BASIC CONCEPTS AND AN OVERVIEW - Foundations of Nanotechnology: Pore Size in Carbon-Based Nano-Adsorbents - page 102

1.2.2.16 COMPARISON BETWEEN DIFFERENT MATHEMATICAL
MODELS FOR PSD CALCULATING AND ADVANTAGES AND
DISADVANTAGES OF THESE MODELS
This chapter has reviewed modeling over the years and the progress that can be
identified. It has been demonstrated that modelers, using modem computational
systems, base their calculations on structure systems, usually around the graphitic
microcrystallite. An objective of such modelers is the simulation of adsorption
isotherms determined experimentally. Whether or not the structural models as-
sumed for their work can be considered to be “realistic” is rarely a matter for dis-
cussion. Accordingly, the considerations of such modelers should not be adopted,
uncritically, by those who have other interests in activated carbon. There is no
unique structure within an activated carbon, which provides a specific isotherm,
For example, the adsorption of benzene at 273 K. The isotherm is a description of
the distribution of adsorption potentials throughout the carbon, this distribution
following a normal or Gaussian distribution. If a structure is therefore devised
whichpermits a continuous distribution of adsorption potentials, and this model
predicts an experimental adsorption isotherm, this then is no guarantee that the
structure of the model is correct. The wider experience of the carbon scientist,
who relates the model to preparation methods and physical and chemical proper-
ties of the carbon, has to pronounce on the “reality” or acceptance value of the
model. Unfortunately, the modeler appears not to consult the carbon chemist too
much, and it is left to the carbon chemist to explain the limited acceptability of
the adopted structures of the modeler. This approach of the modeler is still to be
found in many publications. Different models rely on different assumptions in or-
der to obtain relationships allowing the calculation of the main characteristics of
adsorbent structural heterogeneity. The assumptions are summarized as follows:
(a) The shape of the pores is assumed to be slit like with effects of intercon-
nectivity to be neglected.
(b) The molecule-surface interactions obey the equation of steel.
(c) The adsorbed phase can be considered as liquid-like layers between the
two parallel walls of the pore, as a monolayer or as a double layer.
(d) The thickness of the adsorbed layer is determined by the distance between
two parallel walls of the slit-like porosity, the density of the layers being
constant between the layers and being equal to that of the bulk liquid.
(e) No gas-phase adsorbate exists within the volumes between the layers [1,
2, 133].
Earlier, Olivier [10] used GCMC and DFT as modeling methods based on
slit-like pores with graphitic surfaces, as used by Kowalczyk et al. [15]. Olivier
[10] concluded that the inclusion of surface heterogeneity into the model would
have little effect on obtained simulations. Further, Olivier [10] considers that the
use of alternative geometries maybe worth pursuing. The approach of is interest-
ing in that several quite different approaches have been brought to the modeling