Rutile, water adsorption

M. L. Machesky, D.J. Wesolowski, Ion adsorption at the rutile-water interface Linking molecular and macroscopic properties, Langmuir 20 (2004) 4954-4969. 

Some observations, relevant to the effect of the UV illumination upon energetics of water adsorption on the (100) plane of rutile Ti02, have been made by Lo et al.. The band-gap (hv > 3 eV) irradiation of the prereduced, Ti -rich rutile surface, containing adsorbed water, resulted in a distinct increase of the work function, indicating a possible change of the nature of adsorbed species. 

Hallabaugh and Chessick [34] drew a similar conclusion from the fact that the cross-sectional area of the water molecule in the monolayer on Ti02 (rutile) was extremely large, i.e. 23.5A. Also Morimoto et al. [35] arrived at a similar conclusion in their analysis of water adsorption isotherms on rutile. They proposed the following scheme for the secondary adsorption ... 

The solid line in Fig. 6 is the function x( ) for argon adsorbed onto rutile, obtained by averaging x(c) calculated by two different methods. Thus, it can be clearly seen that x i) should, in this case, be well smoothed by a gaussian-like function. This statement should be true for both argon and water adsorption, because the interaction with surafce oxygens is the key factor in both cases. 

With his model, Sverjensky (2001) predicted different distances for the adsorption of different electrolyte cations (i.e., Rb+ = 3.3 A, Sr2+ = 2.9 A) at the rutile-water interface that compared well to the distances reported from x-ray standing-wave experiments (Fenter et al., 2000). The model also suggested that trace amounts of metals (e g., Sr"" ", Ca +j other than the electrolyte cations should form inner-sphere complexes if adsorbed to the p-plane of rutile and similar sohds, and form outer-sphere complexes if adsorbed to the p-plane of quartz, goethite, and similar solids. These predictions were consistent with the results of x-ray standmg-wave and EXAFS studies (Axe et al., 1998 Fenter et al., 2000 O Day et al., 2000 Saliai et al., 2000). 

Ridley. M.K. et al.. Calcium adsorption at the rutile-water interface A potentiomet-ric study in NaCl media to 250 °C, Geochim. Cosmochim. Acta, 63, 3087, 1999. 

Jang, H.M. and Fuerstenau, D.W., The specific adsorption of alkaline-earth at the rutile/water interface. Colloids Surf., 21, 235, 1986. 

The surface hydration-hydroxylation structure of titania, proved previously mainly by IR studies using dry titania powders, also seems to hold when these powders are dispersed in water. An interesting approach, therefore, is to probe directly the uptake of water from the gas phase by DS-coated rutile surfaces (42). Water adsorption isotherms are presented in Figure 14. The dual nature of titania surface sites, a property not seen... 

The surface hydration-hydroxylation structure of titania, proved previously mainly by IR studies using dry titania powders, also seems to hold when these powders are dispersed in water. An interesting approach, therefore, is to probe directly the uptake of water from the gas phase by DS-coated rutile surfaces [42]. Water adsorption isotherms are presented in Figure 52.14. The dual nature of titania surface sites, a property not seen with other common oxides such as silica and alumina, leads to an unusual type of water adsorption isotherm for titania. The isotherm shows two distinct knees (Figure 52.14) connected by a region where adsorption increases linearly with the partial vapor pressure of water. The explanation for this adsorption behavior is rather complex [42] and beyond the scope of this chapter. This behavior is believed to be due to the presence of hydrated surface cation sites. 

Table 11.10. MSINDO adsorption energies (kJ/mol) for water adsorption on rutile (110) (relaxed cluster calculations), [770]...

Figure 2 Embedded cluster simulation of water adsorption on rutile...

In the pioneering study of rutile by Hollabaugh and Chessick, adsorption isotherms of water were determined after an outgassing at 450°C, and were repeated after evacuation at 90°C at the end of the run. The... 

In Fig. 5.21, from Dawson s paper, the uptake at X for the 250°C-outgassed sample is dose to the calculated value for a monolayer of water with a (H20) = 101 A. Point X has therefore been ascribed to a close-packed monolayer of water on a hydroxylated surface of rutile. The fact that the differential entropy of adsorption relative to the liquid state (calculated from the isosteric heat of adsorption) changes sharply from negative to positive values in this region with A s 0 at X was regarded as supporting evidence. ... 

The surface chemistry of coesite and stishovite was studied by Stiiber (296). The packing density of hydroxyl groups was estimated from the water vapor adsorption. More adsorption sites per unit surface area were found with silica of higher density. Stishovite is especially interesting since it is not attacked by hydrofluoric acid. Coesite is dissolved slowly. The resistance of stishovite is ascribed to the fact that silicon already has a coordination number of six. Dissolution of silica to HaSiFg by hydrogen fluoride is a nucleophilic attack. It is not possible when the coordination sphere of silicon is filled completely. In contrast, stishovite dissolves with an appreciable rate in water buffered to pH 8.2. The surface chemistry of. stishovite should be similar to that of its analog, rutile. 

Hollabaugh and Chessick (301) concluded from adsorption studies with water, m-propanol, and w-butyl chloride that the surface of rutile is covered with hydroxyl groups. After evacuation at 450°, a definite chemisorption of water vapor was observed as well as of n-propanol. The adsorption of -butyl chloride was very little influenced by the outgassing temperature of the rutile sample (90 and 450°). A type I adsorption isotherm was observed after outgassing at 450°. Apparently surface esters had formed, forming a hydrocarbonlike surface. No further vapor was physically adsorbed up to high relative pressures. 

A packing density of 6.6 Ti + ions per 100 A was estimated on a theoretical basis by Hollabaugh and Chessick (301). From the values of irreversible and reversible water vapor adsorption, a surface density of 3.7 OH/100 A was calculated for the substance activated at 450° and of 11.4 OH/100 A for a fully hydroxylated rutile surface. 

Munuera and Stone (142) successfully applied the model of a (110) plane to their adsorption studies of water, isopropanol, and acetone. The dry (110) face exposes 5.1 five-coordinate Ti4+ ions. Half of these were able to chemisorb water dissociatively, leading to the formation of two types of surface OH groups. The activation energy for desorption of this water was 107 kJ mole-1. At this state of hydroxylation the rutile surface consists of isolated five-coordinated Ti4+ ions, which on further addition of water coordinate water molecu-larly, of isolated O2- ions, and of pairs of OH groups, one being monodentate and the other bidentate with respect to the cations. 

IV sites. A protonated pyridine has never been observed, and the Lewis acid sites on titanium oxides cannot be converted into Br0nsted sites by water vapor adsorption (217). Although Jones and Hockey (216) suggest that the chemistry of surface hydroxyl on rutile corresponds more closely to that of the OH" ion rather than that of the hydroxyl group, no surface reactions similar to that observed with alumina [Eq. (14)] have since been reported. 

Fig. 28. Alternatives sites of a molecular adsorption of water on the (110)-rutile surface cluster (146 atoms) on the central, five fold O-coordinated Ti site (the singly coordinated water molecule) and on the bridging oxygen vacancy (the doubly coordinated water molecule). The corresponding reaction coordinates of the dissociative adsorption that produce two OH species involve the bending of water towards the bridging oxygen of the singly coordinated adsorbate and rotation of the doubly coordinated water...

Consider first the molecular adsorption structure. In this case, the essentially antisymmetric modes y = (2, 148) (notice the symmetry breaking due to the oxygen vacancy on the surface) do not participate in charge displacement between water and rutile (see Figs. 31 A and 32 A), so that the CT-reactivity information is basically limited to the four reactive IRM which originate from the two components on each reactant. However, in the transition-state struc-... 

Fig. 31. (Continued). The complementary reactive (delocalized) IRM for the molecular adsorption (Part A) and transition-state (Part B) chemisorption complexes of Fig. 30. The numerical data are reported in the same order as in Fig. 21. The CT parameters have been calculated for the assumed water (base, B) - rutile (acid, A) electron transfer...

It was first thought that the changes in BET area were due to the presence of micro-pores in which water and other molecules could be trapped and removed only by an increase in the outgassing temperature. In the light of further adsorption and spectroscopic measurements, it now seems much more likely that these and other effects are associated with the surface chemistry of rutile rather than its porosity. 

Figures 1.9a and b refer to the adsorption of water on rutile fig. (a) gives isosters (plotted as Inpvs. T ) and fig. (b) the isosteric heats, derived from...

---