In a recent Comment, Decca et al. [Phys. Rev. A 79, 026101 (2009); arXiv:0809.3576] discussed the origin of the anomalies recently reported by us in Phys. Rev. A 78, 036102(R) (2008); arXiv:0812.0028 . Here we restate our view, corroborated by their
considerations, that quantitative geometrical and electrostatic characterizations of the conducting surfaces (a topic not discussed explicitly in the literature until very recently) are critical for the assessment of precision and accuracy of the demonstration of the Casimir force and for deriving meaningful limits on the existence of Yukawian components possibly superimposed to the Newtonian gravitational interaction.
We report on measurements of forces acting between two conducting surfaces in a spherical-plane configuration in the 35 nm-1 micrometer separation range. The measurements are obtained by performing electrostatic calibrations followed by a residual an
alysis after subtracting the electrostatic-dependent component. We find in all runs optimal fitting of the calibrations for exponents smaller than the one predicted by electrostatics for an ideal sphere-plane geometry. We also find that the external bias potential necessary to minimize the electrostatic contribution depends on the sphere-plane distance. In spite of these anomalies, by implementing a parametrixation-dependent subtraction of the electrostatic contribution we have found evidence for short-distance attractive forces of magnitude comparable to the expected Casimir-Lifshitz force. We finally discuss the relevance of our findings in the more general context of Casimir-Lifshitz force measurements, with particular regard to the critical issues of the electrical and geometrical characterization of the involved surfaces.
We have performed precision electrostatic calibrations in the sphere-plane geometry and observed anomalous behavior. Namely, the scaling exponent of the electrostatic signal with distance was found to be smaller than expected on the basis of the pure
Coulombian contribution and the residual potential found to be distance dependent. We argue that these findings affect the accuracy of the electrostatic calibrations and invite reanalysis of previous determinations of the Casimir force.
