Temperature Dependence of the Casimir Effect

I. Brevik, J. B. Aarseth

Published 2005-11-04, updated 2005-12-30Version 2

In view of the increasing accuracy of Casimir experiments, there is a need for performing accurate theoretical calculations. Using accurate experimental data for the permittivities we present, via the Lifshitz formula applied to the standard Casimir setup with two parallel plates, accurate theoretical results in case of the metals Au, Cu and Al. Both similar and dissimilar cases are considered. Concentrating in particular on the finite temperature effect, we show how the Casimir pressure varies with separation for three different temperatures, T={1, 300, 350}K. The metal surfaces are taken to be perfectly plane. The experimental data for the permittivities are generally yielding results that are in good agreement with those calculated from the Drude relation with finite relaxation frequency. We give the results in tabular form, in order to facilitate the assessment of the temperature correction which is on the 1% level. We emphasize two points: (i) The most promising route for a definite experimental verification of the finite temperature correction appears to be to concentrate on the case of large separations (optimum around 2 micrometres); and (ii) there is no conflict between the present kind of theory and the Nernst theorem in thermodynamics.