Anomalous Reflection with Omnidirectional Active Metasurfaces Operating in Free Space

Metasurfaces that manipulate the reflection of impinging sound have recently received significant attention, but virtually all past designs are passive structures with high temporal and spatial dispersion and thus have narrow bandwidths and are unidirectional. In this work, we introduce a method to design omnidirectional and broadband anomalous acoustic reflectors composed of nonresonant active scatterers arranged in a periodic lattice of subwavelength periodicity. The method relies on the manipulation of the wave-vector component parallel to the reflector’s surface (i.e., transverse component) in a broadband manner and for arbitrary impinging sound directions. To illustrate our method experimentally, we design and measure an anomalous reflecting metasurface operating in free space for which the transverse wave-vector components of the incident and reflected waves differ by a prescribed constant. We show that the effect does not depend on the direction of incidence and has a bandwidth an order of magnitude larger than passive anomalous reflectors. This work shows that the active scatterers behave similarly to the atoms of a natural material rather than the unit cells of gratings or phononic crystals in that rearranging the active scatterers changes the metasurface geometry while preserving the device’s desired functionality.

Figure 1

Anomalous reflection. (a) A conventional smooth surface reflects an incident plane wave of wave vector ${\overline{k}}_i$ in the specular direction ${\overline{k}}_s$. The aim is to redirect the reflected wave so that it has the nonspecular wave vector ${\overline{k}}_r$. (b) A metasurface is obtained by embedding active scatterers inside the smooth reflector. Each meta-atom contains a monopole sensor, a dipole-driven transducer and electronics to prescribe the impulse response between the sensor and the driver. The realized metasurface is calibrated by performing measurements on a plane situated at distance $z_{\delta }$ from the metasurface.

Figure 2

Experimental setup. Two metasurface configurations having (a) $3\times 3$ meta-atoms and (b) $6\times 1$ meta-atoms are measured. (c) The metasurface behavior is quantified through measurements along the $y$-$z$ plane perpendicular to the metasurface. (d) The acoustic incident fields are measured for normal (left) and ${30}^{\circ }$ (right) incidence at 2400 Hz.

Figure 3

Experimental results at 2400 Hz. (a) The acoustic response of the first metasurface made of $3\times 3$ meta-atoms to a normally incident spherical wave. The specular reflection obtained with the deactivated meta-atoms (left) is compared against the anomalous reflection obtained with the activated meta-atoms (center) and the expected, prescribed nonspecular reflection obtained in numerical simulations. (b) The same measurements as in (a) performed for ${30}^{\circ }$ incidence. (c),(d) Same measurements as in (a),(b) for the second metasurface composed of $6\times 1$ meta-atoms.

Figure 4

Far-field acoustic pressure amplitude of the reflected wave. The reflections [(a),(b)] and [(c),(d)] are from the metasurfaces composed of $3\times 3$ meta-atoms (e) and $6\times 1$ meta-atoms (f), respectively. The reflections are shown when the incident angles are [(a),(c)] $0^{\circ }$ and [(b),(d)] ${30}^{\circ }$. For each case, the measured (solid line) and simulated (dashed line) reflection are shown for at 2300, 2400, and 2500 Hz.