Density of Thin Films of Vacuum Evaporated Metals

Abstract

IT is well known that the density of a material depends on its method of preparation and subsequent treatment. Values of the densities of bulk materials treated in various ways are given in reference tables¹. Increasing use is, however, being made of materials in the form of thin evaporated films, and in some experiments, such as those involving measurements of transmission or energy loss with atomic particles, the superficial density of the film is a more significant parameter than its linear thickness. It is not always possible to measure the superficial density sufficiently accurately by direct weighing. A knowledge of the density of the evaporated material is needed for the superficial density of the film to be calculated from its linear thickness, measured, for example, by multiple beam interferometry^2–4. Observations⁵ on silver films in which the mass of silver was estimated by a chemical colorimetric method accurate to 3 per cent gave agreement between the density of the film and the bulk density. For aluminium, however, the oscillating quartz crystal thickness monitor described by Lawson⁶ gave an observed thickness sensitivity some 4 per cent greater than the theoretical sensitivity. The thickness sensitivity, measured as the shift in frequency/unit linear film thickness, is directly proportional to the density of the evaporated material. The early measurements by Kahlbaum⁷ seem to have been made with massive samples recast from vacuum distilled material. We report here the results of direct determinations of the density of some metallic films. The linear thicknesses were determined by multiple beam interferometry and the superficial densities were measured gravimetrically.