Following the rapid adoption of automated sorption balances (often referred to as «dynamic vapor sorption» or «DVS» analysers) a large number of studies have been published in the past ten years on the water vapor sorption kinetics in wood, cellulose, and other biopolymers. Nearly all of these papers use the parallel exponential kinetics model (the so called “PEK” model) to fit the kinetic data, see Figure 1. Empirically it has been shown that the PEK model results in high R2 values when fit to sorption data.
Theoretically, the fitting parameters from the PEK model have been used to describe many different wood phenomena including the location of sorption sites, the monolayer moisture content, activation energies, and viscoelastic properties of the wood cell wall. In this paper, we closely examine how the PEK model fit parameters depend upon the method used to collect sorption data and whether these parameters can be physically meaningful.
Keywords: moisture sorption, parallel exponential kinetics model (PEK), dynamic vapor sorption (DVS)
Authors
Emil E. Thybring
Department of Geosciences and Natural Resource Management, University of Copenhagen, Denmark
Charles R. Boardman
Samuel V. Glass
Samuel L. Zelinka
Login to download the PDF