(528g) Modifying the Photocatalytic and Dispersion Properties of TiO2 P25 Nanoparticles By Scalable Molecular Layer Deposition

Titanium dioxide (TiO₂) is used for a range of applications, with a global market of over 6 million metric tonnes per annum. Almost 2/3 of this demand stems from the paints and coatings industries, where the high efficiency with which TiO₂ scatters visible light imparts highly favourable characteristics for its use as a bright white pigment. TiO₂ also acts as an effective UV-screening agent, and nanoscale TiO₂ is an increasingly important constituent of various cosmetic products, most notably as a UV blocker in sunscreens. The excellent performance of TiO₂ as a UV-screening agent stems in part from its semi-conductive properties, which makes it absorptive in the UV spectrum. This property is of particular interest for photocatalytic applications, where nanoscale TiO₂ has been studied as a potential water cleaning agent, degrading organic pollutants through the photocatalytic generation of reactive oxygen species (ROS). Although this generation of free radical species is utilised in certain self-cleaning coatings, for other applications where TiO₂ is used as a white pigment it is highly unfavourable, reducing product life-time through accelerated degradation. The high-molecular weight polymers in the organic binders found in coating applications are particularly susceptible to photochemically induced chain-scission reactions. The production of these oxidative species is also undesirable in cosmetic products, where the use of TiO₂ in sunscreen has raised concerns of potential carcinogenic, and mutagenic effects.

Due to these problems, considerable attention has been placed on finding approaches by which the photocatalytic activity of TiO₂ P25 nanoparticles can be suppressed, preferably without compromising its white pigmentation. Most TiO₂ commercially produced undergoes some form of surface treatment, not only to reduce photo-reduction of final use products, but also to improve the dispersibility of the TiO₂ particles in water and a range of organic liquids. Inorganic coatings are typically used for this purpose, with alumina and silica films the most commonly utilised metal oxide films. Alumina coatings as thin as 1 nm were found to effectively suppress the photodegrading properties of TiO₂. Although less used, organic coatings are of increasing interest for this photo-suppressive effect, often providing advantages in biocompatibility, lower raw material costs, and environmental considerations. The biocompatibility of polymeric films is of particular interest to the cosmetics industry, and also promises applicability to fields such as nanomedicine. A major advantage of using organic rather than inorganic coatings is the opportunity to improve the dispersibility of the underlying particle substrate in organic media. Not only is this promising for the aforementioned applications of TiO₂ to organic-based paints and cosmetics, but it may also be of interest for the production of polymer nanocomposites. Certain industrial polymers such as PVC are UV-sensitive, and various photo-stabilising systems have been investigated as a means to reduce their degradation rates. Two key issues preventing TiO₂ from fulfilling this role are its photocatalytic activity, which can enhance degradation rates, and difficulties in adequately dispersing the nanoparticles throughout the polymer. Therefore a reliable, scalable approach to produce organic coatings on TiO₂ may be of some value to polymer research and manufacture.

Typically, the production of these organic coatings on the particle substrates has been achieved through liquid-phase techniques. Liquid approaches are well established, providing versatility and a large degree of control over experimental conditions, however they provide non-conformal coverage and produce excessive amounts of waste. The requirement for multiple, at times complex synthesis steps, is also undesirable, as is the frequent need for post-processing separation and drying stages. Vapour phase approaches remove these steps, and prevent the undesirable effects of wetting and surface tension which can arise when using liquid-phase approaches. Chemical vapour deposition (CVD) is the most widespread vapour-phase coating process, and the introduction of both monomers in a single step allows simultaneous polymerisation and thin film deposition. However the coatings produced from this method can be non-conformal, reducing their capacity for surface passivation. To combat these issues, thicker films than desirable are typically produced to ensure complete surface coverage, however this is wasteful and expensive. Alternating or sequential vapour deposition polymerisation (VDP) was developed to provide greater control over film growth, and the formation of highly orientated polymer thin films is attributed to the interchanging reaction and deposition process achieved by introducing alternate monomers to the vacuum chamber. Although the control provided by alternating VDP is superior to that of CVD, monomers which remain present in the vapour phase following their introduction will react with sequentially introduced monomers, reducing the extent to which film growth can be quantifiably predicted.

The failings inherent to the above processes led to the development of Molecular layer deposition (MLD) to obtain the controlled growth of ultrathin organic films. Like sequential VDP, MLD is a vapour phase process which involves the cyclical introduction of organic precursors into the reaction chamber, however in MLD each pulse of precursor is separated by a purge step with an inert gas, minimising uncontrolled CVD growth. The consecutive, self-limiting chemical reactions between the substrate surface and the gaseous precursors enables the controlled, layer-by-layer growth of conformal polymeric coatings on a range of substrates.

In this work, we deposit ultrathin PET films on gram-scale batches of TiO₂ P25 nanoparticles (NPs) using terephthaloyl chloride and ethylene glycol as precursors. The MLD process is carried out in an atmospheric-pressure fluidized bed reactor at 150 °C for a broad range of growth cycles (i.e., from 10 to 50). Ex-situ diffuse reflectance infrared fourier transform spectroscopy (DRIFTS-FTIR) shows the presence of the characteristic C=O stretch of ester groups in the range 1730-1750 cm^-1, thus demonstrating the successful MLD reaction. DRIFTS-FTIR and thermogravimetric analysis (TGA) confirm the self-saturating behaviour of the precursors, and the linear increase in mass with the number of cycles. Moreover, TGA highlights the good thermal stability of the PET films, which are stable up to ~225 °C. Transmission electron microscopy (TEM) enables the visualization of uniform PET films on the surface of TiO₂ NPs. TEM observations suggest a low PET growth per cycle (GPC), around 0.06 nm. This value is also in agreement with the GPC estimated by the amount of deposited material on TiO₂ NPs measured by TGA as well as with the GPC of PET on SiO₂ wafers measured by spectroscopic ellipsometry. UV photocatalytic dye degradation tests demonstrate the efficacy of the produced PET films in suppressing the photocatalytic activity of the underlying TiO₂ P25 NPs. Still, UV-Vis diffuse reflectance spectroscopy measurements show that the absorption spectra of the PET-coated TiO₂ are virtually unaffected by the PET coating, indicating that the bulk optical properties (e.g., white color and brightness) of TiO₂ are retained. Furthermore, zeta potential measurements in aqueous solutions indicate greater stability of the suspensions with PET-coated TiO₂ than with uncoated TiO₂. Finally, to demonstrate the improvement of PET-coated TiO₂ in organic media, we disperse both the uncoated and PET-coated powders in mixtures of an apolar organic solvent (diethyl ether) and water. While uncoated TiO₂, being hydrophilic, stays in the aqueous phase, PET-coated TiO₂ stays in diethyl ether, thus indicating clear hydrophobization of the particles upon the MLD process.

TiO₂ P25 is a well-known photo-catalyst, serving as an ideal model substrate to assess the capability of MLD produced organic coatings to suppress the generation of free radical species. PET, a mass produced plastic, was selected on the grounds of its ideal characteristics of good chemical stability, relatively high thermal stability, and biocompatibility. The precursors used to produce the PET coating (terephthaloyl chloride and ethylene glycol) are also widespread in manufacturing, further demonstrating the industrial relevance of this study. Finally, by utilising a fluidized bed reactor operating at atmospheric pressure, we demonstrate a scalable process currently capable of depositing organic coatings on gram-scale quantities of nanopowder. This contrasts with the approach of most MLD studies, utilising highly controlled, low-pressure systems which are both complex and expensive. We therefore believe that this study is among the first demonstrations of a commercially viable MLD process with real world applicability.