Addendum Addendum to ‘ ‘ Large negative Goos – Hanchen shift at metal surfaces ” , [ Opt . Comm . 276 ( 2007 ) 206 ]

In the above communication, we have investigated in details the possibility of observing a large negative lateral Goos–Hanchen (GH) shift from p-polarized light at long wavelength incident onto a bare metal surface, at angles close to grazing incidence. The shift could be up to a few wavelength (e.g. 20 lm at a wavelength of 3390 nm onto silver) and the possible presence of a ‘‘pseudo Brewster angle” will not hinder the observability of such an effect. This effect has since been verified experimentally via reflection of IR light from a gold surface [1]. In reviewing the literature, aside from pointing out that our results were in consistency with those hinted at from previous analysis of surface backward waves [2] and of negative energy flow at interface [3], we should have also noted the first explicit calculation by Wild and Giles of such GH shifts at a silver surface which had obtained a (albeit small) negative shift in the order of one wavelength ( 0.5 lm) [4]. In addition, we would like to emphasize that such large negative GH shift at a metal surface can easily be understood by examining the well-known elementary boundary conditions for the fields across an interface, as applied to the discussion of the continuity of the various components of the Poynting vector in [3]. In fact, if we let the x–z plane be the plane of incidence and apply the following boundary conditions for the normal (N) and tangential (T) field components at the vacuum–metal interface (assuming non-magnetic materials so that B = H and neglecting damping): E0N 1⁄4 eEN; E0T 1⁄4 ET; H 0N 1⁄4 HN; H 0T 1⁄4 HT ð1Þ