Past exam of the materials science course of the University of Cambridge / 2019 / Paper 1 / 2

Features only a few tens of nanometres wide lie below the ordinary visible-light diffraction limit, approximately $\lambda /(2\mathrm{NA})$. The microscope cannot resolve individual rods, and unpolarized illumination does not directly reveal the orientation-dependent optical response. Scanning electron microscopy and atomic force microscopy can resolve the surface pattern. Transmission electron microscopy is another possible choice if a suitably thin specimen can be prepared.

An optically isotropic film changes the phase of both transverse polarization components equally, so crossed polarizers still extinguish the transmitted light. The etched rod-and-gap composite has form birefringence: its effective permittivity depends on the direction of the electric field, even when the constituent material is intrinsically isotropic. In a simple subwavelength lamellar model,The precise effective-medium formula changes with rod geometry, but the anisotropy remains. Resolving incident light along the two principal directions produces a phase retardation. An ideal lossless specimen between crossed polarizers transmitswhere $I_P$ is the intensity after the first polarizer. Extinction occurs when a principal direction follows either polarizer; transmission is strongest near $45^{\circ }$.

At distance $x$ from its tip, the wedge thickness is $t_q=x\tan {\theta }_q$. Its optical path difference is thereforeThe perpendicular slow directions make the two retardances subtract. A white-light black compensation fringe is the zero-order cancellation, so the film retardance has magnitude approximately $393\mathrm{nm}$, with its slow direction along the rods. This interpretation assumes the observed fringe is the white-light compensation fringe; a single monochromatic dark fringe alone would have an integer-order ambiguity. Without the wedge, the wavelength dependence of ${\sin }^2(\pi \Gamma /\lambda )$ selectively transmits visible wavelengths and gives the film its interference colour.

The local slow direction follows the radial rods. Between crossed polarizers, the angular factor is ${\sin }^2(2\phi )$, so radial directions parallel to either polarizer are dark. The pattern is a Maltese cross in polarized-light microscopy, with four bright sectors separated by dark arms parallel to the polarizer directions. Its colours depend on the local optical path difference; tapering rods or changing fill fraction can make this vary with radius. The cross belongs to the local radial-axis pattern and does not imply a conoscopic interference figure.