Porous sound absorbing materials from natural sources: data and modelling

• Sound absorption spectra of naturally sourced material samples are compared. • Tortuosity is estimated using the frequency of the first quarter wavelength resonance. • Tortuosity estimates are higher than obtained by fitting a five parameter model. • A log normal size distribution of non-uniform cylindrical pores model fits some data. • Plant fiber samples are better sound absorbers than samples made from harder sources. This study reports porosity, flow resistivity, and normal incidence sound absorption spectra for samples fabricated from plant fibers, crushed fruit stones, chopped nut shells, and wood-cement composites. The Johnson–Champoux–Allard (JCA) model has been used to fit measured data and extract microstructural parameters such as tortuosity and characteristic lengths. An alternative, semi-analytical approach based on a model for a log normal size distribution of non-uniform cylindrical pores (NUPSD) is proposed. It uses porosity, flow resistivity, and the frequency of the first quarterwavelength resonance to estimate tortuosity. Also, it enables estimates of thermal permeability for making predictions with the Johnson-Champoux-Allard-Lafarge (JCAL) model. It avoids the need for direct fitting and provides comparable predictions. Typically, it predicts higher values of tortuosity than those resulting from fitting the JCA model. For many samples, particularly those with fibrous structure, the NUPSD-based predictions agree better with measured spectra than predictions based on JCA model fitting. The results indicate that the materials made from plant fibers are better sound absorbers than materials made from wood chips, wood cement, crushed fruit stones or chopped nut shells. Those with the higher tortuosity, implying broader pore size distributions, are more useful as sound absorbers for building applications.