Page 197 - Adsorption by Powders and Porous Solids
P. 197

184                             ADSORPTION BY POWDERS AND POROUS SOUQ

   such as
                                N =b-Dh
                                  m                                (6.29)
   where Nm is the number of molecules in the completed monolayer, a is the adsorptive
   molecular area and D, is now the fractal dimension of the accessible surface (F*
   and Avnir, 1989).
     The magnitude of Da is determined inter alia by the degree of  surface roughness
   porosity. In principle, a lower limit of Da = 2 is obtained with a perfectly smooth surface
   on the molecular scale. Most non-porous materials would be expected to exhibit sorne
   surface roughness. With such a material, a constant value of D, between 2 and 3 implies
   that  there is  a degree of  self-similarity:  the  shape of  the surface irregularities thus
   remains invariant over a certain range of resolution. The physical structure of a fractal
   surface will therefore appear similar when viewed at different magnifications.
     Fractal plots of log n,  versus log a for two porous silicas are shown in Figure 6.3
   (here, n,  is the BET monolayer capacity). Both plots are linear, giving D, = 2.98 for
   the silica gel and D, = 2.09 for the controlled pore glass. These values reflect the
   extremes of the fractal scale, the latter being close to the ideal value for a flat surf=,
     Some of the many values of D, compiled by  Farin and Avnir (1989) and Avnir
   et al., (1992) are given in Table 6.2.  Most values are within the theoretical fractal
   range, D, = 2-3  and it is noteworthy that D, = 2 for the graphitized carbon blacks and
    the pillared clays.
     The values of  D, of  (1.89 * 0.09)  and (1.94 rt 0.10) reported by  Van  Darnme and
    Fripiat (1985) for pill&  clays were derived from the multilayer capacities of nitrogen
    and various organic adsorptives. The fact that D,  2 appeared to confirm that the basal
    smectite surface was smooth and that the pillars were regularly distributed. It was argued
    by Van Damrne and Fripiat that a random distribution of  the pillars would necessarily
    lead to some localized molecular sieving and that this in turn would result in D, > 2.
      Various other aspects of fractal analysis have been discussed by Van Damrne and
    Fripiat and their co-workers. For example, by  extending the BET model to fractal
    surfaces, Fripiat et al. (1986) were able to show that the apparent fractal dimension is
    reduced by the progressive smoothing of a molecularly rough surface. Alternatively,
    the effect of a micropore filling contribution is to enhance the fractal dimension.
      Other investigators, including Heifer and Obert (1989), Heifer et al. (1990), Krim
    and Panella (1991), Panella and Krim (1994) and Neimark and Unger (1 993), have also
    studied multilayer adsorption on fractally rough surfaces. In particular, Heifer and his
    co-workers point out that the interpretation of a fractal dimension of a porous surface
          Table 6.2.   Values of  accessible fractal dimension evaluated from adsorption data
          (taken from Farin and Avnir, 1989)
          Adsorbent            Adsorptives         Fractal dimension, D,
          Graphitized carbon blacks   N, and alkanes   1.9-2.1
          Activated carbons    N, and organic molecules   2.3-3.0
          Pillared clays       N, and organic molecules   1.8-2.0
          Silica gels          Alkanes                 2.9-3.4
          Other oxides         N, and alkanes          2.4-2.7
   192   193   194   195   196   197   198   199   200   201   202