Abstract
It is found that due to a simplification in the metaplast model, the original chemical percolation devolatilization (CPD) model does not ensure mass conservation. This is caused by the fact that evacuation of tar is only approximately accounted for in the underlying formulas derived from percolation theory. In this paper, a new version of the metaplast model is developed and implemented in a Julia-based solver. This correction also results in some corrections to the kinetic data for lignin, cellulose, and hemicellulose. The new Julia solver includes two versions, one for small particles where no spatial resolution is required and a spatially resolved version. The spatially resolved version is used to find that, even for pyrolyzing particles, a Biot number of unity is a sufficient criterion to determine when spatial models must be used.