Fractal Analysis Of Adsorption Data Obtained On Soils And Soil Minerals

6.5.1 Soil Minerals

Soil solid phases consist of a very complex mixture of inorganic and organic components that determine the physical and physicochemical properties of the soils. Organic components, which are usually present in much smaller quantities than inorganic components, include plant and animal residues at various stages of decomposition, living cells and tissues of soil organisms, and substances synthesized in soil, the so-called humic substances. Inorganic components include primary and secondary minerals ranging in size from very fine colloids to large rocks. In soils, the sand and silt fractions consist exclusively of primary minerals, whereas minerals in the clay fraction are predominantly secondary and include layered silicates, and various oxides, carbonates and sulfur minerals. The term 'clay mineral' refers to soil inorganic materials less than 2 ^m in effective diameter and comprises minerals with the sheet silicate structure of phyllosilicates. The secondary minerals that are dominant in the soil clay fraction are kaolinites, smectites, vermiculites, mica and especially montmorillonites.

Soil clay minerals often differ in their properties from those of pure minerals. In particular, they are generally much less ordered and smaller in size than pure minerals. Often, neighboring particles or sheets overlap. Furthermore, they exhibit surface irregularities at the molecular scale and a nonuniform microporous structure. Determinations of the fractal dimensions of clay minerals have been reported in several papers [11, 21]. For montmorillonite, a value of Ds « 2 has been measured [11, 21, 91, 92], whereas other authors [87, 93, 94] have obtained a higher value of Ds = 2.7, e.g. by measuring N2 isotherms. Sokolowska and co-workers [15, 16] analyzed the monolayer capacities of montmorillonite with nonpolar and polar adsorbates and found values of Ds = 2.1 and 2.3 respectively. Similar differences in Ds values were also observed for kaolinite [16, 21, 91, 93]. The origin of these differences is probably related to the fact that experiments were carried out on mineral samples of different origins. Small, but variable, amounts of strongly bound water could change surface geometries.

One of the questions that can be answered with the help of adsorption measurements concerns the microtexture of natural clay minerals. Several idealized models for the texture of soil clays (see [5]) have been considered, but rather than assuming one model a priori, one should try to gain useful information from experimental relationships between the size of clay particles and apparent density or surface area and internal porosity, as described in Sections 6.1 and 6.2.1. Experiments aiming at the evaluation of the microtextures of clay minerals were carried out by Ben Ohoud and van Damme [95], who studied kaolinite, sepiolite, palygorskite and 20 monoionic montmorillonite samples. The accessible surface area S of consecutive fractions of size r was measured by N2 adsorption using the classical BET method, whereas the open porosity P was measured from the amounts of adsorbed N2 at a relative vapor pressure of x = 0.99. These authors [95] also determined mass (fragmentation) fractal dimensions by measuring the volume occupied by a mass of powdered particles of a given size. They found, within experimental error, a value of DM = 3 for all clays studied, i.e. the samples were not mass fractals. In contrast, the behavior of the accessible surface area and open porosity of samples was much more interesting [95]. For the fibrous clays, kaolinite and La-saturated montmorillonite, the values of S and P were independent of the grain size, i.e. the surface fractal dimension Ds = 3, which suggested that the ideally porous model was applicable in these cases. All of the other montmorillonite samples were characterized by a linear dependence of ln(S) versus ln(r) over a quite wide range of r values. The slopes of these curves led to nontrivial values of Ds (2 <Ds < 3). Furthermore, the same power laws were obtained for porosity with scaling exponents that were in very good agreement with Ds values. In addition, pronounced correlations between Ds values and the coherence length, i.e. the thickness of the ordered stacks of elementary clay sheets determined from X-ray diffraction, were observed [95].

A question has arisen with regard to the possible existence of a relationship between the properties of exchangeable cations and Ds values, which implies the possible capacity of different cations to generate stacks of smectite lamellae. In this respect, highly charged and/or small cations with a large polarizing power would be expected to have a strong local ordering tendency and, thus, be able to form these stacks. The left panel of Figure 6.2 displays the values of Ds obtained from N2 adsorption data [30, 95], whereas the right panel shows the average adsorption energies of water vapor on different monoionic forms of montmorillonite [96]. The trends shown here

0 0

Post a comment