Sharper Focus for a Radially Polarized Light Beam

We experimentally demonstrate for the first time that a radially polarized field can be focused to a spot size significantly smaller [$0.16(1){\lambda }^2$] than for linear polarization ($0.26{\lambda }^2$). The effect of the vector properties of light is shown by a comparison of the focal intensity distribution for radially and azimuthally polarized input fields. For strong focusing, a radially polarized field leads to a longitudinal electric field component at the focus which is sharp and centered at the optical axis. The relative contribution of this component is enhanced by using an annular aperture.

Figure 1

Experimental setup to measure the focal intensity distribution. MO: Leica plan, apo $100\times$, 0.9; OD: optical diode; HWP: half-wave plate; TL: telescope lenses; PH: pinhole; PC: polarization converter; FL: focusing lens; NCFPI: nonconfocal Fabry-Perot interferometer; CL: collimating lens; AS: aperture stop; M: four mirrors for polarization insensitive deflection (details not shown); MD: monitor diode; PD: photodiode partially covered with gold-zinc alloy.

Figure 2

Defocus series for a radially (a) and azimuthally (b) polarized input field. The insets illustrate how the polarization of the input beam varies spatially across the beam cross section. Note that identical initial intensity distributions lead to different distributions at focus.

Figure 3

Measured projections (a),(c) and cross sections through the tomographically reconstructed intensity distributions (b),(d) in the focal plane for an azimuthally (a),(b) and a radially (c),(d) polarized input field. Open squares: experimental data; calculated curves: solid black line: total field; dashed line: transverse field; solid triangles: longitudinal field.

Figure 4

Projection of (a) and cross section through (b), the tomographically reconstructed intensity distribution for an azimuthally polarized input field focused using an annular aperture ($d_i=3.0\mathrm{m}\mathrm{m}$). Open squares: experimental data; solid black line: calculated curve for the total field.

Figure 5

Measured projection of (a) and cross section through (b), the tomographically reconstructed intensity distribution for a radially polarized input field focused using an annular aperture ($d_i=3.3\mathrm{m}\mathrm{m}$). The calculated power contained in the longitudinal field amounts to $72.8\%$ of the total beam power. Open squares: experimental data; calculated curves: solid black line: total field; dashed line: transverse field; solid triangles: longitudinal field.