N₂-BET specific surface area of bentonites

Abstract

The specific surface areas (SSA_N2BET) of 36 different bentonites had larger values for Ca²⁺/Mg²⁺ bentonites than for Na⁺ bentonites. This trend could not be explained by the different d(0 0 1) values nor by the different microstructures. The investigation of Cu-triene-exchanged smectites, which on drying at 105 °C still had a d(0 0 1) value accounting for approximately 13 Å, proved that the SSA_N2BET of low-charged smectites increased more than that of high-charged smectites. This could be explained by: (i) more space between the permanent charge sites in the case of low-charged smectites and (ii) the fact that the layers of Cu-triene smectites do not collapse at 105 °C. In contrast the SSA_N2BET of Ca²⁺-exchanged bentonites could not be related to the layer charge density (LCD) as in the case of the Cu-triene-exchanged bentonites which is probably due to the varying number of collapsed layers. In conclusion, the SSA_N2BET of bentonites which is known to be largely variable is probably determined by microporosity resulting from the quasi-crystalline overlap region and accessible areas of the interlayer. The number of layers per stack and the microstructure are supposed to play a subordinate role. The larger SSA_N2BET of Ca/Mg bentonites compared to Na bentonites probably can be explained by the larger space between the charges in the case of the presence of divalent cations.

Introduction

Bentonites are fascinating natural raw materials consisting mainly of swelling clay minerals. These clays are used for many industrial applications but are also used to study the properties of smectites (group of swelling clay minerals) which is important in soil science and industrial application of clays.

The external specific surface area of swelling clay minerals is measured in order to: (i) understand and to model surface reactions, (ii) determine surface site densities (e.g., of adsorbed species), and (iii) to calculate the internal specific surface area (external specific surface area value is subtracted from the total surface area) mainly of soils and clay barrier systems (natural clay barriers as well as geotechnical barriers, e.g., for highly radioactive waste disposal). By far the most common method to determine the external specific surface area is nitrogen adsorption in combination with the BET equation (in this study SSA_N2BET). The SSA_N2BET is supposed to represent the external surface area only. Many authors critically discussed the validity of this assumption. A recent summary of this discussion was provided by Michot and Villieras [1] based on the work of Aylmore et al. [2], Rutherford et al. [3], and Sauzéat et al. [4]. The advantage of the nitrogen adsorption measurement is the good reproducibility which allows a comparison of the data of different laboratories. The key question, however, is to identify where N₂ molecules actually adsorb and in turn which bentonite properties determine the N₂ adsorption. The SSA_N2BET is known to depend on: (i) sample pretreatment (such as grinding, temperature and vacuum), (ii) microporosity, and (iii) exchangeable cations. The key questions are: how do these parameters affect the SSA_N2BET, where does N₂ actually adsorb, and what is the physical meaning of the BET values.

Schoonheydt states that “the majority of the SSA_N2BET originates from basal planes and from micro- and mesopores formed by the irregular stacking of the elementary particles and of the aggregates” [5]. The dominant microporosity is believed to exist at the quasi-crystal edges resulting from the imperfect stacking of the TOT layers, e.g., due to turbostratic disorder [1]. The current model is that “slit-shaped micropores at particle edges” [1] exist in the “overlap region” [6] later on termed “quasi-crystalline overlap region” [3].

This model includes the hypothesis that the volume of these micropores increases on swelling due to the increase of the distance of the TOT layers (graphically shown in Fig. 1). The micropores, therefore, are related to the d(0 0 1) value which, of course, is affected by the type of interlayer cation and depends on the sample pretreatment (particularly heating conditions). Accordingly Cs montmorillonites generally show higher SSA_N2BET values than Na⁺ or Ca²⁺ montmorillonites likely owing to the higher d(0 0 1) values of Cs montmorillonites in the dried state in which the SSA_N2BET is determined [3]. The same authors, on the other hand, point out that the two bentonites considered in their study significantly differ with respect to the degree of SSA_N2BET increase on Cs exchange. This difference is explained by different edge structures.

Furthermore, the SSA_N2BET might also be affected by the arrangement of the particles toward each other (microstructure) which in turn is reflected by the mesopore size distribution [1]. Ewald investigated the accessibility of the interlayer space of alkylammonium bentonites and found that N₂ can enter the near-edge interlayer space of smectites treated with small alkylammonium ions [7]. This raises the question if the accessibility of the near-edge interlayer also plays a role in the case of Na/Ca bentonites.

The present study investigates the SSA_N2BET of 36 well-characterized bentonite samples. Considering a significant set of different materials was believed to help to identify the reason for different SSA_N2BET values of different bentonites which would improve the understanding of the physical meaning of the SSA_N2BET values.