BASIC CONCEPTS AND AN OVERVIEW - Foundations of Nanotechnology: Pore Size in Carbon-Based Nano-Adsorbents - page 106

rosity are of imperfect graphene-like layers, with surface irregularities
and other defects.
7. This model also demonstrates that most of the continuous three-dimen-
sional grapheme layer is able to act as an adsorbent surface making use of
both sides. Rarely, do parts of the graphene layer come together (stack) in
two or three layers. The range of such “different” sites is a feature of the
models. Clearly, all the requirements for molecular-sieve properties, the
dynamics and enthalpies of adsorption are present even though the detail
is absent. The effect of increasing HTT would be to remove defects or
irregularities of the carbon atom network resulting in decreased enthal-
pies of adsorption on more homogeneous surfaces. Such a porous system
would respond to SAXS and SANS with the range of ring structures in the
carbon atom network being sufficient to explain observed Raman spectra
and electron spin resonance (ESR) during the carbonization process [1,
133].
There are many methods for calculation of pore size distributions (PSDs) and
most of them are potentially applicable for nanoporous carbons. PSDs for nano-
porous carbons are usually evaluated using methods based on either the Kelvin
equation or the Horvath-Kawazoe method and its modifications. The first group
includes the models of Barrett, Joyner and Halenda (BJH) method, Cranston and
Inkley (CI), Dollimore and Heal (DH), and Broekhoff and de Boer (BdB). Al-
though the BJH, CI, and DH methods are often considered as appreciably dif-
ferent, all of them are based on the general concept of the algorithm outlined in
the BJH work. To implement the algorithms proposed in these three methods, the
knowledge of the relation between the pore size and capillary condensation or
evaporation pressure and the t-curve is required, and a choice needs to be made
which branch of the isotherm is appropriate for the PSD calculation. The original
BJH, CI, and DH models are not fully consistent as far as the selection of these
relations and the choice of the branch of the isotherm are concerned. These incon-
sistencies are capable of affecting the results of calculations much more than the
minor differences in the algorithms, being most likely responsible for claims that
these three methods differ substantially. The BJH, CI and DH methods assume the
same general picture of the adsorption desorption process. Adsorption in meso-
pores of a given size is pictured as the multilayer adsorption followed by capillary
condensation (filling of the pore core, that is, the space that is unoccupied by the
multilayer film on the pore walls) at a relative pressure determined by the pore di-
ameter. The description is pictured as capillary evaporation (emptying of the pore
core with retention of the multilayer film) at a relative pressure related to the pore
diameter followed by thinning of the multilayer. Because the concept underlying
the BJH, CI, and DH models appears to be correct, it is important to:
(i) Establish an accurate relation between the pore size and capillary conden-
sation or evaporation pressure;