An acoustic metadiffuser is a subwavelength locally resonant surface relying on slow sound propagation. Its design consists of rigidly backed slotted panels, with each slit being loaded by an array of Helmholtz resonators. Due to the slow sound properties, the effective thickness of the panel can therefore be dramatically reduced when compared to traditional diffusers made of quarter-wavelength resonators. The aim of this work is to experimentally validate the concept of metadiffusers from the scattering measurements of a specific metadiffuser design, i.e., a quadratic residue metadiffuser. The experimental results reported herein are in close agreement with analytical and numerical predictions, therefore showing the potential of metadiffusers for controlling sound diffusion at very low frequencies.

The anisotropic pore structure and elasticity of cancellous bone cause wave speeds and attenuation in cancellous bone to vary with angle. Previously published predictions of the variation in wave speed with angle are reviewed. Predictions that allow tortuosity to be angle dependent but assume isotropic elasticity compare well with available data on wave speeds at large angles but less well for small angles near the normal to the trabeculae. Claims for predictions that only include angle-dependence in elasticity are found to be misleading. Audio-frequency data obtained at audio-frequencies in air-filled bone replicas are used to derive an empirical expression for the angle-and porosity-dependence of tortuosity. Predictions that allow for either angle dependent tortuosity or angle dependent elasticity or both are compared with existing data for all angles and porosities.

Predictions of a modified anisotropic Biot-Allard theory are compared with measurements of pulses centered on 100 kHz and 1 MHz transmitted through water-saturated stereo-lithographical bone replicas. The replicas are 13 times larger than the original bone samples. Despite the expected effects of scattering, which is neglected in the theory, at 100 kHz the predicted and measured transmitted waveforms are similar. However, the magnitude of the leading negative edge of the waveform is overpredicted, and the trailing parts of the waveforms are not predicted well. At 1 MHz, although there are differences in amplitudes, the theory predicts that the transmitted waveform is almost a scaled version of that incident in conformity with the data. (C) 2010 Acoustical Society of America. [DOI: 10.1121/1.3397581]

Sound diffusion refers to the ability of a surface to evenly scatter sound energy in both time and space. However, omni-directional radiation of sound, or perfect diffusion, can be impractical or difficult to reach under traditional means. This is due to the considerable size required, and the lack of tunability, of typical quarter-wavelength scattering strategies necessary for producing the required complexity of the surface acoustic impedance. As such, it can be a challenge to design sound diffusing structures that can display near perfect diffusion performance within slim dimensions. In this work, we propose a method for obtaining quasi-perfect and broadband sound diffusion coefficients using deep-subwavelength acoustic diffusers, i.e., metadiffusers. The relation among the geometry of the metasurface, the bandwidth, and the diffusion performance is analytically and numerically studied. For moderate bandwidths, around 1/3 of an octave, the method results in nearly perfect sound diffusion, while for a bandwidth of 2.5 octaves, a normalized diffusion coefficient of 0.8 was obtained using panels 1/30th thinner than traditional phase-grating designs. The ratio between the wavelength and the size of the unit cell was identified as a limitation of the performance. This work demonstrates the versatility and effectiveness of metadiffusers to generate diffuse reflections outperforming those of classical sound diffusers.