Optical anisotropy of semiconductor nanowires beyond the electrostatic limit

Semiconductor nanowires with diameters below 100 nm exhibit distinct polarization anisotropies that cannot be explained in the electrostatic limit. Comparing experiments with calculations reveals the effects of diameter-wavelength ratio, material dispersion, and local refractive index. All these parameters need to be taken into account to fully understand optical anisotropies in the size regime between extremely thin nanowires and bulklike materials.

Figure 1

(Color online) [(a) and (b)] Sketches of the wide-field luminescence and combined dark-field/laser microscope, respectively. (c) Absorption (red, triangle) and PL (black, square) spectra of the SNCs in solution (starting material to prepare the NWs) and typical PL spectrum of a single NW (green, circle). (d) TEM image of NWs and (e) Rayleigh scattering image of NWs aligned in a PVA film, inset: distribution of the angle between the long axis of individual NWs and the alignment direction in the PVA film.

Figure 2

(Color online) [(a)–(f)] Wide-field PL images of an individual NW for different excitation polarizations as indicated by the arrow (raw data). The scale bar is $5\mu \text{m}$. [(g) and (h)] PL intensity under linearly polarized excitation or detection, respectively, of the same individual NW shown in (a)–(f). The angle is measured with respect to the long axis of the NW. [(i) and (j)] Rayleigh-scattering intensity of an individual NW under polarized excitation and detection, respectively. Squares are experimental data, and curves are best fits to ${\text{sin}}^2$ functions.

Figure 3

(Color online) PL intensity of NW ensembles aligned in polymer films under: (a) under linearly polarized excitation and (b) linearly polarized detection. [(c) and (d)] Rayleigh scattering of these films with a polarizer in the excitation and detection paths, respectively. Blue squares are experimental data, red curves marked with circles are best fits to ${\text{sin}}^2$ functions, and green curves marked with triangles are the expected polarization anisotropies based on average single NW polarization anisotropy convoluted with the orientational distribution from the inset in Fig. 1e. The shaded areas represent expected standard deviation based on average single NW data.

Figure 4

(Color online) Calculated polarization anisotropy in (a) excitation and (b) emission as function of the NW diameter-to-vacuum wavelength ratio. Blue dashed curves represent the electrostatic limit without a substrate, red solid curves show FDTD calculations including the substrate, and black dashed-dotted curves are FDTD calculations without the substrate. (c) Difference of the polarization anisotropies in emission and excitation, i.e., difference between the solid curves in (a) and (b). Insets display the simulated geometries and snapshots of the field amplitude in the steady state.