J. J. Fripiat

A fractal analysis of adsorption processes by pillared swelling clays

We have applied the concept of fractal dimension D of solid surfaces, developed by Pfeifer and Avnir [J. Chem. Phys. 79, 3566 (1983)] to adsorption on pillared clays, namely, on layer lattice silicates in which the layers are propped apart by large organic or inorganic cations. D was determined either from the radius dependence of the saturation coverages of various yardstick molecules on the same pillared clay, or, alternatively, from the saturation coverages of one type of yardstick molecule (nitrogen) on various surfaces covered with pillars of increasing size. Both methods yield D≂2, or slightly below 2, within experimental error, in the range ∼4–16.5 A. This confirms the common assumption that mica‐type surfaces are molecularly smooth, and shows that the distribution of pillars in the pillared clays that we examined is close to homogeneity. The relation between D, microporosity, and molecular sieving is briefly discussed.

Monolayer adsorption on fractal surfaces: A simple two‐dimensional simulation

We have simulated the adsorption of homologous series of ‘‘molecules’’ on deterministic fractal curves in the whole range 1≤D<2. Adsorption differs from a regular Hausdorf dimension determination in the sense that in the latter process the molecules are put with their center on the curve in order to cover it, whereas in the former process the molecules have to reach the curve, coming from one side only, and are not allowed to cross it. Three series of molecules were used: (i) circles of increasing diameter, r; (ii) rectangles of increasing height, t; (iii) rectangles of high aspect ratio (almost lines) and increasing length, l. The fractal exponents Dr, Dt, and Dl were derived from ln Nm vs ln r, ln t, and ln l plots, respectively, Nm being the number of molecules necessary to reach ‘‘monolayer’’ coverage. With linear molecules, we found Dl=1 in whole range 1≤D<2. With the circular and rectangular molecules, two regimes were observed. In the range 1≤D<1.5, Dr=D and Dt=D−1. These results are consistent wit...

Acid-base and complexation behavior of porphyrins on the intracrystal surface of swelling clays: Meso-tetraphenylporphyrin and meso-tetra(4-pyridyl)porphyrin on montmorillonites

Abstract The reactivity of two simple mesosubstituted porphyrins, TPP and TPyP, in the interlamellar space of montmorillonites has been studied in two respects: (I) the ability of the metal free porphyrins to react with the exchangeable cations (Na + , Mn 2+ , Fe 2+ , Cu 2+ , Co 2+ ) of the mineral, and (II) the stability of some metalloporphyrins (Fe 3+ , Sn 4+ ) on the clay surface. The effect of the hydration state of the clay surface has been particularly investigated. It has been shown that TPP always undergoes a protonation reaction when adsorbed. The amounts adsorbed are small (20 × 10 −6 mole/g). Strongly acid-resistant metallocomplexes, like Sn(IV)TPP, can be adsorbed without demetallating, if enough water is present on the surface. Weaker complexes, like Fe(III)TPP, are irreversibly demetallated and protonated. TPyP shows a more complex reactivity, depending on the water content of the clay. The amounts adsorbed are also much larger (190 × 10 −6 mole/g). Most probably, the four pyridyl substituents are protonated, but, unless the water content of the clay is drastically reduced, the tetrapyrrolic ring remains unprotonated. Moreover, in water suspension, complexation of the interlayer cation is possible with the Cu- and Co-exchanged clays. Sn(IV)TPyP is also stable on the surface under these conditions but air drying the clay induces demetallation. The process is reversible. X-Ray diffraction data on clays having adsorbed TPyP have shown that a continuous layer of these molecules is formed in the interlamellar space. The observed thickness of the intercalated layer suggests that tilting the meso-substituents with respect to the tetrapyrrolic ring is more pronounced than in any related forms of free compounds. As previously proposed (Stone and Fleisher, J. Amer. Chem. Soc. 90, 2735 (1968)) this distortion should increase the resonance interaction between the substituents and the tetrapyrrolic ring and it could be responsible for the large bathochromic shift of the TPyP Soret band in the clay.

On the determination of the surface fractal dimension of powders by granulometric analysis

Abstract We investigate the physical meaning of the exponent derived from the particle size ( R ) dependence of the surface area ( S ) of a solid powder, in cases where a power law S ∼ R α is observed. In general, when the surface area is measured at constant apparent volume, α = D s − 3, as derived by P. Pfeifer and D. Avnir ( J. Chem. Phys. 79 , 3558 (1983)). D s is the surface or textural fractal dimension of the particles, at least in the dimension range spanned by the particle size. On the other hand, when S is measured at constant mass, α = D s − D m , where D m is the mass fractal dimension of the particles, and the knowledge of D m is required in order to determine D s . D m is 3 either for compact or, on the opposite, for very porous particles in which the “external” surface area is negligible with respect to the “internal” surface area. In all the other cases, where the internal morphology of the particles is unknown, D m can be measured from the particle size dependence of the apparent density of the powder, ϱ a , which scales as R D m −3 .

Reaction of molecular hydrogen with the 100 face of MoO3: II. Kinetics initiated by atomic hydrogen and characterization of the surface electronic state

Abstract In order to provide a physical background to the model proposed in Part I, the kinetics of the reaction between the 100 face of a MoO 3 single crystal and, first a mixture of 4% atomic hydrogen in H 2 , and later molecular hydrogen only, has been studied. The rate processes of the activation step and of the stationary step are entirely comparable with those observed for one single crystal surface loaded with platinum particles. Thus atomic hydrogen in the gas phase as well as atomic hydrogen produced by the dissociation of molecular hydrogen on a Pt particle may prepare the favourable surface state able to dissociatively chemisorb molecular hydrogen, ruling out — once again — the classical model of the hydrogen spillover process. This “favourable” state has a Fermi level which is 0.25 eV lower than that of initial MoO 3 , as shown by measuring the work function with a Kelvin probe. This lowering is in good agreement with the variation of the free energy between MoO 3 and H 1.6 MoO 3 , measured electrochemically by others. This suggests that the protons inserted into the surface layers transform the initial MoO 3 layers into layers with composition H 1.6 MoO 3 . The starting material is thus transformed into a biphasic system, the diffusion of the reaction boundary between the two types of layers being the overall rate limiting process. The Fermi energy of H 1.6 MoO 3 being known, it is possible to show that in transformed surface layers the conduction and the valence bands overlap, in agreement with the approximate profile of this band observed by XPS for a reacted single crystal surface. The d character of this band would explain why molecular hydrogen can be dissociatively chemisorbed when this favourable surface state is obtained. The fast electron delocalization within the Mo-O-Mo bonds yields fast oscillations in the oxidation states of the molybdenum atom in the surface layer, accounting for the presence of “oxidized” and “reduced” sites. The formal equation observed in Part I for the rate of stationary step is therefore explained, the impinging H 2 molecules reducing temporarily the oxidized fraction of the surface.

Reaction of molecular hydrogen with the 100 face of MoO3: I. Kinetics in the low pressure range (10−8–10−6 Torr) in the presence of platinum particles

Abstract Three successive processes are observed when the 100 face of a MoO 3 single crystal covered with a specific density of platinum particles of known diameter is exposed to molecular hydrogen in a pressure ( P ) domain between 10 −8 and 10 −6 Torr. During the first step, called the activation step, the rate process is a function of the surface area of the Pt particles and it is slightly thermally activated. During this activation step, H 2 dissociatively chemisorbed on the platinum surface and the hydrogen atoms are inserted into the MoO 3 surface layers. The inserted atom is partially ionized, the electrons given to the lattice reducing Mo 6+ into Mo 5+ and Mo 4+ . The rate of hydrogen insertion ( V a ) in the activation step increases about linearly with time. The activation step is followed by a stationary step where V is independent of the extent of the Pt surface, V = 0.7 Z (1− X ), where Z is the number of collisions between H 2 and the MoO 3 surface and X the electronically reduced fraction of the surface area. X = a P (1 + a P ) , where a is a function of the temperature only. In the stationary step, the platinum particles do not play any role and molecular hydrogen is dissociatively chemisorbed on the MoO 3 surface. The stationary step is followed by a deceleration step. The slowing down of the reaction rate is due to the formation of fractures within the crystal. During all these three steps, the surface is self-cleaning, since protons diffuse into the bulk. The final product of the insertion process would be H 1.6 MoO 3 , the properties of which have been studied in detail elsewhere. The model which is summarized above is very different from that which would result from the classical views on the so-called hydrogen spillover process since the Pt surface producing the H atoms spilling over the surface would operate only during the activation step.

Structural aspects in the photooxidation and photoreduction of water in clay mineral suspensions

Abstract The photosensitized cleavage of water is hampered by fast recombination of usually charged intermediates. Charged surfaces carrying the appropriate catalyst for both the oxidation and reduction of the substrate might be used to compete with this recombination. In this study negatively charged clay minerals and positively charged Al(OH)x gels were examined as catalyst supports capable of exerting such electrostatic interactions. Eu3+ cations were used as quenchers of the excited state of the photosensitizer and vectors of the absorbed energy. It is shown that electrostatic effects do play a role in such systems, and that the tridimensional structure of the catalyst supports must be taken into account when designing a heterogeneous photocatalytic water cleavage system.

Photo-oxidation of water on the surface of hectorite using trans-diaquabis-(2,2′-bipyridine)ruthenium(2+) as catalyst

trans-Ru(bpy)2(H2O)22+(bpy = 2,2′-bipyridine) was shown to be an efficient catalyst for the oxidation of water and allowed the construction of a catalytic water photo-oxidation system within an adsorbed layer.

ADSORPTION OF SAFRANINE BY Na +, Ni 2+ AND Fe 3 + MONTMORILLONITES

The adsorption of the cationic oxidized safranine S+ by a Na+, Ni2+ and Fe3+ montmorillonite has been studied with X-ray powder diffraction, u.v., visible and i.r. spectroscopy. In solution S+ may be protonated: S+, SH2+ and SH23+ have characteristic spectra in the 500–600 nm region where the clay structure does not absorb. In the Na+ as well as in the Ni2+ and Fe3+ clays, the adsorption of S+ is a cation exchange process accompanied by the protonation of the adsorbed dye such as variable concentrations of M+ (Na+, Ni2+ or Fe3+), S+ and SH2+ are simultaneously present. Protonation activity decreases from Fe3+ to Ni2+ and Na+, being the protonation site the amine group as shown by i.r. In the interlamellar space it seems that a SH2+.. S+ association exists that could be described as a sandwich structure 6.5 Å thick.

Clay Minerals: A Molecular Approach to Their (Fractal) Microstructure

For most users, clays are just a class of mineral materials characterized by various properties such as a very small particle size, a strong adsorption capacity, a soft touch, or a plastic behavior when wetted. For geologists and soil scientists, clays are commonly defined as the fine fraction of rocks and soils, with an upper limit for the particle size at 2 μm. It turns out that this purely granulometric definition corresponds rather closely to a particular class of hydrous silicates with layer structures, which belong to the larger group of phyllosilicates.