Smectites adsorbed water

The view that the clay surface perturbs water molecules at distances well in excess of 10 A has been largely based on measurements of thermodynamic properties of the adsorbed water as a function of the water content of the clay-water mixture. There is an extensive literature on this subject which has been summarized by Low (6.). The properties examined are, among others, the apparent specific heat capacity, the partial specific volume, and the apparent specific expansibility (6.). These measurements were made on samples prepared by mixing predetermined amounts of water and smectite to achieve the desired number of adsorbed water layers. The number of water layers adsorbed on the clay is derived from the amount of water added to the clay and the surface area of the clay. 

The assumption that the water is adsorbed in uniform layers on all the clay surfaces for a wide range of mixtures has been criticized (2, 20). The argument is that the individual clay particles in the clay-water mixture do not expand beyond a certain distance regardless of the quantity of water which is added. The clay layers group themselves into tactoids resulting in two populations of water those molecules which are found between the tactoids and those directly perturbed by the clay layers. If true, this would invalidate the procedure used to calculate the thermodynamic properties of the adsorbed water. However, other workers have reported complete delamination of certain smectites (21., 22). It is not clear under what conditions tactoids will form, or not, and this uncertainty is underlined in (21) (see remarks by Nadeau and Fripiat, pages 146-147). 

The validity of the assumption that the various thermodynamic properties of the smectite remain invariant, regardless of the state of hydration, has been addressed in detail by Sposito and Prost (1). They point out that one would, for example, expect hydrolysis of the clay to occur at high water contents, and also, it is likely that the exchangeable cations will change their spatial relationship with the clay layers. Thus, the derived thermodynamic properties of the adsorbed water would not represent correct values. 

In view of the problems associated with the expanding 2 1 clays, the smectites and vermiculites, it seemed desirable to use a different clay mineral system, one in which the interactions of surface adsorbed water are more easily studied. An obvious candidate is the hydrated form of halloysite, but studies of this mineral have shown that halloysites also suffer from an equally intractable set of difficulties (JO.). These are principally the poor crystallinity, the necessity to maintain the clay in liquid water in order to prevent loss of the surface adsorbed (intercalated) water, and the highly variable morphology of the crystallites. It seemed to us preferable to start with a chemically pure, well-crystallized, and well-known clay mineral (kaolinite) and to increase the normally small surface area by inserting water molecules between the layers through chemical treatment. Thus, the water would be in contact with both surfaces of every clay layer in the crystallites resulting in an effective surface area for water adsorption of approximately 1000 tor g. The synthetic kaolinite hydrates that resulted from this work are nearly ideal materials for studies of water adsorbed on silicate surfaces. 

Clay minerals with their own surface properties affect the near surface water in different ways. The adsorbed water in the case of kaolinite consists only of water molecules ( pure water), whereas water adsorbed on a smectite-type mineral is an aqueous solution, due to the presence of exchangeable cations on the 2 1 layer sihcate. Sposito (1989) noted the generally accepted description that the spatial extent of adsorbed water on a phyUosilicate surface is about 1.0 nm (two to three layers of water molecules) from the basal plane of the clay mineral. 

The important role of the exchangeable cation in determining the structure of adsorbed water on smectites was discussed in the review of experimental studies of the structure of water adsorbed on smectites [32], It was concluded that the spatial arrangement of the adsorbed water molecules indeed derives mainly from the solvation of exchangeable cations. Despite the great... 

Swelling is a further important property of clays. In fact, many clay minerals adsorb water between their layers, which move apart and the clay swells. For efficient swelling, the energy released by cations and/or layer solvation must be sufficient to overcome the attractive forces (such as hydrogen bonding) between the adjacent layers. In 2/1 (TOT) clay minerals such as smectite, during swelling, the interlayer cations can... 

R. Prost, Interactions between adsorbed water molecules and the structure of clay minerals Hydration mechanism of smectites, Proc. Int. Clay Conf. 1975, p. 351 (1976). 

The spatial extent of adsorbed water on smectite surfaces is a matter of some controversy. Infrared spectroscopy, NMR relaxation, and X-ray and neutron diffraction experiments all point to a thickness of the adsorbed water film of around 1.0 nm. However, certain thermodynamic data, summarized for Na-montmorillonite in Table 2.5, suggest a thickness as great as 10 nm or more." These data are for partial and apparent specific properties of montmorillonite-water systems whose variation with water... 

The definition of adsorbed water adopted in Sec. 2.3 requires an arrangement of water molecules that differs significantly from that in an appropriate reference aqueous phase. For water on the surfaces of kaolinite group minerals the reference phase is bulk liquid water, whereas for water on vermiculite and smectite surfaces the reference phase is an aqueous solution because of the presence of exchangeable cations on the 2 1 layer silicates. On the ba,si.s of this definition, the consensus developed in Sec. 2.3 is that the spatial extent of adsorbed water on a phyllosilicate... 

Another important chemical property of adsorbed water on vermiculite and smectite surfaces is its Br nsted acidity. This property should refer principally to the acidity of the solvated exchangeable cations, as described by the reaction... 

See, e.g., D.M.C. MacEwan and M. J. Wilson, Interlayer and intercalation compounds of clay minerals, in G. W. Brindley and G. Brown, op cit. That 1 1 electrolytes cannot be dissolved completely in adsorbed water on smectites and illitic micas has been shown by A. M. Posner and J. P. Quirk, The adsorption of water from concentrated electrolyte solutions by montmorillo-nite and illite, Proc. Royal Soc. (London) 278A 35 (1964). 

The representation of the smectite-water system by these model potential functions does involve significant approximation, but we have found that it identifies the most important features of the clay mineral-water-cation interaction leading to the experimentally observed V structure of adsorbed water in smectite interlayers (10-18). 

The balance between the polarity of the adsorptive, the solvent, and the interlamellar environment is instructively illustrated by the adsorption of phenol and trichloro-phenol from water and hexane on tetramethylammonium (TMA) and hexadecyl trimethylammonium (HDTMA) montmorillonite (Fig. 9) [96]. Most striking is the adsorption of trichlorophenol from water it is very low for TMA but high for HDTMA cations. Both smectites adsorb comparable amounts from hexane solutions. Evidently, the pronounced hydrophobic character of the interlayer space of HDTMA montmorillonite favors adsorption of the less hydrophilic trichlorophenol from water. The adsorption of pentachlorophenols from water is also distinctly higher on montmorillonite modified with long chain cations than on montmorillonite with short chain cations [97]. 

Kowalska M, Gtiler H, Cocke DL (1994) Interactions of clay minerals with organic pollutants. Sci Total Environ 141 223-240 Kukkadapu RK, Boyd SA (1995) Tetramethylphosphonium-smectite and tetramethylammonium-smectite as adsorbents of aromatic and chlorinated hydrocarbons - effect of water on adsorption efficiency. Clays Clay Miner 43 318-323... 

An interaction potential between the surface and ions may also be needed in simulating counterion diffusion for the smectite and mica surface models. The form of such an interaction potential remains to be determined. This may not pose a significant problem, since recent evidence (40) suggests that over 98% of the cations near smectite surfaces lie within the shear plane. For specifically adsorbed cations such as potassium or calcium, the surface-ion interactions can also be neglected if it is assumed that cation diffusion contributes little to the water structure. In simulating the interaction potential between counterions and interfacial water, a water-ion interaction potential similar to those already developed for MD simulations (41-43) could be specified. 

Halloysite-10A represents a structure with few if any interlayer cations, allowing one to investigate the relatively simple case of water interacting with a clay surface. Similarly, ice-like models have been proposed for water adsorbed on smectite and vermie-ulite surfaces (2, 12, 12). These represent cases of charged clay layers with adsorbed exchangeable cations. 

Our approach has been to study a very simple clay-water system in which the majority of the water present is adsorbed on the clay surfaces. By appropriate chemical treatment, the clay mineral kao-linite will expand and incorporate water molecules between the layers, yielding an effective surface area of approximately 1000 m2 g . Synthetic kaolinite hydrates have several advantages compared to the expanding clays, the smectites and vermiculites they have very few impurity ions in their structure, few, if any, interlayer cations, the structure of the surfaces is reasonably well known, and the majority of the water present is directly adsorbed on the kaolinite surfaces. 

---