Large-area wide-angle spectrally selective plasmonic absorber

Chihhui Wu, Burton Neuner, III, Gennady Shvets, Jeremy John, Andrew Milder, Byron Zollars, and Steve Savoy

Phys. Rev. B 84, 075102 – Published 1 August 2011

Abstract

A simple metamaterial-based wide-angle plasmonic absorber is introduced, fabricated, and experimentally characterized using angle-resolved infrared spectroscopy. The metamaterials are prepared by nano-imprint lithography, an attractive low-cost technology for making large-area samples. The matching of the metamaterial’s impedance to that of vacuum is responsible for the observed spectrally selective “perfect” absorption of infrared light. The impedance is theoretically calculated in the single-resonance approximation, and the responsible resonance is identified as a short-range surface plasmon. The spectral position of the absorption peak (which is as high as $95 %$ ) is experimentally shown to be controlled by the metamaterial’s dimensions. The persistence of “perfect” absorption with variable metamaterial parameters is theoretically explained. The wide-angle nature of the absorber can be utilized for subdiffraction-scale infrared pixels exhibiting spectrally selective absorption/emissivity.

Received 15 April 2011

DOI:https://doi.org/10.1103/PhysRevB.84.075102

Authors & Affiliations

Chihhui Wu, Burton Neuner, III, and Gennady Shvets^*

Department of Physics, The University of Texas at Austin, Austin, Texas 78712, USA

Jeremy John, Andrew Milder, Byron Zollars, and Steve Savoy

Nanohmics, Inc. 6201 East Oltorf, Suite 400, Austin, Texas 78741, USA

^*gena@physics.utexas.edu

Article Text (Subscription Required)

Click to Expand

References (Subscription Required)

Click to Expand

Issue

Vol. 84, Iss. 7 — 15 August 2011

Reuse & Permissions

Access Options

Author publication services for translation and copyediting assistance advertisement

Images

Figure 1
(Color online) Schematic of the plasmonic absorber with (a) parallel strips and (b) square patches placed above the ground plate. (c, d) Reduced radiative loss rate due to destructive interference between ground plate reflection and magnetic dipole radiation. (c) Strong magnetic dipoles are induced when $G$ is small; (d) negligible magnetic dipoles are induced when $G$ is large.Reuse & Permissions
Figure 2
(Color online) Resistive ( $ω_{i Q}$ ) and radiative ( $ω_{i R}$ ) loss rates of the eigenmodes as functions of (a) the dielectric gap size ( $G$ ), (b) metal strip thickness ( $D$ ), and (c) metal strip width ( $W$ ). The parameters are varied around $L = 350$ nm, $W = 250$ nm, $D = 20$ nm, and $G = 20$ nm. Equal resistive and radiative loss rates correspond to critical coupling and perfect absorption. Insets: Dependence of the resonant wavelengths of the corresponding eigenmodes. (d) Peak absorbance calculated from eigenmode (markers) and driven simulation (solid lines), with the parameters in the same range as (a–c).Reuse & Permissions
Figure 3
(Color online) Angular dependence of the plasmonic resonance responsible for “perfect” absorption. (a) Peak absorbance remains above $80 %$ for all incidence angles $θ$ . (b) Real and (c) imaginary parts of the eigenfrequency as a function of the wave number $k_{y} = ω / c * \sin (θ)$ in the periodicity direction. The imaginary part is separated into radiative and Ohmic loss rates.Reuse & Permissions
Figure 4
(Color online) (a) SEM image of the strip absorber structure. (b) Simulated field profile at the resonance. Color: $| H |$ , arrow: E field. (c) Measured and (d) simulated absorbance with polarization perpendicular (solid lines) and parallel (dashed lined) to the strips. The dimensions are [ $L$ , $W$ ] $=$ [300 nm,230 nm] (blue), [330 nm,250 nm] (green), and [450 nm,350 nm] (red). $D = 30$ nm and $G = 22$ nm for all three cases. Incident beams are $25^{\circ}$ P polarized for both experiments and simulations. The dotted lines in (d) are predicted absorbances for square patch arrays with identical parameters and normal incidence.Reuse & Permissions
Figure 5
(Color online) Angular-resolved absorption spectra $A (θ, λ)$ for the strip-based plasmonic absorber shown in Fig. 4. Experimental absorbance with (a) S- and (b) P-polarized incidence [illustrated in Fig. 1a] is given in the upper panel, and calculated absorbance with (c) S- and (d) P-polarized incidence is given in the lower panel.Reuse & Permissions
Figure 6
(Color online) (a) Absorbance by a finite number of unit cells illuminated by a Gaussian beam with an intensity FWHM of 2.6 $μ$ m. (b) Peak absorbance versus patterned-area width. Dashed line: Fractional overlap between the patterned area and the laser beam. Inset: A simulation of eight unit cells (only half are shown) under the incoming Gaussian beam. Color: out-of-page magnetic field, arrows: Poynting flux.Reuse & Permissions

Physical Review B

covering condensed matter and materials physics