As most home-theater buffs know, acoustic treatments can significantly improve the sound of most rooms. One of the most common treatments is a diffuser panel, which scatters—or diffuses—sound waves that strike it. This reduces or even eliminates strong reflections that can make the frequency response of the room very uneven.
The most common type of diffuser is the Schroeder diffuser (SD), named for its inventor, German physicist Manfred Schroeder. Proposed over 40 years ago, the 2D Schroeder diffuser consists of an array of evenly spaced square depressions that are identical in length and width but different in depth, as seen in the left portion of the drawing above. (A 1D Schroeder diffuser consists of long channels with different depths.)
The overall thickness of a conventional Schroeder diffuser must be about half the wavelength of the lowest frequency it needs to diffuse. However, that means it can be too thick to be practical. For example, a typical male voice can reach down to 85 Hz, which has a wavelength of about 13 feet. A Schroeder diffuser must be about 6.5 feet deep to diffuse that frequency.
A team of researchers at North Carolina State University and Nanjing University in China recently published a paper in Physical Review X that describes a very exciting development in the field of diffusers. Dubbed a “metasurface Schroeder diffuser” (MSD), the new design requires only one tenth the thickness of a conventional SD.
Instead of open cavities with different depths, the MSD uses square openings of different sizes with identical cavities behind them, as seen in the right portion of the drawing above. In this case, the overall depth needs to be only about one twentieth of the lowest wavelength to be diffused, ten times thinner than the conventional approach. Returning to the previous example, an MSD needs to be less than eight inches thick to diffuse frequencies down to 85 Hz.
The team simulated the effect of a conventional SD and an MSD, then built an MSD using a 3D printer and performed a series of measurements. Here are some of their results:
With sound arriving at an angle that is perpendicular (aka “normal”) to the surface, the MSD and SD diffuse it almost identically. By contrast, a flat plate reflects most of the sound back the way it came. In this diagram, (a) represents the simulated scattering effects, (b) represents the measured (upper) and simulated (lower) scattered acoustic field distributions in the x/z plane, and (c) reveals the simulated (black) and measured (red) scattering field distributions.
When the sound arrives at an angle other than perpendicular—in this case, 45°— the diffusion characteristics of the SD and MSD are virtually the same, while the flat plate reflects the sound as you would expect (angle of incidence = angle of reflection). In this diagram, (a) represents the simulated scattering effects, (b) represents the measured (upper) and simulated (lower) scattered acoustic field distributions in the x/z plane, (c) represents the measured (upper) and simulated (lower) scattered acoustic field distributions in the y/z plane, and (d) reveals the simulated (black) and measured (red) scattering field distributions.
Here you can see the simulated and measured diffusion of the prototype MSD at 5772 Hz, 6860 Hz, and 8153 Hz with normal (perpendicular) and 45° incidence.
Clearly, the MSD has great potential for providing effective diffusion at much lower frequencies with far thinner panels than conventional Schroeder diffusers. I eagerly await the day when this technology becomes available to consumers who want to acoustically treat their home theater without taking up so much space.