Membranes titania

When these ID nanocrystals are deposited on a surface, typically a random direction exists for the elongating axis of each ID nanocrystal. However, if the preparation method leads to an alignment of the elongating axis in one single direction, e.g. when an ordered array is present, quasi-2E) nanostructures are present. This class of nanostructured Titania thin film may be extended to include also ordered mesoporous Titania membranes, e.g. when the pores are aligned in one single direction. 

Figure 2B. Typical examples of a pore size distribution for (a) y-alumina membranes desorption branch (b) anatase titania membranes desorption branch.

The results obtained by Burggraaf, Keizer and van Hassel (1989a, b) differ markedly from those reported above by Cot, Guizard and Larbot (1988). The preparation of defect-free titania membranes was found to be much more... 

Characteristic microstructural properties of TiOj membranes produced in this way are given in Table 2.5. Mean pore diameters of 4-5 nm were obtained after heat treatment at T < 500°C. The pore size distribution was narrow in this case and the particle size in the membrane layer was about 5 nm. Anderson et al. (1988) discuss sol/gel chemistry and the formation of nonsupported titania membranes using the colloidal suspension synthesis of the type mentioned above. The particle size in the colloidal dispersion increased with the H/Ti ratio from 80 nm (H /Ti = 0.4, minimum gelling volume) to 140 nm (H /Ti " — 1.0). The membranes, thus prepared, had microstructural characteristics similar to those reported in Table 2.5 and are composed mainly of 20 nm anatase particles. Considerable problems were encountered in membrane synthesis with the polymeric gel route. Anderson et al. (1988) report that clear polymeric sols without precipitates could be produced using initial water concentrations up to 16 mole per mole Ti. Transparent gels could be obtained only when the molar ratio of H2O to Ti is < 4. Gels with up to 12 wt.% T1O2 could be produced provided a low pH is used (H /Ti + < 0.025). 

Less information is available about the stability of ceramic membranes. It is generally thought that ceramic membranes have excellent solvent stability. Acid conditions may be more problematic it was shown [57] that an alumina nanofiltra-tion membrane was very sensitive to corrosion effects in dynamic experiments, whereas the performance of a similar titania membrane was stable in the pH range from 1.5 to 13. 

Similar trends are developing for ceramic membranes applied in pervaporation and nanofiltration, although much slower because ceramic pervaporation and nanofiltration membranes are still sparsely available more experimental observations and experience with applications are needed in this field. Promising results were obtained by Sekulic et al. [61] for titania membranes that can be used in pervaporation as well as nanofiltration. 

Permeable layer reactors consist of thin, porous metal or ceramic substrates onto which titanium dioxide is coated in a manner that allows for flowing fhrough the porous substrates. The flow can be either perpendicular to the surface of the porous media or alternatively may combine perpendicular and parallel vectors. An example of such a reactor was presented by Tsuru et al. (2006). Here, a titania membrane (pore sizes 2.5-22 nm) was prepared... 

Figure 4.26 Rejection coefficient of bovine serum albumin (BSA) protein as a function of time for untreated titania membrane and two titania membranes treated with 0.1 M phosphoric acid at room temperature for 1 and 14 hours, respectively. (A) without phosphate buffer (B) with 0.01 M Na2HP04/NaH2P04 buffer [Randon el al.. 1995]...

Although porous glass membranes have been around for some time, alumina membranes are finding more uses than other inorganic membranes. Zirconia membranes are also receiving much attention. Porous metal membranes are available and are used to less extent due to their unit costs. New inorganic membranes such as titania membranes are emerging. 

Apparently at a temperature above 300 C, the oxidation kinetics of NOx and ammonia gas is so fast that slip of the reactants, when fed from the opposite sides of an alumina or alumina-titania membrane, can be avoided. Vanadium oxide, used as the catalyst for the reaction, is impregnated onto the membrane pore surface. The conversion of NOx reach 70% with the selectivity for nitrogen up to 75% in the temperature range of 300 to 350 C [Zaspalis etal., I991d]. 

Titania membranes prepared at a temperature lower than about 350 C are essentially amorphous. At 350 or so, phase transition to a crystalline phase of anatase begins to occur. The transformation to anatase (tetragonal in crystallinity) is complete and the new phase of rutile (also tetragonal) begins at a temperature close to 450 0. Transformation to rutile is complete at about 600 C [Xu and Anderson, 1989]. Thus, at a temperature between 450 and bOO C, both anatase and rutile phases are present. It has been suggested that this temperature range may be lower at 350-550 C [Larbot et al., 1986 Burggraaf etal., 1989]. 

A sharp change in porosity of the titania membranes occurs at 350 C. This may be attributed to the crystallization of the constituent titania particles [Xu and Anderson, 1994]. Thus crystallization is likely to be a key factor affecting thermal stability of sol-gel derived titania membranes. 

Ahmad, A.L., Othman, M.R., and Mukhtar, H. H2 separation from binary gas mixture using coated alumina-titania membrane by sol-gel technique at high-temperature region. International Journal of... 

Nanofiltration. This new process involves a low rejection for salts (monovalent ions) and ionised organics (MW < 100) in combination with a high rejection of salts (multivalent ions) and organics with MW > 300 which should be separated from the earlier ones at low operating pressures compared with reverse osmosis. Examples are given in Chapter 11 and zirconia or titania membranes seem to be especially suitable due to their relatively good chemical stability. 

A variety of membrane materials has been investigated and reported [1,9] and an overview of commercially available systems has been given by Flsieh [9]. Alumina, zirconia and, more recently, titania membranes are used in large-scale applications. The more complex shapes, i.e. monolith and honeycomb, are almost exclusively based on a-alumina or cordierite. 

Stress patterns of titania without additives were always highly irregular with many peaks and rather low stress levels, indicating severe cracking of the top layer [13,31]. Additives are necessary to relax this problem (see below). Alumina-titania membranes with 25-40 mol% titania yield stress (deflection) curves without peaks (and so without observable cracking). 

These observations support the phenomenological experience that production of defect-free titania membranes by sol-gel techniques is more difficult compared with y-alumina membranes. 

PVA addition to zirconia precursor solutions resulted in an increase of the membrane pore volume and the pore size compared to samples without PVA. Similar effects were found for titania membranes but at lower temperatures indicating a relatively small thermal stability of these titania membranes. This can be improved by doping with alumina [39,40]. 

Kumar [39] reports a considerably larger thermal stability for titania membranes in the rutile phase instead of the usual anatase form. The effect of the support on thermal stability has been reported by Kumar et al. [40,41]. Pure, non-supported titania (anatase) membranes lose their porosity completely when calcined at 600°C for 8 h, where as the supported titania membrane retained ca 30% porosity at 900°C (8 h). Unsupported titania-(50 wt%)alumina composite membranes retained a porosity of ca 40% at 700°C (8 h), supported ones retained porosity even at 900°C. 

Finally, other examples of thermal behaviour of zirconia and titania membranes have been reported by Larbot et al. in a series of papers, e.g. Ref. [33]. 

H. Verweij, K.N.P. Kumar, K. Keizer and A.J. Burggraaf, Formation and formation mechanisms of alumina and titania membranes, in Yi Hua Ma (Ed.), Proceedings of the Third International Conference on Inorganic Membranes (ICIM3) July 10-14,1993. Worcester Polytechnic Institue, Worcester, MA 016109, USA, pp. 27-35. 

K.N.P. Kumar, K. Keizer, A.J. Burggraaf, T. Okubo and H. Nagamoto, Textural evolution and phase transformations in titania membranes. Part 2. Supported membranes. /. Mater. Chem., 3 (1993) 1151-1159. 

Labbez. C. et al.. Evaluation of the DSPM model on a titania membrane Measurements of charged and uncharged solute retention, elecUokinetic charge, pore size, and water permeability,. 1. Colloid Interf. Sci., 262, 200, 2003. 

Wang, Y.H. et al., Titania membranes preparation with chemical stability for very hash environments applications, J. Membr. Sci., 280, 261, 2006. 

Tsuru, T. et al., Titania membranes for liquid phase separation Effect of surface charge on flux, Separ. Purif. Technol., 25, 307, 2001. 

TiO2, and casting. The results for water uptake, lEC and single cell performance were compared with the commercial Nafion-115 and home-made, recast Nafion membranes. Power density values of 0.51 and 0.26 Wcm at 0.56 V were obtained at 110 and 130 °C, respectively, for the composite Nafion-titania membrane [46]. 

Sekuli J, ten Elshof JE, Blank DHA. A microporous titania membrane for nanofiltration and pervaporation. Adv Mater. 2004 16(17) 1546-50. 

---