BASIC CONCEPTS AND AN OVERVIEW - Foundations of Nanotechnology: Pore Size in Carbon-Based Nano-Adsorbents - page 152

fullerene-like model to predict the adsorptive and other properties of microporous
carbons. By far the most ambitious program of work in this area has been carried
out by Terzyk et al. [18], whose results have been published in a series of papers
beginning in 2007. In the first of these 36 different carbon structures with in-
creasing microporosity, labeled S 0 -S 35 , were generated. Fragments were then pro-
gressively added to create the 36 structures labeled S 0 -S 35 . Pore size distribution
(PSD) curves for the structures were calculated using the method of Bhattacharya
and Gubbins (BG). This involves determining the statistical distribution of the
radii of the largest sphere that can be fitted inside a pore at a given point. It shown
that the most crowded structure, S 35 , has a much narrower range of pore sizes
than the initial S 0 structure. Argon adsorption isotherms were simulated for these
structures using the parallel tempering Monte Carlo simulation method. These
show that the gradual crowding of the S 0 structure(leading finally to S 35 ) leads to a
decrease in the maximum number of adsorbed molecules. On the other hand, the
S 0 structure exhibits less adsorption at low pressures than the more crowded ones
because the average micropore diameter is larger. Also notable is the increasing
sharpness of the inflection point in the isotherms, a feature which is often reported
for experimental systems. The simulated isotherms were then used to determine
PSD curves, using a range of widely used methods, with the aim of checking the
validity of these methods. Good agreement was found between the PSDs deter-
mined from the isotherms and the PSDs from the BG method. This confirms the
validity of various methods for calculating PSD curves from adsorption data. It
would also seem to confirm the validity of the fullerene-related model for micro-
porous carbon. The densities of these structures were calculated, and values in
the range 2.18-2.24 g cm -3 were found, consistent with typical densities of non-
graphitizing carbons. Once again, pore size distributions for the structures were
determined using the BG method. Good agreement was found between the PSDs
determined from the simulated adsorption data and the original PSDs from the BG
method, where the PSD curve determined from the Bhattacharya-Gubbins model.
The adsorption of Ne, Ar, Kr, Xe, CCl 4 and C 6 H 6 on the S 0 and S 35 carbons was
modeled. The simulated data were compared with the predictions of the Dubinin-
Radushkevich and Dubinin-Astakhov adsorption isotherm equations, and a good
fit was found for the S 35 carbon. For the S 0 carbon the Dubinin-Izotova (DI) equa-
tion gave a better fit because the micropores in this model have a wide distribution
of diameters. The simulated isotherms exhibited a number of features similar to
those seen in experimental results. For example, the isotherms for CCl 4 and C 6 H 6
were temperature invariant, as observed experimentally. It was also noted that
the isotherms obeyed Gurvich's rule, which states that the larger the molecular
collision diameter the smaller the access to micropores, as well as other empirical
and fundamental correlations developed for adsorption on microporous carbons.
The effect of oxidizing the carbon surface on porosity was analyzed in a paper
published in 2009. A virtual oxidation procedure was employed, in which surface