The pore diffusion modeling approach is often used to model the sorpti
on of organic chemicals by natural sorbents, implicitly assuming that
the physical and sorptive properties of the sorbent are homogeneous. B
ut application of such approaches to systems with significant variatio
ns in particle size and sorption equilibrium properties may be inappro
priate; most natural systems have such variations. A general multiple-
particle class pore diffusion model that accounts for variations in ph
ysical and sorptive properties was developed and used to interpret the
results of a batch sorption rate study. Results indicated that multip
le-particle class models provide a more accurate representation of lon
g-term sorption rate data than traditional single-particle class appro
aches, the common approach of adding an instantaneous equilibrium frac
tion to single-particle class models leads to model parameters that ar
e sensitive to the range over which data is collected, and tortuosity
factors obtained with multiple-particle class modeling approaches were
consistent with the range of values expected based upon mechanistic r
easoning.