level may be further reduced, by heating on a hot plate in air to 325 °C for 15 min, to a final thickness of 0.12 nm. Cleaned samples that are then stored in glass or Fluoroware polypropylene containers retain much of their cleanness and the carbonaceous contamination does not exceed 0.26 nm during the 100 days of the study.
No takes yet. Share an insight, caveat, or question.
Seah et al. (2003) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: