Technical Field

The present invention relates to a method for preparing metal oxide thin films. Background

Semiconducting metal oxides are an important class of functional materials especially when prepared into thin films. An example of a commonly used semiconducting metal oxide is titanium dioxide (Ti0 ₂ ), which has been used in applications across vastly different fields, such as in thermal or photocatalytic chemical reactions, gas sensors, pigments, dielectric ceramics, microelectronics, photonic materials, optical devices, solar energy conversion, hydrogen storage, inorganic membranes, environmental remediation, microorganism photolysis, and medical treatments, because of its unusual chemical, electronic, and optical properties. In particular, Ti0 ₂ is chemically stable and non-toxic and has a wide band gap of 3.0-3.2 eV.

The preparative techniques for Ti0 ₂ thin films (and other functional metal-oxide thin films) at the present stage are mainly chemical vapour deposition (CVD), physical vapour deposition (PVD), sputtering, and sol-gel dip- and/or spin-coating processes. CVD and PVD both require large high-cost processing equipments, and the difficulty in obtaining large-area thin films is a major drawback for these techniques.

The problem with directing Ti0 ₂ -based sols or colloids directly onto substrate surfaces and then converting them into thin films by different post-treatments is that equipment such as spin-coaters and electrophoretic pools require a large amount of space. Therefore, the Ti0 ₂ films cannot be expanded arbitrarily and at most reach the equipment-equivalent size. Further, for the traditional routes such as the sol-gel process, there are other disadvantageous factors such as a long aging time to get a workable sol, instability of precursor sol stock, and shrinkage or cracking of film.

Traditional methods have also resulted in a weak adhesion between film and substrates due to the low temperature methods used. In spite of efforts to improve the methods of fabricating Ti0 ₂ films, some barriers are still present to design and synthesise Ti0 ₂ films.

There is therefore a need for an improved process to fabricate Ti0 ₂ films. Summary of the invention

The present invention seeks to address at least one of the problems in the prior art, and provides an improved method for preparing a metal oxide thin film.

According to a first aspect, there is provided a method for preparing a metal oxide thin film comprising the steps of: contacting a surface of a substrate with a colloidal suspension of metal oxide nanoparticles for a pre-determined period of time to enable adsorption of the metal oxide nanoparticles onto the surface of the substrate; and heating the surface of the substrate.

According to a particular aspect, the adsorption may comprise self-assembly of the metal oxide nanoparticles on the surface of the substrate. In particular, the metal oxide nanoparticles may form chemical bonds with the surface of the substrate.

The metal oxide nanoparticles may be a nanoparticie of any suitable metal oxide. For example, the metal oxide nanoparticles may comprise nanoparticles of at least one of titanium oxide, zirconium oxide, hafnium oxide, vanadium oxide, niobium oxide, tantalum oxide, tungsten oxide, manganese oxide, iron oxide, cobalt oxide, nickel oxide, copper oxide, zinc oxide, gallium oxide, indium oxide, tin oxide, cerium oxide, barium titanate, strontium titanate, calcium titanate, barium strontium titanate, barium lanthanum, lithium lanthanum titanate, lead titanate, lead zirconium titanate, barium zirconate, lead zirconate, yttrium ferrite, bismuth ferrite, yttrium barium copper oxide, lanthanum manganese oxide, strontium cerium oxide, or a combination thereof. According to a particular aspect, the metal oxide nanoparticles may be titanium oxide (Ti0 ₂ ) nanoparticles.

The metal oxide nanoparticles may be of any suitable size. For example, the metal oxide nanoparticles may comprise at least one dimension of size < 200 nm. In particular, the metal oxide nanoparticles may comprise at least one dimension of size 1-200 nm, 10-175 nm, 15-150 nm, 20-125 nm, 30-100 nm, 40-75 nm, 50-60 nm. Even more in particular, the metal oxide nanoparticles may comprise at least one dimension of < 10 nm. The metal oxide nanoparticles comprised in the colloidal suspension may be capped or coated metal oxide nanoparticles. For example, an organic coating material may be disposed on a surface of the metal oxide nanoparticles to provide the capped or coated metal oxide nanoparticles. The organic coating material may be any suitable material. For example, the organic coating material may be any one of an alkanoic acid, a saturated or unsaturated fatty acid, decanoic acid, oleic acid, an alkylamine, a fatty amine, oleylamine, an alkanol, a fatty alcohol, oleyl alcohol, an alcohol derivative, or a combination thereof. In particular, the organic coating material may be oleic acid. According to a particular aspect, the colloidal suspension may comprise oleic acid- capped Ti0 ₂ nanoparticles.

The concentration of the metal oxide nanoparticles comprised in the colloidal suspension may be any suitable concentration. For example, the concentration of the metal oxide nanoparticles may be≥5 mg/mL. In particular, the concentration of the metal oxide nanoparticles may be 5-20 mg/mL, 10-18 mg/mL, 12-15 mg/mL. Even more ih particular, the concentration of the metal oxide nanoparticles may be about 8-9 mg/mL.

The substrate may be any suitable substrate. For example, the substrate may be a glass, plastic, silicon wafer, or silicon substrate. The glass may be normal or doped glass. According to a particular aspect, the substrate may be a glass substrate. The substrate may be a patterned or unpatterned substrate. According to a particular aspect, when the substrate is a patterned substrate, the method may further comprise a step of patterning the substrate prior to the contacting.

The contacting may be any suitable form of contacting the surface of the substrate with the colloidal suspension. For example, the contacting may be by partially or completely dipping the substrate into the colloidal suspension.

The pre-determined period of time may be any suitable amount of time for the purposes of the present invention. For example, the pre-determined period of time may be≥15 minutes. In particular, the pre-determined period of time may be 15 minutes - 8 hours, 30 minutes - 6 hours, 45 minutes - 5 hours, 1 -4 hours, 1.5-3 hours, 2-2.5 hours. Even more in particular, the pre-determined period of time may be about 2 hours. The heating may comprise heating the substrate surface at a suitable temperature For example, the heating may comprise heating the substrate surface at a temperature of ≥ 150°C. In particular, the heating may comprise heating the substrate at a temperature of 150-700°C, 200-600°C, 250-550°C. In particular, the heating may be carried out at a temperature of 300-450°C. Even more in particular, the heating may be carried out at a temperature of about 400°C.

The heating may be carried out for a suitable period of time. For example, the heating may be carried out for at least 5 minutes. In particular, the heating may be carried out for 5 minutes - 8 hours, 10 minutes - 7 hours, 30 minutes - 6 hours, 1-5 hours, 2-4 hours, 2.5-3.5 hours. Even more in particular, the heating may be carried out for about 2 hours.

According to a particular aspect, the method may comprise repeating the contacting and the heating to form a further metal oxide thin film on the surface of the substrate having a preceding metal oxide thin film, thereby increasing the thickness of the metal oxide thin film. In particular, the method may comprise repeating the contacting and the heating such that the metal oxide nanoparticles comprised in the further metal oxide thin film adsorb onto the surface of the preceding metal oxide thin film. Iri particular, the contacting and the heating may be repeated at least one more time.

When the contacting and the heating is repeated, the metal oxide nanoparticles comprised in the further metal oxide thin film may be the same or different from the metal oxide nanoparticles comprised in the preceding metal oxide thin film.

According to a second aspect, there is provided a metal oxide thin film obtained from the method according to the first aspect of the present invention. The present invention further provides an article of manufacture comprising the metal oxide thin film prepared from the method according to the method as described above. The article of manufacture may be any suitable article of manufacture. For example, the article of manufacture may be, but not limited to, a photovoltaic device, a window, or a solar cell.

Brief Description of the Drawings

In order that the invention may be fully understood and readily put into practical effect there shall now be described by way of non-limitative example only exemplary embodiments, the description being with reference to the accompanying illustrative drawings. In the drawings:

Figure 1 is a flow chart showing the general method of preparing a metal oxide thin film according to the present invention;

Figure 2 (A) and (B) show alternative embodiments for contacting a substrate surface with a colloidal suspension of metal oxide nanoparticles;

Figure 3 shows a schematic representation of the steps involved in the method of the present invention;

Figure 4 shows a schematic representation of the steps involved in the method of the present invention when the substrate is a patterned substrate;

Figure 5 (A) and (B) show TEM images of colloidal Ti0 ₂ nanoparticles at different magnifications;

Figure 6 shows the FESEM images of Ti0 ₂ films on soda lime glass substrates viewed from the top at different magnifications (all scale bars in the images denote 100 nm). (A) and (B) are images after 1 process cycle (S1 ), (C) and (D) are images after 2 process cycles (S2), (E) and (F) are images after 3 process cycles (S3), and (G) and (H) are images after 4 process cycles (S4);

Figure 7 shows (A) optical transmittance of samples S1 , S2, S3 and S4 (Figure 6). Inset shows higher-resolution transmittance spectra with the normal incidence. Water contact angle image reveals the superhydrophiiic nature of freshly prepared Ti0 ₂ films; and (B) shows optical transmittance of samples versus varied incident angles of light at the wavelength of 400 nm: (a) equipment reference, (b) bare soda lime glass, (c) S1 , (d) S2, (e) S3, and (f) S4;

Figure 8 shows FESEM images of Ti0 ₂ thin films on borofloat glass substrates at different magnifications: (a-b) 1 process cycle, (c-d) 2 process cycles, (e-f) 3 process cycles, and (g-h) 4 process cycles; all scale bars in the images are equal to 100 nm;

Figure 9 shows (a, b) FESEM images of Ti0 ₂ thin films on pillar-patterned borofloat glass substrates at different magnifications, (c, d) FESEM images of Ti0 ₂ thin films on hole-patterned borofloat glass substrates at different magnifications, (e) transmittance spectra before and after formation of Ti0 ₂ film on hole-patterned borofloat glass (images of c and d), and (f) water contact angle photos of hole-patterned glass before and after formation of the Ti0 ₂ film;

Figure 10 shows Ti0 ₂ film prepared on a larger borofloat glass with dimensions of 12.5 cm x 12.5 cm χ 2.0 mm: (a-b) FESEM images (all scale bars are equal to 100 nm) of this film at different magnifications, (c) optical transmittance spectra for bare glass and glass with Ti0 ₂ film, and the inserted photo shows the water contact angle, and (d) visual transparency of the film (size: about 12.5 cm ^χ 12.5 cm ^χ 2.0 mm);

Figure 1 1 shows the XPS spectra of C 1s, O 1s, Si 2p, and Ti 2p photoelectrons of (a) Ti0 ₂ film on soda lime glass substrate, and (b) Ti0 ₂ fiim on borofloat glass substrate;

Figure 12 shows (A) absorption spectra of methyl orange solution (5 mg/L, 4 ml_) without and with various Ti0 ₂ films on soda lime glass substrates under exposure to the UV light lamp for 6 h: (a) initial methyl orange solution, (b) without any catalyst, (c) S1 (prepared by 1 process cycle), (d) S2 (by 2 process cycles), (e) S3 (by 3 process cycles), (f) S4 (by 4 process cycles), (g) sample prepared in the same way as S1 but with a soaking time of 4 h, (h) sample prepared in the same way as S1 but with a soaking time of 6 h, (i) sample prepared in the same way as S1 but with a soaking time of 8 h, (j) sample prepared in the same way as S1 but with a soaking time of 12 days, and (B) the corresponding C _t /C ₀ plot versus the sample series;

Figure 13 shows evaluation of photocatalytic performance of the Ti0 ₂ films for photodegradation of adsorbed methyl orange under UV-irradiation: (a) S-1 (prepared by 1 process cycle), (b) S-2 (by 2 process cycles), (c) S-3 (by 3 process cycles), (d) S- 4 (by 4 process cycles), (e) UC(1 process cycle, commercial Ti0 ₂ nanoparticles), (f) UCTC (3 process cycle, commercial Ti0 ₂ nanoparticles). The results indicate that S-2 gives the best catalytic performance among all the samples under identical experimental conditions;

Figure 14 shows FESEM image of Ti0 ₂ film provided by Haruna, Singapore;

Figure 15 shows frictional forces measured with a diamond tip of 1 Mm diameter on soda lime glass substrates coated with the Ti0 ₂ thin films. Plots are shifted in y-scale from each other for better clarity (the thresholds of crack formation are indicated with vertical dashed lines);

Figure 16 shows the FESEM images (scale bar: 100 nm) of Ag nanoparticles on Ti0 ₂ film prepared by UV lamp irradiation for (a) 0.5 h, (b) 1 h, (c) 2 h, (d) 3 h with 8 mL AgN0 ₃ solution in the presence of 0.04 g PVP;

Figure 17 shows absorption spectrum of Ag nanoparticles on Ti0 ₂ fiim under the same conditions with samples displayed in Figure 16: (a) 0.5 h, (b) 1 h, (c) 2 h, (d) 3 h;

Figure 18 shows FESEM (scale bar: 100 nm) images of Ag nanoparticles on Ti0 ₂ film prepared by UV lamp irradiation for (a) 0.5 h, (b) 1 h, (c) 2 h, (d) 3 h with 8 mL AgN0 ₃ solution;

Figure 19 shows absorption spectrum of Ag nanoparticles on Ti0 ₂ film fabricated by irradiation for (a) 0.5 h, (b) 1 h, (c) 2 h, (d) 3 h;

Figure 20 shows EDX spectra of Ag nanoparticles-decorated Ti0 ₂ film;

Figure 21 shows FESEM images (scale bar: 100 nm) of Ag nanoparticles on Ti0 ₂ film prepared by UV lamp irradiation for 0.5 h under different conditions: (a) 0.04 g Tween- 20, (b) 0.06 g Tween-20, (c) 0.1 g Tween-20, (d) 0.06 g Tween-40 with 8 mL 0.00875 mol/L AgN0 ₃ solution;

Figure 22 shows absorption spectrum of Ag nanoparticles on Ti0 ₂ film prepared under different conditions: (a) 0.06 g Tween-40, (b) 0.1 g Tween-20, (c) 0.06 g Tween-20, (d) 0.04 g Tween-20;

Figure 23 shows (a-b) different-magnification TEM images of Ag nanoparticles and Ti0 ₂ film scratched from glass substrates, (c) HRTEM of a typical Ag nanoparticle, (d) the corresponding FFT-HRTEM of the squared place. Inset in (b) is SAED;

Figure 24 shows (A) study on the antibacterial efficiency over E. Coli: (a) E. Coli suspension with Ti0 ₂ film as a control sample, (b) E. Coli suspension with Ag/Ti0 ₂ film prepared by UV irradiation for 0.5 h with 0.04 g PVP under exposure to UV lamp, (c) E. Coli suspension with Ag/Ti0 ₂ film prepared by UV irradiation for 0.5 h with 0.04 g PVP in the dark, (d) E. Coli suspension with Ag Ti0 ₂ film prepared by UV irradiation for 2 h with 0.04 g PVP under exposure to UV lamp, (e) E. Coli suspension with Ag Ti0 ₂ film prepared by UV irradiation for 8 h with 0.04 g PVP under exposure to UV lamp; and (B) concentration of Ag ⁺ for different samples over E. Coli at the end of bactericidal process after 5 h irradiation: (a)-(e) of (B) are the same samples as in (A);

Figure 25 shows TEM images of E. coli irradiated by UV lamp with Ag/Ti0 ₂ composite film for (a-b) 0 h, (c-d) 2 h, (e-f) 5 h;

Figure 26 shows the absorption spectrum of Fe:Ti0 ₂ suspension with different doping concentrations of Fe;

Figure 27 shows the FESEM images of Fe doped Ti0 ₂ thin films formed on glass substrates with different doping concentrations and at different magnifications in which (a) and (b) are images for Fe doping concentration of 0.3%, (c) and (d) are images for Fe doping concentration of 0.5%, (e) and (f) are images for Fe doping concentration of 0.8%, and (g) and (h) are images for Fe doping concentration of 1.2%; and

Figure 28 shows the C/C ₀ photocatalysis curve for different Fe doping concentrations doped in Ti0 ₂ films in comparison with a pure two-layer Ti0 ₂ film.

Detailed Description of the invention

In general terms, the present invention relates to a simple and scalable method for preparing metal oxide thin films. The metal oxide thin films may be prepared without the involvement of high cost processing equipment and allow for adequate controllability over the product quality.

The metal oxide thin films prepared from the method of the present invention have smooth and conformal surfaces with no cracks. There is also good adhesion between the metal oxide thin film and the substrate surface which is an improvement over metal oxide nanoparticie films prepared by physical vapour deposition and chemical vapour deposition techniques. The metal oxide thin films also have an improved photocatalytic performance, high transmittance and are super-hydrophiiic, which therefore provide opportunities for the metal oxide thin films to be used in various applications including but not limited to as coatings on solar panels. In particular, the metal oxide thin films may provide an omni-directional self-cleaning coating on solar panels with very little loss of power efficiency. In view of the low cost and simplicity of the method, scale-up for large area thin film production may also be obtained.

According to a first aspect, there is a provided a method of preparing a metal oxide thin film, comprising the steps of: contacting a surface of a substrate with a colloidal suspension of metal oxide nanoparticles for a pre-determined period of time to enable adsorption of the metal oxide nanoparticles onto the surface of the substrate; and heating the surface of the substrate.

A method 100 for preparing the metal oxide thin film may generally comprise the steps as shown in Figure 1. In particular, the method 100 is a solution-based adsorptive self- assembly approach to fabricate metal oxide thin films on a substrate. Each of the steps of the method 100 will now be described in more detail.

Step 102 comprises obtaining metal oxide nanoparticles. Any suitable metal oxide nanoparticle may be used for the purposes of the present invention. For example, the metal oxide nanoparticle may be a rare earth metal oxide, semiconductor metal oxide, a nanocomposite of two or more metal oxides, doped metal oxides, or alloys.

In particular, the metal oxide nanoparticles may comprise nanoparticles of at least one of titanium oxide, zirconium oxide, hafnium oxide, vanadium oxide, niobium oxide, tantalum oxide, tungsten oxide, manganese oxide, iron oxide, cobalt oxide, nickel oxide, copper oxide, zinc oxide, gallium oxide, indium oxide, tin oxide, cerium oxide, barium titanate, strontium titanate, calcium titanate, barium strontium titanate, barium lanthanum, lithium lanthanum titanate, lead titanate, lead zirconium titanate, barium zirconate, lead zirconate, yttrium ferrite, bismuth ferrite, yttrium barium copper oxide, lanthanum manganese oxide, strontium cerium oxide, or a combination thereof. The metal oxide nanoparticles may be doped with any suitable metal. For example, the metal oxide nanoparticles may be doped with one or more ions of alkali metals, earth alkali metals, transition metals, noble metals or rare earth metals. Further metal oxide nanoparticles known or obvious to a skilled person are also encompassed by the scope of the present invention. The step 102 may comprise obtaining two or more different types of metal oxide nanoparticles. According to a particular embodiment, the step 102 may comprise obtaining titanium oxide (Ti0 ₂ ) nanoparticles. For the purposes of the present invention, a metal oxide nanoparticle is defined as being one which has at least one dimension in the nanoscaie. The step 102 of obtaining metal oxide nanoparticles may comprise obtaining metal oxide nanoparticles of any suitable size. For example, the metal oxide nanoparticles obtained in the step 102 may comprise at least one dimension of size <200 nm. In particular, the metal oxide nanoparticles obtained in step 02 may comprise at least one dimension of size 1-200 nm, 3-190 nm, 5-180 nm, 10-175 nm, 15-150 nm, 20-125 nm, 30-100 nm, 40-75 nm, 50-60 nm. Even more in particular, the metal oxide nanoparticles obtained in step 102 may comprise at least one dimension of < 10 nm. For the purposes of the present invention, the dimension may refer to the average diameter of the metal oxide nanoparticle obtained in the step 102.

The metal oxide nanoparticles obtained in the step 102 may be capped or coated metal oxide nanoparticles. For example, the metal oxide nanoparticles obtained in the step 102 may comprise an organic coating material disposed on the surface of the nanoparticles. The organic coating material may be bonded to the surface of the metal oxide nanoparticle, typically by non-covaient interactions. The organic coating material may not be volatile to a great extent and may require heating the metal oxide nanoparticle to a relatively high temperature to remove the organic material. The organic capping material may be a ligand. For the purposes of the present invention, a ligand may be defined as an organic material capable of compiexing, or forming typically non-covalent bonds, with an inorganic molecular or atomic entity.

The organic coating material when coated on the surface of the metal oxide nanoparticles may prevent the metal oxide nanoparticles from aggregating. Further, the metal oxide nanoparticles capped or coated with the organic coating material may be more easily dispersed in certain types of liquids, such as non-polar organic solvents including aliphatic and aromatic hydrocarbons.

Any suitable organic coating material may be used for the purposes of this invention. For example, the organic coating material may be, selected from the group consisting of, but not limited to, organic surfactants such as alkanoic acid, a saturated or unsaturated fatty acid, decanoic acid, oleic acid, an alkylamine, a fatty amine, oleylamine, an alkanol, adsorbed species derived from an alkanol, a fatty alcohol, oleyl alcohol, derivatives of an alcohol, polymeric coating materials such as, but not limited to, polyvinylpyrrolidone (PVP), and a combination thereof. Further organic coating materials known or obvious to a skilled person are also encompassed by the scope of the present invention. In particular, the organic coating material may be oleic acid. The organic coating material may comprise organic functional groups which may enhance or suppress the chemical interaction between the metal oxide nanoparticles and the substrate, or between the different kinds of adsorbed nanoparticles. According to a particular embodiment, the step 102 may comprise obtaining Ti0 ₂ nanoparticles capped with oleic acid.

The metal oxide nanoparticles comprising the organic coating material may be prepared using any suitable method. For example, the organic coating material may be added during the synthesis of the metal oxide nanoparticles.

Step 104 involves preparing a colloidal suspension 1 12 comprising the metal oxide nanoparticles obtained from the step 102. For the purposes of the present invention, a colloidal suspension may be defined as a mixture of metal oxide nanoparticles and a liquid wherein the predominant portion of the metal oxide nanoparticles are suspended in the liquid siich that the metal oxide nanoparticles do not readily precipitate out. In particular, the colloidal suspension 1 12 may comprise one more different types of metal oxide nanoparticles. The colloidal suspension 1 12 may remain stable for a period of time, which can range from one hour to about one year. While the metal oxide nanoparticles comprised in the colloidal suspension 1 12 can be caused to precipitate by centrifugation, under normal gravitational conditions, they do not rapidly collect at the bottom of the vessel containing the colloidal suspension 1 12. In this way, it may be easier to control the mole concentration of the metal oxide nanoparticles in the colloidal suspension 1 12. The colloidal suspension 1 12 may be transparent, or can be opalescent, or cloudy, depending at least in part on the size of the dispersed metal oxide nanoparticles.

When the step 102 comprises obtaining two or more different types of metal oxide nanoparticles, the colloidal suspension 1 12 prepared according to step 104 may comprise the two or more different types of metal oxide nanoparticles.

The liquid in which the plurality of the metal oxide nanoparticles is dispersed may be an organic solvent. The organic solvent may be a polar organic solvent, non-polar organic solvent or a mixture thereof. A mixture of organic solvents may also be used as the liquid for preparing the colloidal suspension 1 12. For example, the polar organic solvent may be acetone, an alcohol, ether, or a mixture of water-miscible solvent and water. Examples of the non-polar organic solvent include, but are not limited to, aliphatic or aromatic hydrocarbons and chlorocarbons. In particular, the liquid may be a non-polar organic solvent selected from hexane, toluene, benzene, or a mixture thereof. According to a particular embodiment, the step 104 comprises preparing a colloidal suspension 1 12 of Ti0 ₂ nanoparticles capped with oleic acid as obtained in the step 102 in toluene.

The colloidal suspension 1 12 may be low in viscosity. In particular, the colloidal suspension 1 12 used for the method 100 may have a viscosity which is < -| .n Pa.s- Since the colloidal suspension 2 does not require the addition of any polymer matrix, the viscosity of the colloidal suspension 1 12 remains relatively low. Despite the lack of addition of a polymer matrix to make the colloidal suspension 1 12 viscous, the method 100 provides formation of a metal oxide thin film which adheres strongly onto a substrate surface, making the method 100 an improvement over other coating and film- formation methods such as dip coating, spin coating, and electrospinning.

The colloidal suspension 1 12 may comprise the metal oxide nanoparticles in a suitable concentration. The concentration of the metal oxide nanoparticles in the colloidal suspension 1 12 may affect the quality of the thin film subsequently obtained from the colloidal suspension 1 12. For example, if the concentration of the metal oxide nanoparticles in the colloidal suspension 1 12 is too high, a thicker and more porous film will be obtained subsequently, in particular, the concentration of the metal oxide nanoparticles in the colloidal suspension 1 12 may be≥5 mg/mL. In particular, the concentration of the metal oxide nanoparticles in the colloidal suspension 1 12 may be 5-20 mg/mL, 10-18 mg/mL, 12-15 mg/mL. Even more in particular, the concentration of the metal oxide nanoparticles may be about 8-9 mg/mL.

The step 104 may further comprise adding chemical additives into the colloidal suspension 1 12. Any chemical additive suitable for the purposes of the present invention may be added into the colloidal suspension 112. For example, chemical additives that may affect the organization fashion of the metal oxide nanoparticles and/or generate additional chemical reactions during the thin film growth upon removal of the organic coating material may be added to the colloidal suspension 1 12. The chemical additives may include a ligand, a surfactant, a polymer, a liquid crystal forming material, a phospholipid, or a mixture thereof. For example, the ligand may be an organic ligand.

Once the colloidal suspension 1 12 of metal oxide nanoparticles is prepared in the step 104, a surface of a substrate may be contacted with the colloidal suspension 1 2 for a pre-determined period of time according to step 106. During the step 106, the metal oxide nanoparticles comprised in the colloidal suspension 1 12 may adsorb onto the surface of the substrate in contact with the colloidal suspension 1 12. In particular, the adsorption of the metal oxide nanoparticles may comprise self-assembly of the metal oxide nanoparticles on the surface of the substrate. Even more in particular, the metal oxide nanoparticles may form chemical bonds with the surface of the substrate.

The contacting may be any suitable form of contacting the surface of the substrate with the colloidal suspension 1 12. For example, the contacting may be by partially or completely dipping the substrate into the colloidal suspension 1 12. When the substrate is completely dipped into the colloidal suspension 1 12, both the top and bottom surfaces of the substrate may be contacted with the colloidal suspension 1 12. Accordingly, the metal oxide nanoparticles comprised in the colloidal suspension 1 12 may adsorb onto both the top and bottom surfaces of the substrate, resulting in the formation of a metal oxide thin film on both the top and the bottom surfaces of the substrate.

If it is desired for the metal oxide thin film to be formed on only one surface of a substrate, the substrate may be partially dipped into the colloidal suspension 1 12. Alternatively, other methods of contacting the surface of the substrate with the colloidal suspension 1 12 may be used. Examples of such methods are as shown in Figures 2(A) and 2(B). Referring to Figure 2(A), a solid frame 204 may be added onto the edges of a substrate 202. According to one embodiment, the solid frame 204 may be pasted onto the edges of the substrate 202, thereby defining an area 203 on the surface of the substrate 202. The colloidal suspension 1 12 may be deposited or poured into the area 203 so that a metal oxide thin film forms only at area 203 of the substrate 202. The actual height of the solid frame 204 may be varied, depending on the amount of colloidal suspension 1 12 to be deposited within the area 203. The solid frame 204 may be made of any suitable material. In particular, the solid frame 204 may be made of a non-reactive material. By this method, the precise area on which the metal oxide thin film is to be formed may be provided. In this way, it may also be possible to predetermine and quantify the amount of metal oxide nanoparticles to be loaded onto the substrate surface 203.

An alternative method may be as shown in Figure 2(B). Figure 2(B) may be scaled up to large scale production of metal oxide thin films. The method shown in Figure 2(B) may be applied as a batch process or as a continuous process. In particular, a substrate 202a may be hung over a pool of colloidal suspension 1 12 such that the bottom surface 203a of the substrate 202a is in contact with the colloidal suspension 1 2. In this way, the adsorption of the metal oxide nanoparticles comprised in the colloidal suspension 1 12 may take place in the direction as shown by the arrow in Figure 2(B). Thus, it would be understood by a person skilled in the art that in the method 100, adsorption of the metal oxide nanoparticles onto the surface of the substrate may be bi-directional, i.e. the adsorption of the metal oxide nanoparticles may take place in line with or against the force of gravity.

When the method as shown in Figure 2(B) is applied as a continuous process, a plurality of substrates 202a may be arranged in a row. When the process is started, surface 203a of each substrate 202a in the row contacts a pool of colloidal suspension comprising the metal oxide nanoparticles one at a time. The adsorption time of the metal oxide nanoparticles onto the surface 203a of the substrate 202a may depend on the shifting speed of the substrates 202a across the surface of the pool of colloidal suspension and the length of the pool of the colloidal suspension.

The substrate which contacts the colloidal suspension 1 12 in step 106 may be any suitable substrate for the purposes of the present invention. The substrate may be of an electrically insulating, conductive or semi-conductive material. For example, the substrate may be a normal or doped glass, plastic, silicon substrate. For example, the glass substrate may be soda lime glass or Borofloat ^® 33 glass. Examples of plastic substrates include substrates containing polypropylene, polycarbonate, polyimide, polyethersulfone, polyethylene terephthalate or a mixture thereof. The silicon substrate may comprise silicon, silicon nitride or silica. According to a particular embodiment, the substrate may be a glass substrate. In particular, the glass substrate may be an optically transparent glass. The surface of the substrate may be cleaned before contacting the colloidal suspension 1 12. The surface of the substrate may be cleaned by any suitable method: In particular, the surface of the substrate may be cleaned to remove contaminants from the surface. The surface of the substrate may also be cleaned to remove any hydrophobic material which may be on the surface of the substrate. In particular, the surface of the substrate may be cleaned to remove oil from the surface. For example, the surface of the substrate may be cleaned using Piranha solution. The method 100 may therefore comprise a further step of cleaning the surface of the substrate prior to the step 104 of contacting.

The substrate may comprise a flat, curved, or irregular surface. For example, the substrate may be a patterned or unpatterned substrate. In the case of a patterned substrate, the surface of the substrate may comprise a pattern. Accordingly, the method 100 may further comprise a step of patterning the surface of the substrate prior to the step 106 of contacting the surface of the substrate with the colloidal suspension 1 12. Any suitable method may be performed to pattern the surface of the substrate. For example, the surface of the substrate may be patterned by etching.

During the step 106, the contacting of the surface of the substrate with the colloidal suspension 1 12 takes place for a pre-determined period of time. The pre-determined period of time may be any suitable amount of time for the purposes of the present invention. For example, the pre-determined period of time may be≥15 minutes. In particular, the pre-determined period of time may be 15 minutes - 8 hours, 30 minutes - 6 hours, 45 minutes - 5 hours, 1-4 hours, 1.5-3 hours, 2-2.5 hours. According to a particular embodiment, in the step 106, the contacting of the surface of the substrate with the colloidal suspension 1 12 is for a pre-determined period of about 2 hours.

After the step 106 of contacting the surface of the substrate with the colloidal suspension 1 12, the surface of the substrate is heated according to the step 108 to form a metal oxide thin film 1 14 on the surface of the substrate. During the step 108, the organic coating material disposed on the surface of the metal oxide nanoparticles adsorbed onto the surface of the substrate may be removed from the surface of the metal oxide nanoparticles. In particular, the organic coating material may be removed partially or completely from the surface of the metal oxide nanoparticles adsorbed onto the surface of the substrate. The heating in the step 108 also promotes the crystallite growth and/or interconnectivity of the metal oxide nanoparticles adsorbed onto the surface of the substrate, thereby generating better adhesion and adsorption of the metal oxide nanoparticles to the surface of the substrate and creating better adhesion of the metal oxide thin film 1 14 to the surface of the substrate.

According to a particular embodiment, the organic coating material may be partially removed from the surface of the metal oxide nanoparticles adsorbed onto the surface of the substrate during the step 108 of heating the substrate. In this way, the metal oxide thin film 1 14 formed may be an inorganic-organic hybrid film. For example, for inorganic-organic hybrid films, the organic coating material may comprise special functionalities which enable the organic coating material to be retained during the thin film fabrication.

If the colloidal suspension 1 12 comprises two or more different types of metal oxide nanoparticles, during the step 108 of heating the surface of the substrate, the two or more different types of metal oxide nanoparticles may undergo solid-solid or gas-solid chemical reactions or interdiffusion between the two or more different types of metal oxide nanoparticles may take place, resulting in the formation of a metal oxide thin film 1 14 which is a complex of two or more different types of metal oxide nanoparticles.

The heating may be under conditions suitable for the purposes of the present invention. For example, the heating may be carried out in laboratory air or in a chemically controlled ambience. For example, the heating may be carried out in the presence of non-reactive gases such as helium, argon, or nitrogen, or in the presence of reactive gases such as water vapour, oxygen, carbon monoxide, or carbon dioxide.

The temperature at which the heating of the step 108 is carried out may be any suitable temperature. For example, the heating may be at a temperature of ≥ 150°C. In particular, the heating may comprise heating the substrate at a temperature of 150- 700°C, 200-600°C, 250-550°C, 300-450°C, 350-425°C, 375-400°C. According to a particular embodiment, the heating may be carried out at a temperature of about 400°C.

The heating in the step 108 may be carried out for a suitable period of time. For example, the heating may be carried out for at least 5 minutes. In particular, the heating may be carried out for 5 minutes - 8 hours, 10 minutes - 7 hours, 30 minutes - 6 hours, 1-5 hours, 2-4 hours, 2.5-3.5 hours, 2.75-3 hours. According to a particular embodiment, the heating may be carried out for about 2 hours.

The method 100 may further comprise an optional step 110 of repeating steps 102 to 108 to form a further metal oxide thin film 116. In particular, the steps 106 and 108 may be repeated to form a further metal oxide thin film 116 on the surface of the substrate having the preceding metal oxide thin fiim 1 14, thereby increasing the thickness of the metal oxide thin film 114. In particular, the method 100 may comprise repeating the steps 106 and 108 such that the metal oxide nanopartides comprised in the further metal oxide thin film 116 adsorb onto the surface of the preceding metal oxide thin film 1 14.

As a result of the step 0 of repeating the steps 102-108, the method 100 allows control over the surface texture and the thickness of the metal oxide thin films prepared from the method 100.

When the steps 102 to 108 are repeated, the metal oxide nanopartides comprised in the further metal oxide thin film 1 16 may be the same or different from the metal oxide nanopartides comprised in the preceding metal oxide thin film 114. When it is desired for the metal oxide nanopartides comprised in the further metal oxide thin film 116 to be different from the metal oxide nanopartides comprised in the preceding metal oxide thin film 114, different metal oxide nanopartides may be selected during the step 102. Accordingly, a colloidal suspension of a different type of metal oxide nanoparticie will be prepared in the step 104. In this way, method 100 may allow for the formation of layered nanocomposite metal oxide thin films in a layer-by-layer sequence. In this way, the surface texture, thickness, and chemical composition of the further metal oxide thin film 116 may be controlled. In particular, the method 100 may therefore allow the incorporation of metal and alloy nanopartides such as gold and silver nanopartides into or onto the further metal oxide thin film 116.

The metal oxide thin film 1 14 and the further metal oxide thin film 116 formed on the surface of the substrate may have any suitable thickness. In particular, the thickness of the metal oxide thin film 1 14 and the further metal oxide thin film 116 may depend on a number of factors such as the pre-determined period of time of the contacting in the step 106, the concentration of the metal oxide nanopartides comprised in the colloidal suspension 2 and the temperature at which the heating in the step 108 is carried out. The thickness of the further metal oxide thin film 1 16 may further depend on the number of times the steps 106 and 108 are repeated.

For example, the thickness of the metal oxide thin film 1 14 and the further metal oxide thin film 1 16 may each be about 20-500 nm. In particular, the thickness of the metal oxide thin film 1 14 and the further metal oxide thin film 1 16 may each be about 30-450 nm, 50-400 nm, 60-350 nm, 70-300 nm, 80-250 nm, 90-200 nm, 100-150 nm. Even more in particular, the thickness of the metal oxide thin film 1 15 and the further metal oxide thin film 16 may be < 100 nm .

The metal oxide thin film 1 14 and the further metal oxide thin film 1 16 prepared from the method 100 may demonstrate desirable and useful properties. For example, the metal oxide thin film 1 14 and the further metal oxide thin film 1 16 exhibit good transmittance, photocatalytic activity, superhydrophilicity as will be shown in the Examples beiow. The metal oxide thin film 114 and the further metal oxide thin fiim 1 16 may be a conformal thin film and may be spread uniformly across the surface of the substrate.

In view of the simplicity of the method 100, the method 100 is a scalable method which may be used on an industrial scale.

According to a particular embodiment, the method 100 may be schematically shown in Figure 3. Figure 3 shows a method 300 in which there is provided a colloidal suspension 302 comprising metal oxide nanoparticles 302a. The metal oxide nanoparticles 302a are coated with an organic coating material 302b. In particular, the metal oxide nanoparticles 302a may be Ti0 ₂ nanoparticles and the organic coating material may be a surfactant such as oleic acid. The colloidal suspension 302 is then contacted with a surface of a substrate 306. The substrate 306 may be a glass substrate. When the colloidal suspension 302 is contacted with the surface of the substrate 306, the metal oxide nanoparticles self-assemble and adsorb onto the surface of the substrate. The substrate 306 is then subjected to a heating step during which the organic coating material 302b coated on the metal oxide nanoparticles 302a is removed and the metal oxide nanoparticles 302b form a thin film 308 on the surface of the substrate. The steps of contacting the colloidal suspension 302 to the surface of the substrate 306 and the heating of the substrate 306 are repeated again (i.e. second cycle) which result in the formation of a further thin film 308 on the thin film 308 formed during the first cycle, thus forming a thicker film 308. The steps of contacting and heating may be repeated until the desired thickness of the thin film 308 is achieved.

According to another embodiment, the method 100 may be schematically shown in Figure 4. In Figure 4, there is provided a method 400 of preparing a metal oxide thin film on a patterned substrate. In particular, there is provided a substrate 402 which may be a glass substrate. The method 400 may comprise a step of etching the substrate 402 so that the surface 402a of the substrate 402 is a patterned surface. A colloidal suspension 404 of metal oxide nanoparticles 404a comprising a coating of organic coating material 404b on the surface of the metal oxide nanoparticles 404a is then contacted with the patterned surface 402a to allow the metal oxide nanoparticles 404a comprised in the colloidal suspension 404 to self-assemble and adsorb onto the patterned surface 402a. The substrate 402 is then heated as previously described in relation to method 300 to form a metal oxide thin film (not shown) on the patterned surface 402a. By patterning a substrate surface, the antireflection characteristics and the light transmission of the metal oxide thin film formed on the surface may be further enhanced.

According to another aspect of the present invention, there is provided a metal oxide thin film obtained from or obtainable by the method described above. The metal oxide thin film obtained may have desirable properties. For example, the metal oxide thin film may be used in photovoltaic devices, window glasses, solar cells such as dye- sensitized solar cells and solid-state semiconductor metal oxide solar cells, catalytic thin films for use in applications such as wastewater treatments, bactericidal applications (i.e. advanced oxidation processes), and for general photocatalytic chemical reactions (i.e. synthesis of fine chemicals), and for indoor air cleaning (i.e. catalytic decomposition of volatile organic compounds). In particular, the metal oxide thin film may be as described in relation to the metal oxide thin film 1 14 and the further metal oxide thin film 1 6.

In particular, the metal oxide thin film may be a Ti0 ₂ thin film, which may be employed in photocatalytic degradation of harmful organic contaminants such as dyes and pesticides and in the photocatalytic inactivation of microorganisms such as £. coli. The Ti0 ₂ thin films may also be applied to photovoltaic devices, solar cells and window glasses. The present invention further provides an article of manufacture comprising the metal oxide thin film prepared from the method according to the method as described above. The article of manufacture may be any suitable article of manufacture. For example, the article of manufacture may be, but not limited to, a photovoltaic device, a window, or a solar cell.

Having now generally described the invention, the same will be more readily understood through reference to the following examples which are provided by way of illustration, and are not intended to be limiting.

EXAMPLES

Example 1

Preparation of Ti0 ₂ nanoparticles

Monodispersed Ti0 ₂ nanoparticles were synthesized based on a modified experimental procedure. Briefly, 0.20 g of titanium (IV) n-propoxide (98%, Alfa Aesar) was added into 10.0 mL of toluene (AR, Aldrich). Then 6.0 mL of oleic acid (OA, CH ₃ (CH ₂ ) ₇ CH=CH(CH ₂ ) ₇ COOH, 90%, Aldrich) was added into the above mixture. A second solution containing 0.10 mL of tert-butylamine (98%, Aldrich) and 10.0 mL of deionized water was also prepared. The above two solutions were mixed and magnetically stirred for 15 minutes and then transferred into a Teflon-lined stainless steel autoclave. The solvothermal reaction was performed in an electric oven which had already been heated to 180°C for a period of 7-12 hours. After the treatment, the product mixture was separated into oil and water phases. The oil phase was taken out and added with an equivalent volume of ethanol in order to precipitate Ti0 ₂ nanoparticles. The precipitated Ti0 ₂ nanoparticles were then centrifuged and the solvent was decanted in order to obtain OA-capped Ti0 ₂ nanoparticles.

Preparation of colloidal suspension

The OA-capped Ti0 ₂ nanoparticles were then were redispersed in 16.0 mL of toluene as a colloidal Ti0 ₂ suspension. The concentration of Ti0 ₂ nanoparticles in the thus prepared colloidal suspension was about 8.9 mg/mL. The colloidal Ti0 ₂ suspensions were held in glass bottles. Therefore, Ti0 ₂ nanoparticies also adsorbed on the wall of the glass bottles. Alternatively, the colloidal suspension may be held in plastic bottles.

Fabrication of Ti0 ₂ film on plane glass substrate

Soda lime glass substrates (1.5 cm x 2.5 cm x 1.0-1.2 mm; Sail brand, China) were cleaned in a Piranha solution. The Piranha solution was prepared by mixing sulphuric acid (H ₂ S0 ₄ , 95-97%, Merck) and hydrogen peroxide (H ₂ 0 ₂ , 30%, Merck) in a volume ratio of 3:1. After the cleaning treatment, the glass substrates were dipped in the Ti0 ₂ colloidal suspension for 2 hours to adsorb the Ti0 ₂ nanoparticies. Samples were taken out of the suspension and washed with toluene, followed by 2 hours of heat treatment in an electric furnace at 400°C in laboratory air. This combined soaking for 2 hours and heating for 2 hours is called one process cycle for the purposes of this example. Further Ti0 ₂ films were fabricated by repeating this process cycle, that is, by soaking the glass substrate every time in a freshly prepared Ti0 ₂ colloidal suspension followed by annealing in the laboratory air. Alternatively, the same Ti0 ₂ colloidal suspension as previously may be used when repeating the process cycle.

Fabrication of Ti0 ₂ film on patterned glass substrate

Patterned glass substrates (Borofloat glass 33; Schott AG) which were etched with inductively coupled plasma (ICP) were also used. Similar to the method described above in relation to plane glass substrate, the patterned glass substrate was dipped in the Ti0 ₂ colloidal suspension for 2 hours followed by heat treatment for 2 hours to fabricate Ti0 ₂ film on the patterned glass substrate.

Characterisation of the Ti0 ₂ film

The size and crystalline structure of colloidal Ti0 ₂ nanoparticies were investigated by transmission electron microscopy (TEM, JEM-2010, accelerating voltage: 200 kV). Structural and morphological profiles of studied samples were carried out with field emission scanning electron microscopy (FESEM, JSM-6700F, accelerating voltage: 5 kV, current: 10 μΑ) and atomic force microscopy (AFM, Dl Nanoscope Multimode). When detected by FESEM equipment, all the samples are coated with Pt at 10 mA for 40 seconds. For the titania films on glass substrates, the surface and bulk chemical compositions of Ti0 ₂ films were analyzed with energy-dispersive X-ray spectroscopy (EDX/FESEM, JEM-6700F, accelerating voltage: 15 kV, working distance: 15 mm) and X-ray photoelectron spectroscopy (XPS, AXIS-HSi, Kratos Analytical) respectively. The working pressure of XPS detection is 5x1 CT ⁹ torr and the X-ray source is monochromatic Al K _a (hv = 1486.71 eV, 5 mA, 15 kV). UV-vis absorption spectra were recorded at different reaction intervals to monitor photocataiytic reactions (as described below) using an UV/vis spectrophotometer (Shimadzu UV-2450). The hydrophilic properties of the grown Ti0 ₂ films were examined from a static contact angle (CA) measurement (VCA optima surface analysis system) at room temperature using a water droplet with a volume of about 1 μΙ_.

Photocataiytic measurements

The photocataiytic investigation for the glass supported Ti0 ₂ films was carried out with a reference reaction of decomposition of methyl orange (Ci H ₁ N ₃ Na0 ₃ S, Merck) in laboratory air at room temperature (25°C) and under ambient pressure. The photocataiytic activity of Ti0 ₂ films was evaluated in two different ways. Firstly, a cylindrical wide-mouth glass bottle (capacity of about 30 mL) was used as the reactor vessel. The glass supported Ti0 ₂ film (area at 1.5 cm ^χ 2.5 cm) was dipped in 4.0 mL of aqueous methyl orange solution with a concentration of 5.0 mg/L and then exposed to ultraviolet (UV) light using a high-pressure mercury lamp (125 W, Philips). A cut-off filter (wavelength of 385 nm) was placed in front of the lamp which removed all emission lines of wavelength less than 385 nm.

For comparison, a control experiment was also conducted without inserting any Ti0 ₂ film catalysts to detect the natural photodecomposition of methyl orange with the same ultraviolet irradiation. UV-vis absorption spectra of methyl orange were recorded at constant intervals to monitor the progress of photocataiytic reactions, as described above.

A second way of studying the photocataiytic performance of the prepared Ti0 ₂ films was designed to simulate the real working condition of Ti0 ₂ films on glass when used as a commercial self-cleaning material under dry conditions (i.e., outside the solution). Namely, the Ti0 ₂ film counterparts were firstly immersed in 8.0 mL of methyl orange solution at a high concentration of 250 mg/L for 2 hours. After this organic deposition, the samples with the dye molecules on their surfaces were taken out and dried in an electric oven. They were then exposed vertically to the same set of the UV light source and cut-off filter. The gradual degradation of this surface organic deposit was also sequentially recorded.

Scratch test

Scratch tests using diamond indenter were performed on the prepared Ti0 ₂ films to analyze the compactness and adherence of the films to the glass substrates. The radius of diamond ball used to make scratch was 1 jim, -which resulted in a pressure of 1.06 GPa at an applied load of 100 mN. Although this kind of pressure is not possible in real world outdoor applications, the film strength was estimated under extreme conditions by using this test. The load was varied progressively from 0.03 to 100 mN over a distance of 1 mm. Frictional force and scratch depths were measured during the measurement and compared with uncoated glass substrates. The surface topography of the films after scratch tests was examined with atomic force microscopy (SPM, Veeco Dimension 3100).

Results obtained

(a) Ti0 ₂ thin films on plane glass substrates

Figure 5 gives two TE images of the Ti0 ₂ colloidal particles which were used as starting building blocks in the self-assembled film preparation as described above. The oxide nanoparticles showed a narrow size distribution of 5-9 nm. A higher-resolution TEM image of Figure 5(B) indicates Ti0 ₂ nanocrystals with clearly resolved lattice spacing of 0.352 nm, corresponding to the interspacing of (101 ) planes of anatase Ti0 ₂ . The above analyses and results affirm that the Ti0 ₂ nanoparticles were single- crystalline in the anatase phase.

Figure 6 presents the representative surface morphology of as-synthesized Ti0 ₂ films, showing a uniform and conformai film spreading on the glass substrate. After the first process cycle, Ti0 ₂ nanoparticles self-assembled evenly and formed a continuous dense film on the glass surface, as shown in the panoramic view of Figure 6(A). Through analyzing the FESEM image of Figure 6(B), it can be seen that the surface Ti0 ₂ particles have an average size of 20 nm after heating, which were grown bigger compared to their pristine Ti0 ₂ counterpart (i.e., the initial colloidal particles in Figure 5). At the same time, partial fusion of Ti0 ₂ nanoparticles also took place, because nanoparticles aggregated into bigger crystal grains and interstitial voids were formed after the thermal treatment. Through repeated process cycles in Ti0 ₂ suspensions, thicker Ti0 ₂ films were deposited on the glass surface in a stepwise fashion, as demonstrated in Figures 6(C) to (H).

Using AFM technique to characterize these films, it was found that the surface of the Ti0 ₂ films were composed of tiny grains and their average roughness was about 1.7 nm for S1 (films formed from one process cycle) and S2 (films formed from two process cycles) samples. With increase in process cycles, the films became somewhat rougher but still without cracking and the average roughness reached around 2.0 nm. Because the phase-transformation from anatase to rutile takes place at a much higher temperature, the formed Ti0 ₂ films which were annealed at 400°C remained in the anatase phase.

At the same time, a pronounced peak at 285.6 eV in C1 s photoelectron spectrum of these Ti0 ₂ films before heat-treatment is attributed to defect-containing sp ² hybridized carbon confirming the existence of OA capping molecules before calcination. During the formation of Ti0 ₂ nanoparticles, tert-butlyamine served as the base to neutralize the OA and thus generated the carboxylic group which can bind on surface Ti0 ₂ nuclei. After heating at 400°C for 2 hours, the peak corresponding to the sp ² hybridized carbon disappeared completely, revealing the complete removal/decomposition of the OA molecules. The calcination/heating step therefore brings about better engagement between the Ti0 ₂ films and glass substrates, enhancing the film adhesion and mechanical performance. Consistently, the water contact angle of Ti0 ₂ film prior to the heat-treatment was found as high as about. 70°, implicating the presence of hydrophobic OA molecules in the films and the pivotal role of thermal treatment to form the Ti0 ₂ coating.

Table 1 summarizes the EDX results and film thickness (determined together with FESEM technique) for the samples shown in Figure 6. As expected, the progressive increase in Ti/Si atomic ratio and film thickness in these Ti0 ₂ /Si0 ₂ samples is in good agreement with the repeated soaking-heating cycles. From the data of Table 1 and the size of pristine Ti0 ₂ nanoparticles (Figure 5(B)), after the each process cycle, the prepared film consisted of about 10-15 layers of adsorbed Ti0 ₂ nanoparticles.

Table 1 : Summary of EDX results and thickness of Ti0 ₂ thin films shown in Figure 6.

In addition, the Ti0 ₂ films prepared with one to four process cycles were all absolutely clear and transparent and had a light transmittance nearly the same as their pristine bare glass substrates, as shown in Figure 7(A) with the normal incidence of light. In fact, most samples showed overlapping transmission curves, which reflect their suitability in solar cell applications. Furthermore, the freshly prepared samples exhibit excellent superhydrophilic nature with a water contact angle of about 0° (the photograph in ^' the inset of Figure 7(A)) and still retained at about 15° even though the samples were kept in the dark for a week.

In order to investigate the applicability of the films for self-cleaning solar cells, the transmitting property of these films were also measured with all-directional incident light. Figure 7(B) gives the transmittance curves versus the incident light at a wavelength of 400 nm. By comparing with uncoated glass, it was found that the self- assembled Ti0 ₂ films indeed showed a good all-dimensional transmittance close to their original glass substrate.

To further demonstrate the versatility of this adsorptive self-assembly route, a series of related experiments were conducted. For example, the soda lime glass substrates were replaced with borofloat glass (borofloat glass 33; Schott AG) to see the reproducibility of the film formation. As expected, very uniform thin films of Ti0 ₂ on the borofloat glass were obtained (see Figure 8) using the same synthetic procedures adopted to obtain the results as shown in Figure 6. However, aggregated crystal particles consisting of these films had a larger size distribution in the range of 9-18 nm compared to those formed on the soda lime glass in Figure 6. Apart from the crystallite size variation, the films were also highly transparent. Similarly, the water contact angle on the borofioat glass decreased from 60° to about 1 ° after the formation of the Ti0 ₂ films, demonstrating that the as-formed Ti0 ₂ films were also superhydrophilic.

(b) Ti0 ₂ thin films on patterned glass substrates

Because infantile Ti0 ₂ particles were very small in size, they likely could adsorb and thus spread over to every point of a substrate. In order to verify this, the same film forming process as described for plane glass substrate was extended to patterned nanostructured glass. In particular, a borofioat glass substrate was etched with inductively coupled plasma (ICP) to obtain two kinds of nanostructures with piilar- and holed patterns. As is expected, FESEM study of Figure 6 revealed resultant topographical surface nanostructures, testifying the formations of Ti0 ₂ films both on the pillar-patterned glass and hole-patterned glass. The transmittance reported in Figure 9(e) decreased a little in contrast with the bare borofioat glass, and at the same time, the reducing water contact angle from about 37° to below 10° (Figure 9(f)) further confirmed the growth of Ti0 ₂ film on holed glass. The above results suggest that Ti0 ₂ nanoparticles self-assemble uniformly on different kinds of patterned glass surfaces.

(c) Scale up of method for preparing Ti0 ₂ thin film

The process for preparing Ti0 ₂ thin films on larger substrates was carried out using the process parameters acquired in the above experiment. A larger piece of borofioat glass plate sized at 12.5 cm ^χ 12.5 cm ^χ 2.0 mm was used. The glass plate was dipped in 350 ml_ of Ti0 ₂ suspension for a soaking (adsorption) period of 2 hours, followed by the same annealing treatment as described above at 400°C for 2 hours. It was found that the surface morphology of the formed Ti0 ₂ film was very flat with a high light transmittance and a small water contact angle of below 10° (see Figure 10). This illustrates the scalability of the method for preparing the Ti0 ₂ thin films.

(d) Surface analysis and adsorptive self-assembly

In order to get more information about the surface and chemical composition of the constituent elements in the prepared films, a comparative XPS analysis was conducted for samples fabricated both on the soda lime glass and borofioat glass. In Figure 1 1 , the binding energies (BEs) of XPS spectra of all studied elements are referred to C 1 s photoelectron peak originating from adventitious carbon (its BE was set at 284.6 eV); other two weak C 1 s peaks located at 286.0-286.2 and 288.2-288.3 eV are assigned to hydroxyl carbon (HO-C) and carboxylate/carbonate anions, respectively. All O 1 s spectra can be deconvoluted into four peaks. The two' identical peaks at 532.7 eV are due to the bridging oxygen in Si-O-Si, namely the main component of glass. The other peaks at 530.9, 531.4-531 .8, and 533.2 eV are attributed to adsorbed hydroxyl (-OH), and Si-O-Ti/M (M denoting other metal elements contained in glass) and water molecules, respectively, indicating the presence of a hydrophilic surface. The peaks at 529.5-529.8 eV are commonly assigned to the lattice oxygen of anatase phase. In this agreement, the doublet peaks of Ti 2p photoelectrons at 458.1-458.5 eV and 463.9-464.3 eV further confirmed the presence of anatase phase before and after calcination on the both glass substrates. As for the Si 2p photoelectrons, the two peaks at 103.1-103.2 eV and 102.1-102.2 eV are ascribed to Si elements of the above- mentioned Si-O-Si and Si-O-Ti/M linkages.

Generally, the layer-by-layer deposition depends on the electrostatic interaction between the nanoparticles from suspensions and substrates with opposite charges. In order to have some insights on this, soda lime glass without cleaning with Piranha process was also used for Ti0 ₂ deposition. In such a case, formation of isolated Ti0 ₂ islands on the glass substrates was found, suggesting that Ti0 ₂ can only be added to certain locations of the glass substrate, and attractive electrostatic force of opposite charges is unlikely to be a major cause for the adsorption because the OA-capped Ti0 ₂ nanoparticles are not charged.

The effect of adsorption time on the formation of Ti0 ₂ thin films was also examined. The initial yellowish transparent Ti0 ₂ colloidal suspension turned white turbid when the soaking duration of glass was beyond 12 days and white precipitates couid then be seen at the bottom of the bottle after 18 days. This observation indicates that Ti0 ₂ nanoparticles became less protected by the oleic acid (OA) capping after the prolonged standing in toluene. Consistent with this postulation (and unlike those shown in Figure 5), the Ti0 ₂ nanoparticles after soaking for 18 days were are indeed more aggregative with a reduction in inter-crystal space, and direct contact among the nanoparticles could be observed. Because no aggregation among the Ti0 ₂ colloidal nanoparticles was observed in the normal film formation process as described above (i.e., in which the adsorption time was much shorter than 12 days), the adsorption due to van der Waals interaction among the surface OA molecules can be ruled out. On the other hand, it was observed that the adsorption of the as-prepared Ti0 ₂ nanoparticles took place readily on the cleaned glass substrate even within 10 seconds, which shows that there must be a partial disengagement of OA-capping from the Ti0 ₂ nanoparticles in order to generate aggregative thin films of the nanoparticles, because unprotected Ti0 ₂ nanoparticles would be very reactive. Accordingly, it can be concluded that during the initial adsorption, the OA surfactant is partially disengaged and surface hydroxyl groups of Ti0 ₂ nanoparticles react with those of silica substrate, forming Ti-O-Si linkages (i.e., Ti-OH + HO-Si - Ti-O-Si + H ₂ 0). Afterwards, the assembly mainly takes place among the Ti0 ₂ nanoparticles. In particular, the OA surfactant is partially removed and surface hydroxyl groups of incoming Ti0 ₂ nanoparticles react with the hydroxyl groups of the previously deposited Ti0 ₂ nanoparticles, resulting in formation of Ti-O-Ti bonding (Ti-OH + HO-Ti → Ti-O-Ti + H ₂ 0) or direct contact among the Ti0 ₂ nanoparticles. This adsorptive self-assembly would go on and the thickness of the Ti0 ₂ nanoparticle film depends on the length of process time or the number of cycles carried out.

Therefore, the surface hydroxyl groups of silica substrate play a crucial role to initiate the Ti0 ₂ adsorption. Accordingly, without cleaning of the surface of the substrate, surface active sites for adsorption may be limited due to a layer of pristine grease/stains that largely depend on opposite charges provided by the nanoparticles and poly-electrolytes used in the film fabrications.

(d) Self-cleaning application

After being cleaned with the Piranha process, the soda lime glass substrates exhibit high hydrophilicity due to the presence of abundant surface -OH groups which is consistent with the XPS results in Figure 1 1. In this regard, the surface cleaning would ensure an even deposition of the Ti0 ₂ nanoparticles onto the Si0 ₂ substrate as well as the subsequent Ti0 ₂ to Ti0 ₂ deposition.

The photocatalytic behaviour of the as-prepared films was investigated using the photocatalytic degradation of methyl orange as a model reaction, during which the characteristic absorption of this dye molecule at about 485 nm was used to monitor its degradation process. Furthermore, to verify the effectiveness of the present adsorptive assembling route, several batches of Ti0 ₂ film samples with a deposition time varying from 4 to 8 h were also prepared.

Figure 12(A) displays a series of absorption spectra of methyl orange aqueous solution (initial concentration of 5.0 mg/L) measured for this degradation reaction in the absence of a catalyst or in the presence of various Ti0 ₂ films. All the sample films were exposed to an UV irradiation (UV lamp, 125 W, as described above) for 6 hours, and their corresponding C _t /C ₀ data after the exposure to the UV light was plotted in Figure 12(B). It was found that films of samples S2, S3 and S4 all had a higher photocatalytic activity compared to their counterpart samples (i.e., samples (g) to (i) of Figure 12(B)) prepared with the same total soaking times of 4, 6 and 8 hours, respectively. When the soaking time was extended to 12 days, the resultant Ti0 ₂ film (sample (j) of Figure 12(B)) still showed a lower activity than S2, which further justified the effectiveness of this multiple-coating route to prepare Ti0 ₂ films. It can be noticed that the Ti0 ₂ film prepared with two process cycles (S2) has a superior photocatalytic performance to other samples with one, three or four process cycles as shown in Figure 12. When only one process cycle of Ti0 ₂ was deposited on the glass substrate (i.e., sample S1 ), the film was too thin to have good photocatalytic activity. After two process cycles, the quantity of Ti0 ₂ film (sample S2), as well as the total surface area, increased as a result of the rise of existing inter-particle space. With more deposited layers, while the Ti0 ₂ phase grew in quantity, the photocatalytic efficiency declined, which is attributed to a significant crystallite coarsening among the Ti0 ₂ nanoparticles when the overall thickness of the films increases.

To understand the advantage of the prepared films in self-cleaning application, the soda lime glass supported Ti0 ₂ films were immersed in 8.0 ml_ of 250 mg/L of methyl orange solution for 2 hours to disseminate methyl orange on the surface and compared its efficiency with commercial Ti0 ₂ films (spray coated samples, proprietary source information; samples supplied by Haruna, Singapore) under the identical reaction conditions. Time-dependent (C _t /C ₀ ) curves of residual methyl orange were recorded (see Figure 13) and once again it was found that the sample prepared with two process cycles had the best photocatalytic activity compared to the other samples and was much more efficient than the commercial Ti0 ₂ films fabricated via three spray process cycles of commercial titania powder, which may be attributable to much smaller grain size and special nanostructure of the Ti0 ₂ films prepared by the method according to the present invention, in comparison to those in the commercial products. A FESEM image of the Ti0 ₂ film provided by Haruna, Singapore is shown in Figure 14.

For the practical application of the Ti0 ₂ films prepared according to the method of the present invention, the mechanical durability is an important factor. In Figure 15, the normalized frictional force curves measured during scratch tests performed on several Ti0 ₂ film samples are reported. Understandably, the oscillations of the curves manifest the formation of cracks in the film and a resultant decrease in the friction measured by the diamond ball. As the diamond ball penetrates deeper, the friction force increases again until the formation of a new crack. The uncoated glass shows formation of cracks at a force of 25-28 mN, which is delayed to higher values with each process cycle of Ti0 ₂ film growth. Samples with three process cycles shows formation of large cracks shifted to 50 mN. In a practical world, a force of 5 mN is equivalent to normal hand cleaning of surfaces, and hence the scratch test indicates a huge increase in the life span of the module packaging. Nevertheless, thicker films show easy peeling-off and breakage of films.

By comparing with bare glass, the prepared Ti0 ₂ fiims according to the method of the present invention show good anti-scratching property. Therefore, the Ti0 ₂ films can serve as cover glasses for self-cleaning solar panels as well as other types of self- cleaning devices and applications.

Remarks

In summary, the method of the present invention relates to an adsorptive self-assembly approach for solution-based preparation of anatase Ti0 ₂ thin films. Using the pre- synthesized anatase nanoparticles as starting building blocks, high quality Ti0 ₂ thin films can be assembled uniformly onto various plane glass substrates, including normal soda lime glass and solar-grade borofloat glass. In addition to the variation in substrate composition, the same Ti0 ₂ thin films can also be prepared onto nanostructure- patterned glass substrates. Owing to the simplicity and architectural flexibility of the method, the present synthetic strategy can be faciiely employed in large-scaled Ti0 ₂ thin film preparations. The results obtained above show that the fiims formed are smooth and conformal with no cracks, and the results from the scratch tests show a good interracial adhesion between the Ti0 ₂ films and glass substrates which is further enhanced by the thermal post-treatment at 400°C. Through repeated adsorption- heating process cycles, the thickness and crystallite morphology of the thin films can be adjusted. The Ti0 ₂ films have excellent super-hydrophilic surface with a water contact angle below 5° and maintain at a low contact angle of about 15° even after being stored in a dark place for a week. Furthermore, the photocatalytic performance of the Ti0 ₂ thin films fabricated by the present approach is found to be better than commercially available spray-coated Ti0 ₂ films. Due to its high transmittance, the Ti0 ₂ thin films prepared from the method of the present invention can indeed act as omnidirectional self-cleaning coating on solar panels with very little loss of power efficiency.

Example 2

Materials used

The chemicals and solvents used for this example are as follows: Titanium (IV) n- propoxide (98%, Alfa Aesar), Toluene (AR, Aldrich), Oleic acid (OA, CH ₃ (CH ₂ ) ₇ CH=CH(CH ₂ ) ₇ COOH, 90%, Aldrich), Tert-butylamine (98%, Aldrich), Sulphuric acid (H ₂ S0 ₄ , 95-97%, Merck), Hydrogen peroxide (H ₂ 0 ₂ , 30%, Merck), AgN0 ₃ (Merck, ISO, Reag. Ph Eur), polyvinylpyrrolidone (PVP; Fluka, K30, average Mw 40,000), Tween-20 (Acros, USA), Tween-40 (Acros, USA), ethanol (Merck, Analytical Reagent grade).

Preparation of Ti0 ₂ film

Ti0 ₂ films on glass substrates were prepared by an adsorptive self-assembly method as described in Example 1 above.

Photo-assisted deposition of Ag nanoparticles on Ti0 ₂ film

0.01 mol/L AgN0 ₃ aqueous solution was mixed with absolute ethanol (the volume ratio is 7:1 ) to obtain 0.00875 mol/L AgN0 ₃ solution (dissolved in water-ethanol). One piece of silica-supported Ti0 ₂ film was dipped in 8 mL of the above 0.00875 mol/L AgN0 ₃ solution with or without surfactant held in a 10 mL sample vial, which was then exposed to a ultraviolet (UV-A) light using a high-pressure mercury lamp (125 W, Philips) for different durations. A cut-off filter (wavelength ¹ of 385 nm) was placed in front of the lamp to remove all emission light less than 385 nm in wavelength. instrumentation and analysis of materials The size and crystalline nature of Ag nanoparticles were detected by transmission electron microscopy (TEM, JEM-2010). For the structural and morphological textures of examined samples, field emission scanning electron microscopy (FESEM, JSM-6700F) was employed. The surface and bulk chemical compositions of Ag nanoparticles were profiled with X-ray photoelectron spectroscopy (XPS, AXIS-HSi, Kratos Analytical) and energy-dispersive X-ray spectroscopy (EDX/FESEM, JEM-6700F) respectively. UV-Vis absorption spectra were recorded for different samples in the wavelength range of 300- 800 nm via a spectrophotometer (Shimadzu UV-2450). The detection of silver ion concentration in E. coli solution possibly dissolved by Ag nanoparticles was conducted via inductively coupled plasma spectrometry (ICP-MS, Agilent 7500). All of the solutions detected were prepared with ultrapure water in this detection.

Anti-bacteria tests of Ag nanoparticles on Ti0 ₂ film

E. coli K-12 was incubated in 30 mL of nutrient broth (0.9 g of TSB dissolved in 30 imL of Milli-Q water) at 37°C overnight at 150-200 rpm in a rotary shaker. Then, 1 mL overnight culture was subcultured in another 30 mL nutrient broth for incubation for 4 hours until reaching log phase. The cells were isolated by centrifugation at 3500 rpm for 13 minutes and washed two times with sterilized Milli-Q water to remove residual culture media components. The as-prepared cells were then resuspended and diluted to the required cell density of around 10 ⁸ colony-forming units per millilitre (CFU/mL) with sterilized Milli-Q water. 10 mL of E. coli suspension and one piece of film sample were added into a bottle and irradiated by a high-pressure mercury lamp (125 W, Philips) with a cut-off filter placed between reactors and UV lamp to shield the short wavelength UV light. For the bactericidal efficiency comparison of different samples, plate-counting method was carried out. Before and during the light irradiation, 0.5 mL of £. coli suspension was taken out and diluted serially with sterilized 0.9% NaCI solution to tune the bacterium concentration for precise and easy counting. 100 aliquots of diluted bacteria solution were spreading on the nutrient bottom agar in petri dish and cultured at 37 °C in oven overnight. All the bactericidal experiments were performed at room temperature; the measured data for each set of experiments were expressed with the mean and standard deviation. The E. coli mixture without any Ag nanoparticles or Ti0 ₂ film or both was exposed to the same UV lamp with the other condition identical as the reference. For the TEM detection of E. coli, all the bacteria samples were stained by 1.2% sodium phosphotungstate aqueous solution (pH of about 7.0) mixed with sodium hydroxide and phosphotungstate acid. The as-stained bacteria was taken by pipette and dropped on the copper grids, air-dried in a natural laboratory environment and investigated by the technique of aforementioned TEM (JEOL JEM- 2010).

Results and discussion

(a) Deposition of Ag nanoparticles/Ti0 ₂ thin film on plane glass substrates

Through the photoreduction of AgN0 ₃ , Ag nanoparticles were desirably deposited on Ti0 ₂ film. Figure 16 shows the top textures of Ag nanoparticles on Ti0 ₂ film prepared in the presence of 0.04 g PVP with different irradiation time under the UV lamp. By comparison with the pure Ti0 ₂ film, after deposition of Ag nanoparticles, the surface profile is much rougher. Even though the irradiation time was extended, the size of Ag nanoparticles grew little. However, the intensity of plasmon peaks increased with the exposure time to UV lamp, as shown in Figure 17, demonstrating the growth of more Ag nanoparticles Ti0 ₂ film. From this graph, it can be observed that the plasmon peak of Ag nanoparticles is located at around 450 nm. At the same time, the peak position moved to the longer wavelength as the irradiation time prolonged. During the preparation process, no extra free Ag nanoparticles were observed in the solution except those anchored on Ti0 ₂ film, which implied the formation of Ag nuclei only in the presence of Ti0 ₂ semiconductor.

Two control experiments were also carried out to examine the growth mechanism of Ag nanoparticles on Ti0 ₂ film. For the first control experiment, a piece of bare glass without Ti0 ₂ coating was used as substrate to deposit Ag nanoparticles with the same amount of AgN0 ₃ mixture. Regardless of the time the reactor was exposed to the UV lamp, no Ag nanoparticles formed. For the second control experiment, a piece of Ti0 ₂ film formed on a glass substrate was dipped into 8 mL 0.01 mol/L AgN0 ₃ aqueous solution without ethanol and was also exposed to the UV lamp. No Ag nanoparticles were deposited on the Ti0 ₂ film despite the longer irradiation time.

Based on the above analyses and experimental results, it was found that under the irradiation of UV lamp, the electron-hole pairs in Ti0 ₂ catalysts are photo-generated and can recombine quickly. Therefore, in the absence of ethanol, no Ag nanoparticles are formed. Ethanol serves as the hole-scavengers to allow hole-electron to efficiently separate and extend the lifetime of charge carriers. Thus, electrons disseminating on the surface of Ti0 ₂ film can initiate the reduction of cations absorbed on the surface of the Ti0 ₂ film.

The role of PVP in the silver formation process was studied by also studying the texture of samples without PVP. When the irradiation was for 0.5 hours, the size of silver nanoparticies was very small. As exhibited in Figure 18(a), Ag nanoparticles were not detected by SEM. However, as the exposure duration extended to 1 hour, Ag nanoparticles obviously extruded and spread on the Ti0 ₂ film as shown in Figure 18(b). An increase in the irradiation time resulted in the serious aggregation of newly deposited Ag nanoparticles with the former ones. As shown in Figure 18(d), the bulk nanoparticles dispersed on the film were as large as 80 nm when the irradiation time reached 3 hours. PVP, as the common structure-director of nanostructure formation, played a vital role in manipulating and controlling the morphology and size of nanomaterials by selectively binding on the surfaces of newborn nuclei. Accordingly, because of the addition of PVP in the synthesis of Ag nanoparticles, the size of the nanoparticles was well-controlled as represented in Figure 16. Figure 19 gives the absorption spectra of Ag nanoparticles on Ti0 ₂ film prepared without PVP. In accordance with the previous samples, the plasma absorption intensity increased with prolonging the irradiation time. At the same time, elementary analysis as shown in Figure 20 describes the atomic ratio of Ti, O, Ag of the studied samples shown in Figure 18. Based on the data obtained by EDX, silver content in the samples increased with longer growth time, which is consistent with the enhanced absorption intensity.

In view of the above, it was observed that the type of surfactant used determines the size of the silver nanoparticles. Therefore, Tween-20 and Tween-40 were also applied in the synthesis system of Ag nanoparticles. Figure 21 shows the top surface of different samples fabricated with Tween-20 and Tween-40. Figure 21 (a) and (b) shows Ag nanoparticles formed with Tween-20 as the structure-director while Figure 21 (c) and (d) shows Ag nanoparticles formed with Tween-40. As shown in the figure, the particle size shows no large difference even though different amounts of Tween-20 were employed between Figures 21 (a) and (b) and between Figures 21 (c) and (d). The size of the Ag nanoparticles for all amounts of Tween shown in Figure 21 was about 20 nm. However, as shown in Figure 21 (d), some ditches separate the Ag nanoparticles. Based on the absorption spectra of Figure 22, the intensities of plasma peaks decreased with the larger amount of Tween-20, which showed that more Tween- 20 retarded the growth rate of Ag.

In order to demonstrate the phase and structure of Ag nanoparticles spreading on the Ti0 ₂ film, the samples from glass were scratched using a sharp knife and the scratched samples were examined using TEM. The results obtained as shown in Figure 23. Figures 23(a) and (b) shows some wavelike structure marked by the arrow which confirms the presence of Ti0 ₂ film on glass substrate. The inset in Figure 23(b) is the corresponding selected-area electron diffraction (SAED) showing a series of diffraction rings to confirm the polycrystaliine nature. They can be indexed from inside to outside to be Ti0 ₂ (101 ), Ag (1 1 1 ), Ag (101 ), further demonstrating the anatase Ti0 ₂ and face- centred cubic Ag. An individual Ag nanoparticle is displayed in high-resolution Figure 23(c) with distinct lattice fringes. In order to examine the microstructure of Ag nanoparticles, the FFT-HRTEM of a portion of the nanoparticle of Figure 23(c) is shown in Figure 23(d) where the lattice spacing of 0.239 nm is corresponding to the plane distance of {1 1 1} of the fcc-structured Ag.

The Ag nanoparticles composited with Ti0 ₂ film was subjected to anti-bacteria tests. Figure 24 shows the bactericidal investigation results of different samples with the pure Ti0 ₂ film as the control experiment. From Figure 24(A), it can be seen that Ag nanoparticles enhance the anti-bacteria efficiency of Ti0 ₂ film to a large extent. By comparison between curve a and b in Figure 24(A), a small amount of Ag nanoparticles prepared by irradiation for just 0.5 h has better antibacterial behaviour. However, the UV lamp plays vital role in the bactericidal enhancement of Ag nanoparticles, as demonstrated by curves b and c. As for the sample, the bactericidal rate under exposure of UV lamp is higher than that in the dark. Electrons and holes will be generated by transferring electrons in valence band to conduction band of Ti0 ₂ on irradiation of incident lights. By the contact interface, the plasma metals can serve as the electron conductors to transport the electrons away from Ti0 ₂ surface and thus retard the possibility of electron-hole recombination. The residual free holes can react with oxygen, water to generate various oxygen species as immense oxidizer which was the main mechanism for Ti0 ₂ semiconductor to kill bacteria by the assistance of light. Therefore, the inactivation efficiency over E. Coli of Ag nanoparticles composited with Ti0 ₂ film can be improved by the irradiation of visible light. At the same time, with increasing the content of Ag loading on Ti0 ₂ film, the inactivation behaviour of samples becomes more effective, as evidenced by curve e in Figure 24(A).

The Ag ion concentrations of different samples after anti-bacteria were identified by ICP-MS technique and Figure 24(B) displays the corresponding results. For the small quantity of Ag nanoparticles prepared for shorter irradiation time (b, d points in Figure 24(B)), the Ag ion concentrations changed a little by comparison with pure Ti0 ₂ film (point "a" in Figure 24(B)). However, for the longer preparation time, Ag ion concentration increased fiercely to point "e" in Figure 24(B), which indicates that the dissolution of Ag ions from Ag nanoparticles occurred for this sample. The enhanced inactivation against E. Coli can also be explained because Ag ions own strong inhibitory and antimicrobial properties (for example, against E. coli and Staphylococcus aureus) even though oxygen-containing radicals have preferable bactericidal effect to the intrinsic cytotoxicity of Ag ions eluted from the compounds.

At the same time, in order to detect the possibility of E. Coli structure transformation during the inactivation process, TEM technique was applied for the morphology investigation of E. Coli. Figure 25 exhibits the TEM images of E. Coli experiencing different inactivation duration. From Figure 25(a) and (b), the pristine E. Coli cells showed integrated structure and shape. After bactericidal operation for 2 hours, the structure of E. Coli cells showed some defects, such as holes marked by the arrows in Figure 25(c). In particular, judging from the clear contrast of higher-resolution Figure 25(d), the even distribution of E. Coli cell composition has been destroyed. As the inactivation time was extended, the structure of E. Coli cells went through further destruction and the whole cell texture was completely crushed, as demonstrated by the areas marked by the arrows in Figure 25(e). Figure 25(f) shows a typical TEM showing the complete destruction of a single E. Coli cell.

Conclusion

From this example, it can be seen that Ag nanoparticles anchored on Ti0 ₂ film can improve the bactericidal efficiency when compared to pure Ti0 ₂ films. As the content of Ag nanoparticles increase, the Ag ions eluted from the Ag nanoparticles take part in the antibacterial process, which makes the Ti0 ₂ film with more Ag loading have better bactericidal behaviour. Example 3

Preparation of doped Ti0 ₂ nanoparticles

0.20 g of titanium (IV) n-propoxide (98%, Alfa Aesar) was added into 10.0 mL of toluene (AR, Aldrich). Then 6.0 mL of oleic acid (OA, CH ₃ (CH ₂ ) ₇ CH=CH(CH ₂ )7COOH, 90%, Aldrich), 0.10 mL of tert-butylamine (98%, Aldrich) and 10.0 mL of fresh Fe(N0 ₃ ) ₃ aqueous solution with different concentrations were added into the above mixture. The resultant solution was magnetically stirred for 15 minutes and then transferred into a Teflon-lined stainless steel autoclave. The solvothermal reaction was performed in an electric oven which had already been heated to 180°C for a period of 7-12 hours. After the synthesis, the product mixture was separated into oil and water phases. The oil phase was taken out and added with an equivalent volume of ethanol in order to precipitate the Fe(lll) doped Ti0 ₂ nanoparticles. After centrifuging, the solvent was decanted and Fe(lll) doped Ti0 ₂ nanoparticles were redispersed in 16.0 mL of toluene as a colloidal Ti0 ₂ suspension.

To calculate the molar ratio of Fe(lll) in the doped Ti0 ₂ suspension, the formula used is as follows:

% of Fe(lll) = ^ x l00%

[Fe + Ti]

Accordingly, the Fe(N0 ₃ ) ₃ aqueous solution having the calculated concentration will be prepared for use in preparing the monodispersed Fe(lll) doped Ti0 ₂ nanoparticles.

Fabrication of Fe-doped Ti0 ₂ films on glass substrate

Soda lime glass substrates (1 .5 cm ^χ 2.5 cm ^χ 1.0-1.2 mm; Sail brand, China) were cleaned in a Piranha solution as described in Example 1 above. After the cleaning treatment, the glass substrates were first dipped in a pure colloidal Ti0 ₂ suspension (as prepared in Example 1 ) for 2 hours to enable adsorption of Ti0 ₂ nanoparticles onto the substrate surface. The samples were removed from the colloidal suspension and washed with toluene, followed by a 2 hour heat treatment in an electric furnace at 400°C in laboratory air. Each glass substrate with the Ti0 ₂ film on the glass substrate surface was then placed in Fe(lll) doped Ti0 ₂ colloidal suspensions of different molar ratios (i.e., 0.3, 0.5, 0.8 and 1.2%) respectively, to adsorb Fe(lll) doped Ti0 ₂ nanoparticles. The glass substrates were then taken out and heated in the same manner as described above. Ti0 ₂ /Fe doped Ti0 ₂ non-symmetric films were obtained on the glass substrates.

Absorption spectra

The absorption spectra of the prepared Fe doped Ti0 ₂ films were measured. All the sample films were exposed to UV irradiation (UV lamp, 125 W). The results obtained are shown in Figure 26. The percentage indicated in Figure 26 refers to the % of Fe present in the Ti0 ₂ film. The samples were measured against a sample of pure Ti0 ₂ film.

Characterisation of the film

The structural and morphological profiles of the Fe-doped Ti0 ₂ thin films formed on the glass substrates were obtained using field emission scanning electron microscopy (FESEM, JSM-6700F, accelerating voltage: 5kV, current: 10 μΑ). The FESEM images obtained for each of the samples are shown in Figure 27, in which Figures 27(a) and (b) provide the images for 0.3% Fe:Ti0 ₂ thin film at different magnifications respectively, Figures 27 (c) and (d) provide the images for the 0.5% Fe:Ti0 ₂ thin film at different magnifications, Figures 27 (e) and (f) provide the images for the 0.8% Fe:Ti0 ₂ thin film at different magnifications and Figures 27 (g) and (h) provide the images for the 0.8% Fe:Ti0 ₂ thin film at different magnifications.

It can be observed that the Fe doped Ti0 ₂ nanoparticles spread uniformly on the glass substrate, thus forming a conformal thin film. Thus, while the doping can improve the properties of the thin film, the conformity of the thin film formed from the doped nanoparticles is not affected. Further, the average size of the nanoparticles remain small and are about 23 nm. Even when doping concentration is increased from 0.3% to 1.2% Fe, cracking is not observed, demonstrating the effective adsorptive self- assembly of the nanoparticles in the film formation.

Photocatalysis

The photocatalytic investigation for the glass supported Ti0 ₂ /Fe doped Ti0 ₂ films was carried out with a reference reaction of decomposition of methyl orange (C ₁₄ H ₁₄ N ₃ Na0 ₃ S, Merck) in laboratory air at room temperature (25°C) and under ambient pressure. A cylindrical wide-mouth glass bottle (capacity of about 30 mL) was used as the reactor vessel. The glass supported Ti0 ₂ films (area at 1.5 cm ^χ 2.5 cm) with the difference Fe concentrations were dipped in 4.Ό mL of aqueous methyl orange solution with the concentration of 2.0 mg/L and then exposed to ultraviolet (UV) light using a high-pressure mercury lamp (125 W, Philips). A cut-off filter (wavelength of 385 nm) was placed in front of the lamp which removed all emission lines of wavelength less than 385 nm. UV-vis absorption spectra of methyl orange were recorded at constant intervals to monitor the progress of photocatalytic reactions, as shown in Figure 28.

Fabrication of 3% and 5% Fe doped Ti0 ₂ films

The same process as described above for Fe doped Ti0 ₂ films was used to prepare 3% and 5% Fe doped Ti0 ₂ nanoparticles. After the solvothermal reaction, the upper oil phase was collected and washed using ethanol in order to precipitate the Fe(ili) doped Ti0 ₂ nanoparticles. After centrifugtng, the solvent was decanted and Fe(lll) doped Ti0 ₂ nanoparticles (3% and 5% Fe doped) were transferred to a ceramic crucible. The crucibles were heated in a box furnace at 400°C in laboratory air for 2 hours. The x-ray diffraction analysis carried out on the samples shows that even after the heating, the nanoparticles remained in the anatase phase.

Whilst the foregoing description has described exemplary embodiments, it will be understood by those skilled in the technology concerned that many variations in details of design, construction and/or operation may be made without departing from the present invention.