Broadband Reflectionless Metasheets: Frequency-Selective Transmission and Perfect Absorption

Energy of propagating electromagnetic waves can be fully absorbed in a thin lossy layer, but only in a narrow frequency band, as follows from the causality principle. On the other hand, it appears that there are no fundamental limitations on broadband matching of thin resonant absorbing layers. However, known thin absorbers produce significant reflections outside of the resonant absorption band. In this paper, we explore possibilities to realize a thin absorbing layer that produces no reflected waves in a very wide frequency range, while the transmission coefficient has a narrow peak of full absorption. Here we show, both theoretically and experimentally, that a thin resonant absorber, invisible in reflection in a very wide frequency range, can be realized if one and the same resonant mode of the absorbing array unit cells is utilized to create both electric and magnetic responses. We test this concept using chiral particles in each unit cell, arranged in a periodic planar racemic array, utilizing chirality coupling in each unit cell but compensating the field coupling at the macroscopic level. We prove that the concept and the proposed realization approach also can be used to create nonreflecting layers for full control of transmitted fields. Our results can have a broad range of potential applications over the entire electromagnetic spectrum including, for example, perfect ultracompact wave filters and selective multifrequency sensors.

Popular Summary

Electromagnetic waves play a pivotal role in our everyday lives; they are what we see and how we communicate. The manipulation of waves has led to a myriad of applications and has revolutionized society through communications, medicine, and entertainment. New electromagnetic and optical devices with properties that exceed those achieved by conventional lenses and mirrors are currently being developed, and planar films of advanced artificial engineered materials have opened up new degrees of freedom to guide and bend electromagnetic waves. Such films, so-called metasurfaces, are capable of modifying the intensity, direction, and even the shape of wave beams passed through them. However, metasurfaces are all characterized by the same major drawback: They efficiently manipulate transmitted radiation of a desired frequency, but they do not fully transmit electromagnetic waves of other frequencies. This limitation creates a perceptible shadow, while the overwhelming majority of applications require wave-controlling devices that do not create a shadow and are transparent to radiation at nonoperational frequencies. In this work, we demonstrate the first realization of such devices that are free of shadows.

We employ a single-layer metasheet that contains inclusions that can be polarized both electrically and magnetically. These inclusions, whose sizes are on the order of the impinging microwave radiation, are smooth helices with one or two turns. We design and experimentally test metasurfaces that totally absorb incident radiation at the operational frequency and are fully transparent elsewhere in the spectrum. We prove that such unparalleled operation can be achieved in wave-manipulating devices using only structural inclusions whose electric and magnetic properties are strongly coupled.

We expect that our results will motivate future work to extend the functionality of the proposed structures and design new devices to fully control electromagnetic waves.

Figure 1

A schematic illustration of a generic, electrically thin metasheet.

Figure 2

Illustration of two absorption regimes in a metasurface with unit cells containing electrically and magnetically polarizable inclusions resonating (a) at different frequencies and (b) at the same frequency. Red and blue lines depict, respectively, normalized electric and magnetic polarizabilities. Solid and dashed lines show the real and imaginary parts of the polarizabilities, respectively. Calculation parameters: (a) ${\omega }_{\mathrm{m}}=0.86{\omega }_{\mathrm{e}}$, $A_{\mathrm{e}}=1.28{\omega }_{\mathrm{e}}^2$, $A_{\mathrm{m}}=1$, ${\gamma }_{\mathrm{e}}=0.2{\omega }_{\mathrm{e}}$, ${\gamma }_{\mathrm{m}}=0.1{\omega }_{\mathrm{m}}$, ${\omega }_{\mathrm{a}}=0.74{\omega }_{\mathrm{e}}$; (b) ${\omega }_{\mathrm{e}}={\omega }_{\mathrm{m}}$, $A_{\mathrm{e}}={\omega }_{\mathrm{e}}^2$, $A_{\mathrm{m}}=1.56$, ${\gamma }_{\mathrm{e}}=0.2{\omega }_{\mathrm{e}}$, ${\gamma }_{\mathrm{m}}=0.2{\omega }_{\mathrm{m}}$, ${\omega }_{\mathrm{a}}=0.8{\omega }_{\mathrm{e}}$.

Figure 3

(a) Single-turn and (b) double-turn helical inclusions. (c) Arrangement of the single-turn and (d) double-turn helical inclusions in the arrays. Blue and red colors denote right- and left-handed helices, respectively.

Figure 4

Absorption coefficient versus conductivity of the inclusion material at the operating frequency of 3 GHz.

Figure 5

Individual normalized polarizabilities of the designed (a) single- and (b) double-turn helical inclusions.

Figure 6

Simulated power reflection $R$, transmission $T$, and absorption $A$ coefficients for infinite arrays of (a) single- and (b) double-turn helical inclusions. (c) Response of the double-turn helix array in a wide frequency range.

Figure 7

An equivalent circuit of the proposed absorbers matched at all frequencies. The resistive load ${\eta }_0=377\mathrm{\Omega }$ denotes the impedance of free space behind the absorber.

Figure 8

The absorption level $A=1-R-T$ in a metasurface with (a) single-turn and (b) double-turn helical elements versus the incident angle.

Figure 9

Fabricated metasurfaces of (a) single-turn and (b) double-turn helices comprising, respectively, 480 and 324 elements embedded in plastic foam.

Figure 10

Measured and the corresponding simulated reflection, transmission, and absorption coefficients for the fabricated metasurfaces with (a) single- and (b) double-turn helical inclusions. The dots denote measured values. The solid lines are the envelope curves of the measured data. The dashed lines show the corresponding simulated coefficients of the measured samples.