Optical fiber meta-tips

Abstract

We report on the first demonstration of a proof-of-principle optical fiber 'meta-tip', which integrates a phase-gradient plasmonic metasurface on the fiber tip. For illustration and validation purposes, we present numerical and experimental results pertaining to various prototypes implementing generalized forms of the Snell's transmission/reflection laws at near-infrared wavelengths. In particular, we demonstrate several examples of beam steering and coupling with surface waves, in fairly good agreement with theory. Our results constitute a first step toward the integration of unprecedented (metasurface-enabled) light-manipulation capabilities in optical-fiber technology. By further enriching the emergent 'lab-on-fiber' framework, this may pave the way for the widespread diffusion of optical metasurfaces in real-world applications to communications, signal processing, imaging and sensing.

Keywords: Fiber optics; metasurfaces; plasmonics; wavefront manipulation.

Figure 1

Illustration of the idea and geometry. (a) Pictorial sketch (not in scale) of an optical-fiber MT. A plasmonic metasurface (with details magnified in the inset) is laid on the tip of an optical fiber, covering the entire core region. The metasurface impresses a linear-phase profile (constant gradient, along the x-direction) in the wavefront of a given component of an impinging beam. This yields splitting in the transmitted beam, with an ordinary (co-polarized) component propagating along the incidence direction, and an anomalous component (with generally different polarization) undergoing a phase-gradient-induced steering of an angle. A similar phenomenon (not shown for better visibility) occurs in reflection as well. (b) Illustration of the generalized Snell's refraction/reflection laws in Equation (1). (c) Geometry (not in scale) of the unit cell: a rectangular nanohole (rotated of 45° in the x – y plane) milled in a gold layer.

Figure 2

Results from the design procedure (MT₁ and MT₃). (a, b) Numerically-synthesized phase and magnitude distributions, respectively, of the transmission coefficient pertaining to the single nanoholes in the supercell (shown on top) of the MT₁ design (with parameters as given in Table 1), for the co-polarized (blue square markers) and cross-polarized (red circle markers) components, assuming an infinite periodic array of period l_x = l_y = 1 μm, under normally-incident x-polarized plane-wave illumination at λ = 1.56 μm. Element #1 is chosen as phase reference. Continuous curves are guides to the eye only. (c, d) Same as above, but for MT₃ design.

Figure 3

Example of a fabricated sample. (a) SEM image of the MT₃ sample, displaying the entire fiber cross-section. (b, c) Two magnified details, showing the entire metasurface and two unit cells, respectively.

Figure 4

Far-field characterization (MT₁ and MT₃) without polarization control. (a) Simulated electric field-intensity profiles (at z = 8 mm and y=0) of the ordinary and anomalous beams (blue and red curves, respectively), for MT₁ sample (with parameters as given in Table 1). Results are obtained by averaging the co-polar and cross-polar responses, respectively, under normally-incident x- and y-polarized illuminations at λ = 1.56 μm. (b) Measured field-intensity map at z = 8 mm. (c) Transverse cuts at y = 0 comparing the measured (black-solid curve) and simulated (magenta-dashed curve) results. Numerical results are obtained by averaging the total electric-field intensities for normally-incident x- and y-polarized illuminations. The structure is considered as infinitely-periodic along y, whereas, along the x-direction, a finite-size ∼20 μm is assumed, together with a Gaussian-beam taper (with waist size of 5μm) in the illumination. (d–f) Same as above, but for MT₃ sample.

Figure 5

Far-field characterization (MT₃) with polarization control. (a–c) Measured field-intensity maps at z = 5.9 mm for MT₃ sample (with parameters as given in Table 1) pertaining to the y-, oblique (45°) and x-polarized components, respectively, and assuming a y-polarized incident field (see also Supplementary Movie 1 for a finer sampling of the selected transmitted polarization).

Figure 6

Examples of fabricated samples. (a, b) SEM images (magnified details) of the MT₅ and benchmark (phase-gradient-free) samples, respectively.

Figure 7

Perspectives in sensing applications (MT₅). (a, b) Measured and simulated, respectively, reflectivity spectra in the absence (red-dashed curves) and presence (blue-solid curves) of a 40 nm overlay of SiO_x. (c, d) Simulated electric-field magnitude map over a supercell (at y=0, nearby the MT) for the co-polarized and cross-polarized component, respectively, at λ = 1.46 μm, assuming a normally-incident x-polarized plane-wave illumination. (e, f) Corresponding longitudinal cuts at x = 0.265 μm. The inset shows the MT₅ supercell (with parameters as given in Table 1). Fields are normalized with respect to the incident-field amplitude.

Figure 8

Perspectives in sensing applications (benchmark MT). (a, b) Measured and simulated, respectively, reflectivity spectra in the absence (red-dashed curves) and presence (blue-solid curves) of a 40-nm-thick overlay of SiO_x. (c, d) Simulated electric-field magnitude map over a supercell (at y=0, nearby the MT) for the co-polarized and cross-polarized component, respectively, at λ = 1.62 μm, assuming a normally-incident x-polarized plane-wave illumination. (e, f) Corresponding longitudinal cuts at x = 0.265 μm. The inset shows the benchmark supercell. Fields are normalized with respect to the incident-field amplitude.

Figure 9

Resonant field distributions (MT₅ and benchmark). (a, b) Simulated electric-field magnitude maps over a supercell at the MT interface z = 50nm, for the MT₅ (at λ = 1.46 μm) and benchmark (at λ = 1.62 μm) configurations, respectively, assuming a normally-incident x-polarized plane-wave illumination. (c, d) Corresponding transverse cuts at y=0. Fields are normalized with respect to the incident-field amplitude.