Role of spatial coherence in the Goos-Hänchen shift

Summary form only given. Amongst the various nonspecular phenomena that affect the reflection of a beam of light by a smooth surface, the Goos-Hänchen (GH) shift is certainly the most investigated one. Experimental studies of this effect have been performed so far only with fully spatial coherent beams.Recently theorists have addressed the question whether the spatial coherence of the light source does or does not affect the GH shift. Wang et al. investigated this problem and found that the GH shift depends on spatial coherence, decreasing along with the latter [1,3]. Aiello et al. find that the GH shift is not affected by the spatial coherence of the incident light [2, 4]. In this work [5] we report the experimental comparison of the GH shift suffered from a spatially coherent Gaussian beam and Gaussian Schell-model (GS) beams with low spatial coherence. Beams at a wavelength of 635 nm are totally reflected from a BK7 prism. With a quadrant detector we measure the displacement (D) of the laser beam in the plane of incidence when the polarization of the beam is switched from p to s. Our experimental results for the spatial coherent beam are reported as solid dots in Fig. 1. Experimental results for three different GS beams are reported as well. GS beams are generated from three planar sources with 1 \ e intensity distribution width s and 1 \ e degree of spatial coherence width g (see Fig. 1). Data are compared with the theoretical GH shift for a fully coherent Gaussian beam (solid line in Fig. 1). We do not find any decrease of the GH shift along with spatial coherence, whereas in our most extreme case (triangles in Fig 1) Wang et al predict a reduction bigger than a factor five. Our data support theories predicting that the spatial coherence does not affect the GH shift.