A method of oxidising an inorganic amine to nitrate Technical Field [1] The invention relates to a method of oxidising an inorganic amine to nitrate. The method comprises contacting an aqueous solution comprising inorganic amine with a titanium dioxide photocatalyst and irradiating the aqueous solution to photocatalytically oxidise the inorganic amine to nitrate, wherein the titanium dioxide photocatalyst is contacted with a phosphorous-based species before or during the oxidising. The invention also relates to a method of fertilizing a crop. Background of Invention [2] The invention of the Haber-Bosch process in the 20 ^th century provided for the first time an industrial route to produce large volumes of synthetic ammonia from dinitrogen (N ₂ ). However, due to the exceptional stability of the N ₂ triple bond (N≡N, 942 kJ mol ^-1 ), the Haber-Bosch process requires extreme reaction conditions of elevated pressure (150-350 atm) and temperature (400-550°C), as well as a supply of pure H ₂ which is typically sourced from the steam reforming process of natural gas. Consequently, the process consumes approximately 2% of global energy supply and contributes ~1.5% of global greenhouse gas emissions. [3] A major use of industrially produced ammonia is in the manufacture of synthetic nitrate-based fertilizers. Ammonia produced in the Haber-Bosch process is thus oxidised to nitrate (as nitric acid) in the Ostwald process. In this process, ammonia is reacted with oxygen in a stepwise process to optimise the yield of nitric acid according to the overall reaction in equation (1). However, the Ostwald process also produces nitrogen oxide by-products such as N ₂ O, which is a potent greenhouse gas. NH ₃ + 2 O ₂ → H ₂ O + HNO ₃ . (1) [4] Given the large environmental impacts of the Haber-Bosch and Ostwald processes, there is an urgent need for sustainable and decentralized technologies for converting N ₂ to ammonia, and then oxidizing this ammonia to the nitrate form useful for fertilizers. [5] The sustainability of ammonia synthesis can be improved to an extent by replacing fossil fuel-derived H ₂ in the Haber-Bosch process with renewable H ₂ produced by water hydrolysis. More fundamentally, significant recent progress has also been made towards a commercially viable electrochemical nitrogen reduction reaction (NRR) process, in which renewable electricity is directly converted into ammonia in a simple electrolytic cell. The cathodic half-reaction of the NRR is shown in equation (2): N ₂ + 6H ⁺ + 6 e- → 2NH ₃ (2) [6] Examples of NRR processes having faradaic efficiencies of over 50% for ammonia are described in the international patent applications published as WO2017/132721 and WO2022/020904. In another approach, N ₂ is electrochemically reduced to haloamines at very high rates and faradaic efficiencies, as described in the international patent application published as WO2021/108859. [7] The use of electrochemical approaches to N ₂ reduction provides the opportunity for small scale and distributed production of inorganic amines such as ammonia and haloamines. However, there remains a need for similarly-scaled technology capable of selectively oxidising the resultant ammonia or haloamines to nitrates useful in fertilizers. [8] Ammonia is known as an environmentally harmful pollutant in aqueous streams such as wastewater. Various technologies have been proposed for remediating ammonia-contaminated aqueous compositions, including photocatalytic and photoelectrocatalytic methods. Thus, for example, Lee et al (Environ. Sci. Technol. 2002, 36, 5462-5468) reported selective photocatalytic oxidation of aqueous ammonia to N ₂ on platinum-doped titanium dioxide (TiO ₂ ). [9] Photocatalytic oxidation of aqueous ammonia using undoped TiO ₂ as the photocatalyst has also been investigated for water-remediation applications, for example as reported by Zhu et al (Environ. Sci. Technol.2005, 39, 3784-3791). In this case, ammonia is preferentially oxidised via a nitrite (NO ₂ -) intermediate to nitrate (NO ₃ -). However, significant conversions of ammonia were only achievable at very high pH values (> 10) and no conversion was obtained at neutral or acidic pH values (≤ 7). [10] For fertilizer applications, however, photooxidation of ammonia to nitrate at high pH values is undesirable. Sufficiently high pH values (>10) for significant ammonia conversion are generally provided and maintained during the photooxidation reaction by addition of an inorganic base, which must be added in excess to neutralise the protons released during ammonia oxidation. The resultant nitrate-containing solution thus contains higher levels of metal cations than are generally desirable for fertilizer applications. Furthermore, oxidation of the nitrite intermediate to nitrate is relatively inhibited at high pH values. Therefore, there is a risk that an unacceptably high amount of nitrite remains present in the fertilizer solution, particularly considering that only partial conversion of ammonia would typically be targeted (i.e. to produce ammonium nitrate). [11] There is therefore an ongoing need for methods of oxidising an inorganic amine to nitrate which at least partially address one or more of the above-mentioned short-comings, or provide a useful alternative. [12] A reference herein to a patent document or other matter which is given as prior art is not to be taken as an admission that the document or matter was known or that the information it contains was part of the common general knowledge as at the priority date of any of the claims. Summary of Invention [13] The inventors have discovered that the TiO ₂ -photocatalyzed oxidation of inorganic amines, including ammonia and haloamines, can be enhanced in aqueous solution by contacting the TiO ₂ photocatalyst with a phosphorous-based species, such as a phosphate or phosphonate. Typically, the phosphorous-based species is dissolved in the aqueous solution. Surprisingly, the phosphorous-based species improves the rate of photooxidation at weakly alkaline pH values and provides efficient and highly selective conversion of ammonia to nitrate at neutral and even weakly acidic pH values. Advantageously, aqueous nitrate-containing solutions suitable for fertilizer applications can thus be produced without excessive, or indeed any, metal cations in solution. Only a small amount of the phosphorous-based species (sub-stoichiometric relative to ammonia) is required to achieve this improvement, although certain phosphorous-based species (such as phosphates) are in fact a useful component of fertilizer solutions and may be included above the minimum functional requirement to achieve a desirable P:N ratio. [14] The TiO ₂ photocatalyst for the process is typically an undoped TiO ₂ comprising both rutile and anatase phases. The inventors have also discovered that surprisingly significant improvements in photocatalytic performance can be obtained by heat-treating a TiO ₂ precursor in a low oxygen environment, such as in flowing Ar gas at 600°C, to convert a portion of the anatase phase to rutile. Optionally, the TiO ₂ photocatalyst further comprises a co-catalyst, typically decorated in the form of metallic nanoparticles decorated on the TiO ₂ surface. The co-catalyst has also been found to significantly increase the rate of inorganic amine conversion. [15] In accordance with a first aspect the invention provides a method of oxidising an inorganic amine to nitrate, the method comprising: contacting an aqueous solution comprising at least one inorganic amine selected from ammonia and haloamine with a titanium dioxide photocatalyst; and irradiating the aqueous solution with light, thereby photocatalytically oxidising at least a portion of the inorganic amine to nitrate, wherein the titanium dioxide photocatalyst is contacted with at least one phosphorous-based species before or during the oxidising. [16] In some embodiments, the at least one phosphorous-based species is selected from the group consisting of phosphorous oxoacids and salts and esters thereof. [17] In some embodiments, the at least one phosphorous-based species is dissolved in the aqueous solution. [18] In some embodiments, at least a portion of the inorganic amine is oxidised to nitrate at a pH of below 8.5, or below 8, such as below 7.5. [19] In some embodiments, at least 15 mol% of the inorganic amine is oxidised to nitrate at a pH of below 8.5, or below 8, such as below 7.5. [20] In some embodiments, at least a portion of the inorganic amine is oxidised to nitrate at a pH of below 7, preferably below 6.5, more preferably below 6. [21] In some embodiments, the at least one phosphorous-based species is selected from the group consisting of phosphoric acid and salts and esters thereof. [22] In some embodiments, the at least one phosphorous-based species comprises a phosphate. The phosphate may comprise at least one selected from orthophosphate (PO ₃ ^3- ), hydrogen phosphate (HPO ₄ ^2- ) and dihydrogen phosphate (H ₂ PO ₄ -). [23] In some embodiments, the molar ratio of N:P in the aqueous solution is greater than 1:1, or greater than 2:1, such as greater than 5:1. [24] In some embodiments, at least 30 mol%, or at least 40 mol%, such as in the range of 40 to 60 mol%, of the inorganic amine is oxidised to nitrate. [25] In some embodiments, the aqueous solution is substantially free of nitrite after oxidising the at least a portion of the inorganic amine to nitrate. [26] In some embodiments, no more than 20% of the inorganic amine, such as no more than 10% of the inorganic amine, is oxidised to gases selected from N ₂ , N ₂ O and NOx. [27] In some embodiments, the at least one inorganic amine has an initial concentration in the aqueous solution of at least 1 mM, or at least 2 mM, such as at least 5 mM. [28] In some embodiments, the molar ratio of N:M in the aqueous solution is greater than 1:1 after oxidising the at least a portion of the inorganic amine to nitrate, where M is the total amount of alkali and alkali earth metal cations. [29] In some embodiments, the aqueous solution is substantially free of alkali and alkali earth metal cations after oxidising the at least a portion of the inorganic amine to nitrate. [30] In some embodiments, the titanium dioxide photocatalyst comprises rutile and anatase phases in a ratio (w/w) of 20:80 to 45:55, such as 25:75 to 40:60. [31] In some embodiments, the titanium dioxide photocatalyst comprises undoped TiO ₂ . [32] In some embodiments, the titanium dioxide photocatalyst is a nanoparticulate titanium dioxide. [33] In some embodiments, the titanium dioxide photocatalyst is prepared by heat-treating a titanium dioxide precursor in a low-oxygen environment, thereby converting a portion of anatase in the titanium dioxide precursor to rutile. The titanium dioxide precursor may be heat-treated at a temperature of above 400°C, or above 500°C, or above 550°C, such as about 600°C. [34] In some embodiments, the titanium dioxide photocatalyst comprises particulate TiO ₂ , preferably nanoparticulate TiO ₂ , decorated with a metallic co-catalyst. In some embodiments, the metallic co-catalyst comprises a noble metal. The noble metal may be selected from the group consisting of platinum, silver, gold, copper, palladium, rhodium, ruthenium, iridium, osmium and combinations thereof. In some embodiments the noble metal is silver or platinum. In some embodiments, the metallic co-catalyst comprises metallic nanoparticles comprising a noble metal selected from the group consisting of platinum, silver, gold, copper, palladium, rhodium, ruthenium, iridium, osmium and combinations thereof. [35] In some embodiments, the at least one inorganic amine is selected from ammonia and monochloroamine. [36] In some embodiments, the aqueous solution is irradiated with sunlight. [37] In some embodiments, the aqueous solution comprises dissolved dioxygen (O ² ) when irradiated. The aqueous solution may be in contact with air when irradiated. [38] In some embodiments, the titanium dioxide photocatalyst is dispersed in the aqueous solution or coated on a solid substrate, optionally a transparent substrate, which is in contact with the aqueous solution. [39] In some embodiments, the aqueous solution, after irradiating the aqueous solution with light and oxidising the at least a portion of the inorganic amine to nitrate, is a fertilizer solution comprising: total nitrogen in a concentration range of 1 – 200 mM, nitrate in a concentration range of 1 – 200 mM, phosphate in a range of 0.5 – 20 mM, and optionally metal cations selected from K ⁺ , Ca ²⁺ and Mg ²⁺ in a total concentration range of 0.1 – 200 mM. [40] In accordance with a second aspect, the invention provides a method of fertilizing a crop, the method comprising: contacting an aqueous solution comprising at least one inorganic amine selected from ammonia and haloamine with a titanium dioxide photocatalyst; irradiating the aqueous solution with light, thereby photocatalytically oxidising at least a portion of the inorganic amine to nitrate, wherein the titanium dioxide photocatalyst is contacted with at least one phosphorous-based species before or during the oxidising; and applying the aqueous solution containing nitrate to a crop as a fertilizer. [41] In some embodiments, the at least one phosphorous-based species comprises a phosphate. [42] In some embodiments, the method further comprises adding at least one mineral base comprising K, Ca or Mg to the aqueous solution before or during the oxidising. [43] In some embodiments, the aqueous solution containing nitrate, when applied to the crop as a fertilizer, comprises total nitrogen in a concentration range of 1 – 200 mM, nitrate in a concentration range of 1 – 200 mM, phosphate in a range of 0.5 – 20 mM, and optionally metal cations selected from K ⁺ , Ca ²⁺ and Mg ²⁺ and combinations thereof in a total concentration range of 0.1 – 200 mM. [44] Other embodiments of the second aspect may generally be in accordance with embodiments of the first aspect as disclosed herein. [45] Where the terms “comprise”, “comprises” and “comprising” are used in the specification (including the claims) they are to be interpreted as specifying the stated features, integers, steps or components, but not precluding the presence of one or more other features, integers, steps or components, or group thereof. [46] Further aspects of the invention appear below in the detailed description of the invention. Brief Description of Drawings [47] Embodiments of the invention will herein be illustrated by way of example only with reference to the accompanying drawings in which: [48] Figure 1 is a graph depicting the concentration of reactant and product nitrogen species in aqueous solution over time when photocatalytically oxidising monochloroamine with unmodified TiO ₂ , as done in Example 2. [49] Figure 2 is a graph depicting the concentration of reactant and product nitrogen species in aqueous solution over time when photocatalytically oxidising monochloroamine with TiO ₂ heat-treated in Ar gas, as done in Example 2. [50] Figure 3 is a graph depicting the concentration of ammonia in aqueous solution over time when photocatalytically oxidising different ammonia sources with a TiO ₂ photocatalyst (heat-treated in Ar gas) at different pH values and in the presence or absence of phosphate, as done in Examples 3 and 4. [51] Figure 4 is a graph depicting the concentration of nitrate in aqueous solution over time when photocatalytically oxidising different ammonia sources with a TiO ₂ photocatalyst (heat-treated in Ar gas) at different pH values and in the presence or absence of phosphate, as done in Examples 3 and 4. [52] Figure 5 is a graph depicting the concentration of ammonia in aqueous solution over time when photocatalytically oxidising ammonia with a TiO ₂ photocatalyst (heat-treated in Ar gas) in the presence of either phosphate or borate, as done in Examples 4 and 5. [53] Figure 6 is a graph depicting the concentration of nitrate in aqueous solution over time when photocatalytically oxidising ammonium with a TiO ₂ photocatalyst (heat- treated in Ar gas) in the presence of either phosphate or borate, as done in Examples 4 and 5. [54] Figure 7 is a graph depicting the concentration of reactant and product nitrogen species in aqueous solution over time when photocatalytically oxidising ammonia with a TiO ₂ photocatalyst (heat-treated in Ar gas) in the presence of 4 mM phosphate, as done in Example 6. [55] Figure 8 is a graph depicting the concentration of reactant and product nitrogen species in aqueous solution over time when photocatalytically oxidising ammonia with a TiO ₂ photocatalyst (heat-treated in Ar gas) in the presence of 4 mM borate, as done in Example 6. [56] Figure 9 is a graph depicting the concentration of reactant and product nitrogen species in aqueous solution over time when photocatalytically oxidising monochloroamine with a TiO ₂ photocatalyst (heat-treated in Ar gas) in the presence of phosphate, as done in Example 7. [57] Figure 10 is a graph depicting the initial rate of photocatalytic oxidation of ammonia, with various initial ammonia concentrations, for three different TiO ₂ catalysts in the presence of phosphate, as investigated in Example 8. [58] Figure 11 is a graph depicting the first order rate constant of photocatalytic oxidation of ammonia, with various initial ammonia concentrations, for three different TiO ₂ catalysts in the presence of phosphate, as investigated in Example 8. [59] Figure 12 is a graph depicting the selectivity of photocatalytic oxidation of ammonia, for three different TiO ₂ catalysts in the presence of phosphate, as investigated in Example 8. [60] Figure 13 is a graph depicting the concentration of ammonia in aqueous solution over time when photocatalytically oxidising ammonium with a TiO ₂ photocatalyst (heat-treated in Ar gas) in the presence of phosphate, under either air or argon atmosphere, as done in Example 9. [61] Figure 14 is a graph depicting the concentration of nitrate in aqueous solution over time when photocatalytically oxidising ammonium with a TiO ₂ photocatalyst (heat-treated in Ar gas) in the presence of phosphate, under either air or argon atmosphere, as done in Example 9. [62] Figure 15 is a graph depicting the concentration of reactant and product nitrogen species in aqueous solution over time when photocatalytically oxidising ammonia with a TiO ₂ photocatalyst (heat-treated in Ar gas) in the presence of phosphate, as done in Example 10. [63] Figure 16 is a transmission electron microscope (TEM) image depicting the deposited Ag nanoparticles on the TiO ₂ photocatalyst particles, as produced in Example 14. [64] Figure 17 is a graph depicting the concentration of reactant and product nitrogen species in aqueous solution over time when photocatalytically oxidising ammonia with an Ag/TiO ₂ photocatalyst (TiO ₂ heat-treated in Ar gas) in the presence of phosphate, as done in Example 15. [65] Figure 18 is a graph depicting the concentration of reactant and product nitrogen species in aqueous solution over time when photocatalytically oxidising ammonia with a TiO ₂ photocatalyst (heat-treated in Ar gas) in the presence of phosphonate, as done in Example 16. Detailed Description [66] The present invention relates to a method of oxidising an inorganic amine to nitrate. The method comprises contacting an aqueous solution comprising at least one inorganic amine selected from ammonia and haloamine with a titanium dioxide photocatalyst, and irradiating the aqueous solution with light so that at least a portion of the inorganic amine is photocatalytically oxidised to nitrate. The titanium dioxide photocatalyst is contacted with at least one phosphorous-based species before or during the oxidising, for example because it is dissolved in the aqueous solution. Aqueous solution comprising at least one inorganic amine [67] The photooxidation process is conducted in an aqueous solution, which thus contains water. In some embodiments, the aqueous solution comprises at least 10 wt.% water. In some embodiments, water is the main constituent of the aqueous solution, so that the aqueous solution comprises at least 50 wt.% water, or at least 90 wt.% water. In some embodiments, the aqueous solution comprises water as the only solvent. [68] The aqueous solution comprises the inorganic amine(s) to be oxidised. In some embodiments, the inorganic amine comprises or consists of ammonia. It will be appreciated that ammonia exists in aqueous solution as an equilibrium between the neutral ammonia species (NH ₃ ) and the ammonium cation species (NH ₄ +), with the proportion of each being dependent on the pH. As used herein, the term “ammonia” refers to both NH ₃ and NH ₄ +, and the “ammonia concentration” refers to the combined concentration of both neutral and cationic ammonia species. [69] In some embodiments, the inorganic amine comprises or consists of one or more haloamines. Haloamine molecules generally have the formula NH _y X _3-y , where X is a halogen and y is selected from 0, 1 or 2. In some embodiments, X is selected from Cl and Br. In some embodiments, the inorganic amine comprises one or more chloroamines. Suitably, the chloroamines may comprise, or consist of, monochloroamine. [70] In some embodiments, the inorganic amine(s) are present at an initial concentration in the aqueous solution (i.e. prior to oxidation) of at least 1 mM, or at least 2 mM, such as at least 5 mM. If the nitrate product is to be used in the aqueous solution, the initial concentration of the inorganic amine may be selected to achieve a desired concentration of nitrate, or a desired total nitrogen concentration (including both unreacted inorganic amine, ammonia and nitrate product), for the target application. For example, the aqueous solution following oxidation may be used as a fertilizer solution, with little or no further work-up following the oxidation step. In this case, the concentration of inorganic amine(s) may be present initially in a corresponding amount to the total nitrogen concentration desired in the fertilizer solution, such as in the range of 1 to 20 mM. [71] The inorganic amine(s) may be obtained from any source. However, in some embodiments, the inorganic amine is produced by reduction of dinitrogen (N ₂ ), for example electrochemical reduction of N ₂ . Previously reported methods for the electrochemical reduction of N ₂ to ammonia or haloamines may be suitable, for example those reported in WO2017/132721, WO2022/020904 and WO2021/108859. [72] In some embodiments, the at least one phosphorous-based species is dissolved in the aqueous solution, together with the inorganic amine. The phosphorous-based species may be dissolved in the aqueous solution before or after the solution is contacted with the titanium dioxide photocatalyst, and indeed before or after commencing the oxidation of the inorganic amine provided that at least a portion of the photocatalytic oxidation takes place after the phosphorous-based species is dissolved in the aqueous solution. [73] When the at least one phosphorous-based species is dissolved in the aqueous solution, the solution will contain both (i) nitrogen species (N), including the at least one inorganic amine and oxidised derivatives thereof, and (ii) phosphorous species (P), including the at least one phosphorous-based species that was added and any derivatives thereof formed in situ. In principle, any molar ratio of N:P may be used. In some embodiments, however, the molar ratio of N:P in the aqueous solution is greater than 1:1, for example greater than 2:1, such as greater than 5:1. This may be preferred in some embodiments because the P species are simply used as photooxidation promotors, and it is desirable to minimise the amount of such promotor species relative to the reactant and product N species. In other embodiments, in particular where the nitrate-containing aqueous solution is to be used in certain fertilizer applications, both N and P are important fertilizer components but elevated N:P ratios are still desirable based on the target composition of the fertilizer. The inventors have demonstrated that excellent conversions of ammonia (close to 50% conversion to nitrate) can be achieved in an aqueous solution having an N:P molar ratio of 10:1, based on the initial ammonia and phosphate concentrations. Titanium dioxide photocatalyst [74] The methods of the present disclosure use a titanium dioxide (TiO ₂ ) photocatalyst to photocatalytically oxidise at least a portion of the inorganic amine to nitrate. [75] TiO ₂ is known as a photocatalyst for many chemical conversion processes. In some embodiments, the TiO ₂ photocatalyst is a particulate solid material, for example a nanoparticulate material such as P25 TiO ₂ powder available from Degussa Co., Ltd. (Germany). Such particulate materials are considered particularly suitable in the presently disclosed methods because they have a high surface area and can be well-dispersed in the aqueous solution, thus facilitating high reaction rates in the photocatalytic oxidation reaction. [76] Particulate TiO ₂ photocatalysts may be used at any suitable concentration when dispersed in the aqueous solution. Higher concentrations may be preferred to increase the reaction rate, although excessive concentrations may reduce the transparency of the solution and thus inhibit the transmission of actinic light through the solution. In some embodiments, the concentration is in the range of 0.01 to 10 g/litre, such as in the range of 0.1 to 1.5 g/litre. Alternatively, the particulate titanium dioxide photocatalyst may be coated onto or otherwise formed on the surface of a solid substrate which the aqueous solution contacts. Optionally, the substrate may be transparent, such as glass, thus allowing the photocatalyst to be irradiated through the substrate. [77] Both doped and undoped TiO ₂ materials are known as photocatalysts. While doped TiO ₂ photocatalyst are not excluded from the methods disclosed herein, some doped TiO ₂ materials may be less suitable, or undesirable, if the selectivity of the reaction to the desired nitrate product is adversely affected. In some embodiments, therefore, the titanium dioxide photocatalyst is an undoped TiO ₂ . As used herein, an undoped TiO ₂ refers to a TiO ₂ which lacks dopant atoms that significantly affect the electronic band structure of the semiconductive TiO ₂ . [78] In some embodiments, the titanium dioxide photocatalyst comprises both rutile and anatase phases. It has been found by experiment that the relative abundance of rutile and anatase phases in the photocatalyst is correlated with the photocatalytic performance. Accordingly, the ratio (w/w) of rutile to anatase in the titanium dioxide photocatalyst may be in the range of 20:80 to 45:55, or in the range of 25:75 to 40:60, such as in the range of 25:75 to 35:65. The phase composition of the titanium dioxide photocatalyst can be determined by X-Ray Diffraction (XRD) with Rietveld refinement. [79] The methods disclosed herein may be performed using as-received TiO ₂ , for example P25 TiO ₂ , as the titanium dioxide photocatalyst. However, a significantly improved performance has been obtained when the TiO ₂ is subjected to a heat- treatment step, and in particular a heat treatment step in a low-oxygen environment. Without wishing to be limited by any theory, it is proposed that the heat treatment improves photocatalytic performance by (i) converting amorphous TiO ₂ to crystalline phases, (ii) by decreasing the number of defects which can act as recombination centres for photogenerated charge carriers (holes and electrons) and thus limit the rate of photocatalysis, and (iii) by transforming a portion of the anatase phase to rutile. In at least some commercially available TiO ₂ materials, including P25, the ratio (w/w) of rutile to anatase may be undesirably low. By subjecting the material to a heat treatment step in a low oxygen environment, a desirable ratio (w/w) of rutile to anatase is obtained. In oxygen-rich environments, such as air, there is a risk of exceeding the preferred range of rutile to anatase ratios. [80] In some embodiments, therefore, the titanium dioxide photocatalyst is prepared by heat-treating a titanium dioxide precursor, for example an as-received TiO ₂ material such as P25 TiO ₂ , in a low-oxygen environment to convert a portion of anatase in the titanium dioxide precursor to rutile. As used herein, a low-oxygen environment refers to an environment with little or no O ₂ present, for example an inert gas with less than 1000 pm O ₂ (preferably less than 100 ppm O ₂ , or less than 10 ppm O ₂ ), or a vacuum. The titanium dioxide precursor may be heat-treated at a temperature of above 400°C, or above 500°C, or above 550°C, such as about 600°C. The titanium dioxide precursor may be heated at such temperatures for a time sufficient to produce a desired ratio (w/w) of rutile to anatase, for example in the range of 20:80 to 45:55, or in the range of 25:75 to 40:60, such as in the range of 25:75 to 35:65. [81] In some embodiments, the titanium dioxide photocatalyst comprises a co- catalyst. As used herein, a co-catalyst refers to a solid composition present as a distinct phase from the titanium dioxide phase of the photocatalyst but which modulates the photocatalytic behaviour of the titanium dioxide phase, for example as observed by an increased rate of photocatalytic oxidation of inorganic amine. As such, a co-catalyst may be distinguished from a dopant which is incorporated into the titanium dioxide phase. [82] The titanium dioxide photocatalyst may comprise particulate and preferably undoped TiO ₂ , as disclosed herein, which is decorated with a co-catalyst. In other words, the co-catalyst is supported on the surface of the TiO ₂ particles. Preferably, the co-catalyst is present in the form of nanoparticles which are dispersed over the surface of the titanium dioxide particles. In some embodiments, the nanoparticles are predominantly have an average particle size of below 10 nm, such as below 5 nm, for example as measured by transmission electron microscopy. [83] The co-catalyst may comprise a metallic composition, being a composition comprising one or more metal elements in reduced, i.e. zero-valent, metallic form. The co-catalyst may comprise a transition metal, preferably a noble metal. Suitable noble metals may include platinum, silver, gold, copper, palladium, rhodium, ruthenium, iridium, osmium and combinations thereof. Particularly good experimental results have been obtained with silver and platinum. [84] Advantageously, the presence of a suitable co-catalyst has been found to increase the initial reaction rate of inorganic amine oxidation and to reduce the time required to obtain a commercially significant conversion of inorganic amine to nitrate (e.g. in the order of 50% as required for oxidation of ammonia to ammonium nitrate). [85] Without wishing to be limited by any theory, the enhanced photocatalytic performance may be attributed to the better utilization of incident irradiation. Metal particles when in contact with semiconductors create Schottky barriers (φB) by the different work function of metal and the electron affinity of the semiconductor conduction band. This barrier may promote enhanced charge separation by efficient electron transfer between TiO ₂ and metallic co-catalyst. Moreover, plasmonic co- catalyst materials could further facilitate reactions by plasmon resonance effects. Metallic nanoparticles comprising noble metal elements, as disclosed herein, are known to exhibit plasmonic resonance and it is expected that this may contribute to enhanced photocatalytic oxidation of inorganic amine. [86] Metallic co-catalysts may be produced on the surface of titanium dioxide photocatalysts by photoreduction of an aqueous solution of a suitable metal precursor salt in the presence of the titanium dioxide. Contacting the titanium dioxide photocatalyst with a phosphorous-based species [87] The titanium dioxide photocatalyst is contacted with a phosphorous-based species before or during the photocatalytic oxidising of the inorganic amine. [88] In some embodiments, as already disclosed herein, the phosphorous-based species may be contacted with the titanium dioxide photocatalyst because it is a dissolved component of the aqueous solution. However, it is not excluded that the titanium dioxide photocatalyst is pre-treated with the phosphorous-based species, for example to modify the surface properties of the solid photocatalyst, before contacting the aqueous solution with the titanium dioxide photocatalyst. [89] Without wishing to be limited by any theory, the inventors propose that the phosphorous-based species interacts with the surface of the titanium dioxide photocatalyst, thus affecting its photocatalytic properties when in contact with the aqueous solution and irradiated by light. In particular, the phosphorous-based species may lower the point of zero charge (pH _pzc , also known as the isoelectric point) of the titanium dioxide photocatalyst, thus enhancing the rate of photocatalyst oxidation at mildly alkaline conditions and allowing catalytic photooxidation to occur at neutral and mildly acidic conditions. [90] The pH _pzc of unmodified TiO ₂ is reported to be in the range of 6.25 to 7. When the pH of an aqueous solution in contact with TiO ₂ is greater than pH _pzc , the surface of TiO ₂ is deprotonated and thus negatively charged, as shown in equation (3). When the pH is less than pH _pzc , the surface of TiO ₂ is protonated and thus positively charged, as shown in equation (4): TiOH ⇋ TiO ⁻ + H ⁺ pH > pH _pzc (3) TiOH + H ⁺ ⇋ TiOH ₂ ⁺ pH < pH _pzc (4) [91] At pH values below pH _pzc , the photocatalytic surface will thus repel cationic species, which have the same charge as the surface. At pH values above pH _pzc , the photocatalytic surface will strongly interact with cationic species. [92] When photocatalytically oxidising an inorganic amine in an aqueous solution, the pH of the solution tends to reduce because protons are released from the reacting inorganic amine. This in turn causes the amount of cationic NH ₄ + relative to neutral NH ₃ to increase due to the protonation equilibrium of aqueous ammonia (even if the initial inorganic amine reactant is a haloamine, since the haloamine converts initially to ammonia). Since the pKa of ammonia is 9.3 (at 25°C), the ammonia will be primarily cationic NH ₄ + at neutral pH values. Thus, as the pH approaches pH _pzc , the photocatalytic oxidation reaction is impeded by repulsion between the photocatalytic surface and the cationic ammonia reactant. For this reason, it has previously been necessary to perform photocatalytic oxidation of ammonia at high pH values (>10) if significant conversions are to be obtained. [93] By using a phosphorous-based species to modify the pH _pzc of the titanium dioxide photocatalyst, it is believed that the photocatalytic surface remains negatively charged at neutral and even mildly acidic pH values. Thus, the inorganic amine can be photocatalytically oxidised at significant rates across a wider range of pH values. [94] The phosphorous-based species may in principle be any phosphorous- based species capable of enhancing the photocatalytic oxidation rate of inorganic amine at neutral pH values, according to the principles disclosed herein. Thus, in embodiments, the phosphorous-based species reduces the point of zero charge of the titanium dioxide photocatalyst in aqueous media. The reduction in the point of zero charge as a result of the phosphorous-based species can be measured through titration or zeta potential (ζ), for example as disclosed by Chen et al, Res. Chem. Intermed. 2003, 29, 733-748 or Kormann et al, J. Phys. Chem.1988, 92, 5196-5201. [95] In some embodiments, the phosphorous-based species is selected from the group consisting of phosphorous oxoacids, and salts and esters thereof. As used herein, a phosphorous oxoacid is a phosphorous compound wherein the phosphorous is bonded to an oxygen which is bonded to hydrogen, and which produces a conjugate base by deprotonation of the oxygen. Examples of phosphorous oxoacids include phosphoric acids, hypophosphoric acids, phosphonic acids, and the like. [96] In some embodiments, the phosphorous-based species is selected from the group consisting of phosphoric acid, phosphonic acid and salts and esters thereof. The phosphoric acid may be orthophosphoric acid (H ₃ PO ₄ ). [97] Orthophosphoric acid exists in equilibrium with its phosphate salts – dihydrogen phosphate (H ₂ PO ₄ -), hydrogen phosphate (HPO ₄ ^2- ) and orthophosphate (PO ₄ ^3- ) – in aqueous solution, with the abundance of each species dependent on the pH. At mildly acidic, neutral and mildly basic pH values, H ² PO ^4- and HPO ^42- species are expected to predominate. Thus, even if phosphoric acid is added to the aqueous solution which contains the inorganic amine, phosphate will form in the solution. In some embodiments, the at least one phosphorous-based species which contacts the titanium dioxide photocatalyst comprises a phosphate. [98] Phosphates are considered particularly advantageous in embodiments of the methods disclosed herein because they are important nutrients in many fertilizer applications. In such applications, their presence therefore serves the dual function of promoting the oxidation of inorganic amine to nitrate and contributing to the fertilization of a crop. A secondary advantage of phosphate is that the pH may be buffered in the range of 5.8 to 8.0 when H ₂ PO ₄ - and HPO ₄ ^2- are present in solution, allowing enhanced conversion of the inorganic amine to be achieved before the point of zero charge is reached. However, it has been shown by experiment that the phosphate or other phosphorous-based species plays an additional role, beyond buffering, in that the rate of photooxidation is enhanced at neutral and weakly acidic conditions pH values where reaction is suppressed in the absence of a phosphorous-based species. [99] Esters of phosphorous oxoacids, including esters of phosphoric acid, may also be used as the phosphorous-based species. Sugar esters of phosphoric acids are useful fertilizer components and may be used in the methods disclosed herein as both photooxidation promotor and fertilizer nutrient. [100] The phosphorous-based species may be used in any amount suitable to promote the photocatalytic oxidation of the inorganic amine. In some embodiments, as already disclosed herein, the phosphorous-based species is dissolved in the aqueous solution and the amount of phosphorous-based species may then be selected based on either or both of the following imperatives: (1) to minimise the amount of promotor needed to achieve a desirable conversion of inorganic amine to nitrate, and (2) to achieve a desirable N:P ratio in the aqueous solution for the target application. Irradiating the aqueous solution [101] The methods disclosed herein involve a step of irradiating the aqueous solution with light, thereby photocatalytically oxidising at least a portion of the inorganic amine to nitrate. [102] The light may be any light including at least a fraction of photons with an energy greater than or equal to the bandgap of the titanium dioxide photocatalyst (≥ 3.0 eV, equal to 400nm wavelength). In some embodiments, the light is sunlight, for example concentrated sunlight. This advantageously provides the opportunity to reduce the input of external energy into the process. In other embodiments, the light is UV light. [103] The titanium dioxide photocatalyst may be dispersed in the aqueous solution during the irradiation, for example by mixing. Alternatively, the TiO ₂ may be coated or printed onto a solid material, such as a glass or plastic material; this allows facile separation of the TiO ₂ from the aqueous solution containing the products of the photooxidation. [104] One mode of operation of the photocatalytic process involves a batch photoreactor cell where the cell itself is sufficiently transparent to transmit the desired wavelength of light, around 360 nm to 400 nm into the cell. Alternatively, a window of transparent material is arranged in an otherwise opaque cell to receive and transmit the light. [105] A coated form of the catalyst may be used in a flow-through version of the photo-reactor cell, in which the aqueous solution containing the ammonia or chloramine is passed continuously through the cell. Such an apparatus may involve a larger tank of the solution to be reacted and the solution is recirculated through the photo-reactor cell until a desired extent of reaction is achieved. [106] The aqueous solution has a suitable pH during the irradiation to allow the photocatalytic oxidation to take place. The pH may decline during the reaction from a high initial pH as protons are released. This will occur, for example, when the oxidation process is conducted as a batch reaction. Advantageously, the phosphorous-based species allows the reaction to proceed at higher rates, to higher conversions and/or to lower final pH values than in the absence of the phosphorous-based species. This may be desirable because it avoids or minimises the need to add a mineral or other base to maintain an elevated pH. Moreover, for certain applications, such as fertilizer solutions, it is desirably that the nitrate-containing product solution is only weakly alkaline, neutral or even weakly acidic. [107] Thus, in some embodiments, at least a portion of the inorganic amine, such as at least 15 mol% or at least 30 mol%, is oxidised to nitrate at a pH of below 8.5, or below 8, such as below 7.5. In such pH ranges, the photooxidation rate may be higher than the rate in the absence of the phosphorous-based species and/or the selectivity to nitrate may be higher. Preferably, the portion of the inorganic amine oxidised to nitrate at a pH of below 8.5 (or below 8, such as below 7.5) is a final portion of the inorganic amine that is oxidised, such that the pH of the aqueous solution is below 8.5 (or below 8, such as below 7.5) when the oxidation reaction is terminated. A low pH at the end of the oxidation reaction may advantageously ensure that nitrite is eliminated or minimised in the aqueous solution. [108] In some embodiments, at least a portion of the inorganic amine is oxidised to nitrate at a pH of below 7, or below 6.5, such as below 6. Again, this portion is preferably the final portion of the inorganic amine that is oxidised. In such pH ranges, the photooxidation reaction may proceed when the photocatalytic oxidation rate would be negligible or zero in the absence of the phosphorous-based species. It has been observed by experiment that photooxidation of ammonia continues to at least pH 4.1, with the rate slowing down but not yet stopping at this acidic condition. [109] In some embodiments, the pH remains substantially constant during the photocatalytic oxidation reaction. This may be the case, for example, when the oxidation process is conducted as a continuous reaction, i.e. with continuous addition of reactant and withdrawal of product from the reactor. In continuous processes the phosphorous-based species may also allow the reaction to proceed at higher rates, to higher conversions and/or at lower pH values than would be the case in the absence of the phosphorous-based species. [110] The use of the phosphorous-based species may minimise or avoid the need for a mineral base to produce high pH values in the aqueous solution. In some embodiments, therefore, the aqueous solution is substantially free of alkali and alkali earth metal cations after oxidising the inorganic amine to nitrate. [111] In other embodiments, a base, such as a mineral base, may be added to the aqueous solution to elevate the pH, either initially or during the course of the photooxidation reaction. An advantage of the methods disclosed herein is that the amount of base needed to achieve a desired oxidation rate or conversion may be less than would be the case in the absence of the phosphorous-based species. Thus, in some embodiments, the molar ratio of N:M in the aqueous solution is greater than 1:1, or greater than 2:1, such as greater than 5:1, after oxidising the inorganic amine to nitrate, where M is the total amount of alkali and alkali earth metal cations. [112] The addition of base, and particularly potassium hydroxide (KOH), may be desirable in some applications. Potassium is the third main macronutrient for plants (together with nitrogen and potassium). Thus, the addition of potassium via KOH may advantageously allow a fertilizer solution with a desired ratio of N:P:K to be produced. The use of a phosphorous-based species, such as phosphate, to promote the photocatalytic oxidation reaction, provides the opportunity to add KOH in suitable amounts to produce nitrate-containing fertilizer solutions having a very wide range of N:P:K ratios. Magnesium and calcium are other useful fertilizer components for some applications, and these can also be added via appropriate bases such as magnesium hydroxide or calcium hydroxide. [113] The photocatalytic oxidation may proceed to a desired conversion of the inorganic amine. For example, at least 30 mol%, or at least 40 mol%, such as 40 to 60 mol%, of the inorganic amine may be oxidised to nitrate. Although very high conversions may be targeted, for some applications a conversion of about 40 to 60 mol% inorganic amine to nitrate may be preferred because the product is then essentially ammonium nitrate, a common form of nitrogen in fertilizers. [114] The photocatalytic oxidation is preferably highly selective to nitrate. In particular, it is preferred that there is little or no nitrite (NO ₂ -) present after the oxidation reaction reaches the target conversion. Nitrite is believed to be an intermediate in the photooxidation of ammonia to nitrate. Thus, achieving a low nitrite concentration in the product solution, particularly when the inorganic amine reactant is incompletely converted, requires that oxidation of nitrite to nitrate is strongly kinetically favoured over the initial step of oxidising the inorganic amine to nitrite. The nitrite oxidation reaction is inhibited at high pH values, so that higher nitrite concentrations can be expected when the photocatalytic oxidation reaction is conducted at higher pH values. Therefore, a further advantage of using the phosphorous-based species is that nitrite formation can be suppressed by conducting all or part of the oxidation reaction at lower pH values. In some embodiments, less than 2 %, preferably less than 1 % of the total N in the aqueous solution is present as nitrite after oxidising the inorganic amine to nitrate. In some embodiments, the aqueous solution is substantially free of nitrite after oxidising the inorganic amine to nitrate, for example when 40 to 60 mol% of the inorganic amine is oxidised to nitrate. [115] Preferably, no more than a small proportion of the inorganic amine is converted to gaseous by-products. In some embodiments, no more than 20% of the inorganic amine, such as no more than 10% of the inorganic amine, is oxidised to gases selected from N ₂ , N ₂ O and NOx. [116] The aqueous solution may comprise dissolved dioxygen (O ₂ ) when irradiated, for example because the aqueous solution is in contact with air. It has been found by experiment that the rate of oxidation is elevated in the presence of oxygen compared to an inert atmosphere. This effect is ascribed to the better ability of O ₂ , compared to water, to be photocatalytically reduced by accepting a valence band electron from TiO ₂ . Fertilizer solutions [117] The methods disclosed herein may be particularly useful for producing aqueous fertilizer solutions, for example for use in hydroponic horticulture. As disclosed herein, the methods may be used to produce aqueous solutions with a wide range of N:P ratios, and N:P:K ratios, as required for optimum fertilization of various crops. [118] In some embodiments, the aqueous solution, after irradiating the aqueous solution with light and oxidising the at least a portion of the inorganic amine to nitrate, is thus a fertilizer solution comprising: total nitrogen in a concentration range of 1 – 200 mM, nitrate in a concentration range of 1 – 200 mM, phosphate in a range of 0.5 – 20 mM, and optionally metal cations selected from K ⁺ , Ca ²⁺ and Mg ²⁺ in a total concentration range of 0.1 – 200 mM. Typically, the higher nitrate content along with the metal cations is preferred for fruiting plants, while higher total N is preferred for green leaf crops. [119] The present invention thus also provides a method of fertilizing a crop, comprising contacting an aqueous solution comprising at least one inorganic amine selected from ammonia and haloamine with a titanium dioxide photocatalyst; irradiating the aqueous solution with light, thereby photocatalytically oxidising at least a portion of the inorganic amine to nitrate, wherein the titanium dioxide photocatalyst is contacted with at least one phosphorous-based species before or during the oxidising; and applying the aqueous solution containing nitrate to a crop as a fertilizer. [120] In this method, the phosphorous-based species is preferably in a chemical form which facilitates its uptake by plants as a phosphorous macronutrient. For example, the phosphorous-based species may comprise a phosphate. EXAMPLES [121] The present invention is described with reference to the following examples. It is to be understood that the examples are illustrative of and not limiting to the invention described herein. Materials. [122] TiO ₂ powder (P25: ca.85% anatase and 15% rutile, average size ≈ 20 nm) was purchased from Degussa Co., Ltd. (Germany). Sodium citrate, sodium hypochlorite, ammonium chloride, ammonium hydroxide solution (28 wt%), ammonium sulfate, sodium hydroxide, sodium nitroprusside, sodium salicylate, boric acid, sodium tetraborate (borax), monosodium phosphate, disodium phosphate, sodium hydroxide, potassium hydroxide, hydrochloric acid (32%) were of analytical grade and supplied by Sigma-Aldrich. Ultrapure water from the Sartorius Aurium Comfort system was used throughout the study (>18 MΩ·cm). [123] Solutions of monochloramine were freshly prepared before each experiment by dropwise addition of sodium hypochlorite into well-stirred ammonium chloride solution which was adjusted to pH 8 at N:Cl ₂ = 1:1 (molar ratio). The molarity of N is the molarity of ammonium cations, while molarity of Cl ₂ (free chlorine in NaOCl) was determined by the widely available DPD-titration method. Photooxidation procedure. [124] Ammonia/monochloramine oxidative reactions were conducted in cylindrical borosilicate test tubes (diameter = 25 mm, height = 150 mm) sealed with rubber septa at ambient temperature and pressure. A 150 W Xenon lamp (UXL-150SO, Ushio Inc., Japan) was operated in a commercial Arc lamp housing (67005, Newport, USA) with an AM 1.5G filter to provide simulated solar light, while the illumination intensity was monitored at the cell position by a Model 10.0 Global Solar Power Meter from Solarmeter® (USA) to maintain 1 SUN (100 mW·cm ^-2 ) intensity. The photoreactor cells were placed in a black painted steel box to eliminate uncontrolled light leakage. The cells were cooled by air ventilation with an intake port on the side of the black box and an exhaust fan on the top of the box. Before use, all parts of the photoreaction cell were soaked and sonicated in alkaline Extran MA05 detergent solution (5 wt%) and water for at least 1 h in each medium, and then washed copiously with water to eliminate any ammonia and other contaminants. Finally, the cell was dried by blowing with compressed pure dinitrogen. Unless indicated otherwise, 0.5 g/l of TiO ₂ photocatalyst was used in the experiments. The TiO ₂ was dispersed by ultrasonication for 10 min in the dark to ensure homogeneity, and the mixture was stirred continuously during the subsequent photooxidation. The septum seal was kept above the irradiation area and wrapped in aluminium foil to protect it from degradation by UV light. Quantification of ammonia (NH ₄ +), monochloramine (NH ₂ Cl), nitrite (NO ₂ -), and nitrate (NO ₃ -) [125] The amount of nitrite and nitrate was monitored by the Griess method (García-Robledo et al, Marine Chemistry 2014, 162, 30-36). Limit of detection of both nitrate and nitrite by this method is estimated to be 0.5 μM. Ammonia was quantified by the indophenol method. In this method, 250 µL of salicylate/NaOH/citrate (10 wt%, 0.125 M and 5 wt% respectively) solution containing 0.04 wt% sodium nitroprusside (Na ₂ [Fe(CN) ₅ NO]) and 250 μL of hypochlorite solution NaClO (0.06%) containing 0.125 M NaOH were added into the obtained 1 mL of the aforementioned centrifuged aliquot. The mixed solution was stored in the dark for 2 h before examination by spectrophotometry in 500-1000 nm spectrum. The adsorption peak of indophenol, at around 655 nm, was recorded and used to calculate the ammonia yield. The concentration-absorbance curve was calibrated using standard NH ₄ Cl solutions with a series of concentrations (y = 0.0109x + 0.0081, R ² = 0.9991). [126] Monochloroamine (NH ₂ Cl) was quantified by a method based on the same indophenol method. In the classic Berthelot/indophenol reaction, ammonia reacts with chlorine to form NH ₂ Cl in the first step. Thus, NH ₂ Cl was quantitatively analyzed, without interference of ammonia, by the same procedure used in the indophenol method (see above) except that the hypochlorite reagent was omitted. The calibration data obtained for NH ₂ Cl quantification by the modified method gave the calibration equation y = 0.0075x + 0.013, R ² = 0.9998. The standard indophenol method detects both ammonia and monochloroamine quantitatively, so that mixtures of ammonia and monochloroamine can be quantified using both analysis methods, with the amount of ammonia being the difference in detected concentration between the two methods. Example 1. Heat treatment of TiO ₂ [127] P25 TiO ₂ powder was pre-purged for 30 min, and then heat-treated under either pure argon (Ar) gas or dry air (c.a.10 ml/min) in a tube furnace to 600°C for 3 h, with a ramping rate of 5 °C/min, then cooled to room temperature. The heat-treated TiO ₂ was washed with a strong alkaline NaOH solution, then with H ₂ SO ₄ (0.05 M) and ultrapure water several times until the washed solution reached neutral pH. The samples were then separated by centrifuge and dried at 80°C in a vacuum oven overnight. The treated titania powder was kept in ultrapure water before use. [128] The crystal structure of the untreated TiO ₂ , TiO ₂ heat-treated in Ar, and TiO ₂ heat-treated in air was investigated with X-ray diffraction (XRD). All the diffraction peaks match well with either the anatase phase (COD No.96-900-9087) or the rutile phase (COD No.96-900-7433), with no residual peaks for different phases seen. The amount of each phase was calculated from Rietveld refinement (Fullprof suite, version Sep-20) of the diffraction data. The Brunauer-Emmett-Teller (BET) surface area and Barrett-Joyner-Halenda (BJH) pore size were determined using the N ₂ adsorption- desorption isotherm at 77 K on a Micrometrics Tristar II 3020, and the samples were also analysed by scanning electron microscopy (SEM). The results of the analyses are shown in Table 1. Table 1. [129] The heat-treatment induced a partial phase transformation of anatase to rutile, with a greater extent of transformation obtained during the treatment in air than in argon. Only minor effects on the BET surface area and BJH pore size were evident in the TiO ₂ heat-treated in Ar, and no obvious difference in morphology could be detected by SEM. Example 2. Effect of TiO ₂ heat-treatment in Ar on monochloroamine photooxidation [130] Photooxidation of monochloroamine (NH ₂ Cl) was conducted using both the untreated TiO ₂ and TiO ₂ heat-treated in Ar (Example 1) as the photocatalyst. The conversion activity and the selectivity of products is seen in Figure 1 (untreated TiO ₂ ) and Figure 2 (TiO ₂ heat-treated in Ar), where C is the concentration and C0 is the initial concentration of NH ₂ Cl. Rapid, complete degradation of NH ₂ Cl to NH ₃ /NH ₄ + was observed within minutes in both cases. A comparative control experiment without the presence of photocatalyst confirmed that TiO ₂ has photocatalytic activity towards the oxidation/degradation of NH ₂ Cl under visible light. However, the TiO ₂ heat-treated in Ar was notably more active than untreated TiO ₂ : the pseudo-first-order rate of the photodecomposition of NH ₂ Cl was 4.9 ×10 ^-3 s ^-1 vs.1.8×10 ^-3 s ^-1 , respectively. Moreover, the formation of the target oxidation product, NO ₃ -, was obtained sooner and at higher yield for TiO ₂ heat-treated in Ar than untreated TiO ₂ : a conversion of 23.2% vs.10.6% was obtained, respectively, after the 2 h experiment. [131] An increase in total concentration of nitrogen species (total_N in Figure 1) was seen after the first hour in the reaction using untreated TiO ₂ . This was ascribed to contamination of the untreated TiO ₂ by surface-absorbed NOx and ammonia species in the laboratory environment. A similar effect was not seen with the TiO ₂ heat-treated in Ar (Figure 2) because the adsorbed gaseous contaminants were removed by the hot flowing gas during the heat-treatment, and reabsorption was avoided by storing the photocatalyst in ultrapure water. Example 3 (comparative). TiO ₂ -catalyzed photooxidation of ammonia [132] To investigate the effect of pH, photooxidation experiments were conducted using TiO ₂ heat-treated in Ar (Example 1) as the photocatalyst and different sources of ammonia in aqueous solution (100 µM initial ammonium concentration). Thus, experiments were conducted with aqueous ammonium chloride (NH ₄ Cl; initial pH 5.88), aqueous ammonium sulfate ([NH ₄ ] ₂ SO ₄ ; initial pH 5.57) and aqueous ammonium hydroxide (NH ₄ OH; initial pH 9.47). The results are shown in Figure 3 (NH ₄ + conversion) and Figure 4 (NO ₃ - formation), where C is the concentration and C0 is the initial concentration of NH ₄ +. [133] Negligible conversion was obtained after 4 hours at initial pH values of less than 6, which is below the point of zero charge (pHpsc) for TiO ₂ (expected to be about 6.2-7.0). Only limited conversion was obtained at an initial pH of 9.5, with most of the observed conversion taking place in the first 30 minutes of the four hour reaction. Ammonia conversion took place until the pH dropped to approximately 7.0, then became negligible. The drop in pH is ascribed to the release of protons occurring during the photooxidation process, as indicated by equations (5)-(9): TiO ₂ + hν → e _cb− + h _vb + (5) NH _3aq → NH _3ad (6) NH _3ad + h _vb + → ^• NH ₂ + H ⁺ (7) NH ₃ + 2H ₂ O + 6 h _vb ⁺ → NO ₂ - + 7H ⁺ (8) NO ₂ - + H ₂ O + 2 h _vb ⁺ → NO ₃ - + 2H ⁺ (9) [134] This experiment shows that TiO ₂ -catalysed photooxidation of ammonia in aqueous solution is inhibited at pH values of below about 8.5, and ceases at about pH 7.0. Example 4. TiO ₂ -catalyzed photooxidation of ammonia in the presence of phosphate [135] Photooxidation experiments were then performed using a phosphate buffer at pH 8, with the hypothesis that the buffer would maintain a sufficiently alkaline pH during ongoing photooxidation to allow higher ammonia conversions to be obtained. In initial experiments, aqueous 100 µM ammonia solutions (NH ₄ OH or NH ₄ Cl) containing 1 mM phosphate (0.06 mM of H ₂ PO ₄ - and 0.94 mM HPO ₂ ^2- , diluted from 50mM stock phosphate buffer solution (PBS) which minorly adjusted to pH 8 by addition of NaOH) were prepared. The initial pH of both solutions was 8.0. The photooxidation results (TiO ₂ heat-treated in Ar; 4 hour duration) are shown in Figure 3 (NH ₄ + conversion), Figure 4 (NO ₃ - formation), and Table 2. The final pH of the solution was not measured, but it is expected that the pH remained about 8 throughout the reaction due to the buffer capacity provided by the excess phosphate. Table 2. ^b [NO ₂ -] _t /[NH ₃ ] _T,0 x 100; dash in table indicates below Limit of Detection (0.5 μM) ^c [NO ₃ -] _t /[NH ₃ ] _T,0 x 100. ^d ([NO ₂ -] _t + [NO ₃ -] _t + [NH ₃ ] _T,t )/[NH ₃ ] _T,0 x 100. [136] The addition of the phosphate thus provided a dramatic improvement in the conversion of ammonia and formation of nitrate: over 95% conversion was obtained in four hours of reaction (c.f. Example 3). Moreover, the reaction was highly selective to nitrate, with no detectable nitrite present in the final solution. This was ascribed in part to the weakly alkaline conditions, as the nitrite intermediate is more likely to remain unoxidized at higher pH values. [137] Similar results (conversion rate and selectivity) were obtained when the concentration of phosphate was increased to 4 mM (see Figure 5; Table 2). There is no evident adverse effect due to increased ionic strength or competitive interactions between ammonium and other cations on the photocatalyst surface. Example 5 (comparative). catalyzed photooxidation of ammonia in the presence of borate [138] Photooxidation experiments were then conducted using a borate buffer at pH 9, with the hypothesis that the borate would have a similar effect to phosphate (c.f. Example 4). An aqueous 100 µM ammonia (NH ₄ OH) solution containing 4 mM borate (2 mM of boric acid and 2mM of sodium tetraborate, diluted from stock 100 mM borate buffer solution (BBS) which was minorly adjusted to pH 9 by addition of NaOH or HCl) was thus prepared. The initial pH of the solution was 9.0. The photooxidation result (TiO ₂ heat-treated in Ar; 4 hour duration) is shown in Figure 5 (NH ₄ + conversion), Figure 6 (NO ₃ - formation), and Table 3, with comparison against the photooxidation result with phosphate. Table 3. ^b [NO ₂ -] _t /[NH ₃ ] _T,0 x 100; dash in table indicates below Limit of Detection (0.5 μM). ^c [NO ₃ -] _t /[NH ₃ ] _T,0 x 100. ^d ([NO ₂ -] _t + [NO ₃ -] _t + [NH ₃ ] _T,t )/[NH ₃ ] _T,0 x 100. [139] The addition of the borate also improved the conversion of ammonia and formation of nitrate, relative to the experiments without an additive (c.f. Example 2). However, the initial rate decreased by a factor of 3 relative to the counterpart experiment with phosphate, from about 10.4 10 ^-7 M·min ^-1 to 3.6 × 10 ^-7 M·min ^-1 . Furthermore, the nitrite oxidation intermediate was present in significant quantities during the course of the reaction (> 10% selectivity after 1-2 hours), as seen in Figure 6. This is ascribed in part to the inhibition of nitrite oxidation at higher pH values. [140] A similar experiment was conducted but with the borate buffer solution adjusted to pH 8 (close to the edge of the buffer capacity). In this case, little photooxidation was observed over 4 hours of reaction. It is proposed that the initial oxidation rapidly reduced the pH to the point of zero charge (pHpsc) for TiO ₂ (expected to be about 6.2-7.0), after which no further reaction took place. These experiments indicate that the borate additive only promotes photooxidation as a buffer by maintaining the pH above the (unmodified) pHpsc for TiO ₂ . Example 6. catalyzed photooxidation of higher concentration ammonia in the presence of phosphate and borate [141] The effect of the phosphate and borate additives for TiO ₂ photocatalyzed ammonia oxidation was further investigated at higher (x10) initial ammonia concentrations (higher N:P and N:B ratios). Aqueous 1 mM ammonia (NH ₄ OH) solutions containing 4 mM phosphate or borate were thus prepared by the addition of PBS or BBS, with pH values of 8.0 and 9.0 respectively. The photooxidation results (TiO ₂ heat-treated in Ar; 4 hour duration) are shown in Figure 7 (phosphate), Figure 8 (borate), and Table 4. Table 4. ^b [NO ₂ -] _t /[NH ₃ ] _T,0 x 100; dash in table indicates below Limit of Detection (0.5 μM). ^c [NO ₃ -] _t /[NH ₃ ] _T,0 x 100. ^d ([NO ₂ -] _t + [NO ₃ -] _t + [NH ₃ ] _T,t )/[NH ₃ ] _T,0 x 100. [142] When using phosphate in the solution, ammonium conversions of about 50% were obtained, i.e. at the target for converting ammonia to ammonium nitrate. Moreover, high selectivity to nitrate was observed, with no nitrite being detectable. However, when borate was used instead, only 30% ammonia conversion was obtained and the selectivity to nitrate was poor (NO ₃ -:NO ₂ - < 3:1). Example 7. TiO ₂ -catalyzed photooxidation of monochloroamine in the presence of phosphate [143] TiO ₂ -catalyzed photooxidation of monochloroamine was then investigated using a phosphate additive. An aqueous 100 µM NH ₂ Cl solution containing 1 mM phosphate was prepared by the addition of PBS. The initial pH of the solution was 8.0. The photooxidation results (TiO ₂ heat-treated in Ar; 4 hour duration) is shown in Figure 9 and Table 5. Excellent conversion and nitrate selectivity was obtained, similar to the phosphate-mediated photooxidation of ammonia. The difference in photocatalytic performance between untreated TiO ₂ and TiO ₂ heat-treated in Ar (Example 1) was also evident in shorter (2h) experiments (Table 5). Table 5. ^b [NO ₂ -] _t /[NH ₃ ] _T,0 x 100; dash in table indicates below Limit of Detection (0.5 μM). ^c [NO ₃ -] _t /[NH ₃ ] _T,0 x 100. ^d ([NO ₂ -] _t + [NO ₃ -] _t + [NH ₃ ] _T,t )/[NH ₃ ] _T,0 x 100. Example 8. Effect of TiO ₂ heat-treatment on ammonia photooxidation in the presence of phosphate [144] The effect of the TiO ₂ photocatalyst pre-treatment on ammonia oxidation under phosphate-mediated photooxidation conditions was investigated using the untreated TiO ₂ , TiO ₂ heat-treated in Ar, and TiO ₂ heat-treated in air (Example 1). Aqueous solutions with different ammonia (NH ₄ OH) concentrations in the range of 0.1 mM to 1.1 mM, also containing 1 mM phosphate by the addition of PBS, were thus prepared. The initial pH of the solutions was 8.0. The solutions were subjected to photooxidation in the presence of the three different TiO ₂ photocatalyst samples (4-8 hour duration). Comparative results are shown in Figure 10 (initial rate, as measured over the first half-life) and Figure 11 (first order rate constant). It is evident that the best results were obtained for the TiO ₂ that was heat-treated in argon. Heat-treatment to a similar temperature in air degraded the photocatalytic performance relative to the untreated catalyst. [145] Another aqueous solution with 1 mM ammonia concentration and 4 mM phosphate concentration, initial pH 8.0, was subjected to photooxidation in the presence of the three different TiO ₂ catalyst samples (8 hour duration). Comparative selectivity results are shown in Figure 12 and Table 6. Photooxidation in the presence of all TiO ₂ samples gave good conversion of ammonia to nitrate. However, the TiO ₂ which was heat-treated in Ar gave the best conversion and selectivity: no nitrite was detected in any of the samples (the “lost” ammonia is believed to correspond to N ₂ formation). By contrast, the TiO ₂ heat-treated in air gave the lowest conversion, and the untreated TiO ₂ gave the highest losses of starting ammonia. A TiO ₂ catalyst which was heat treated in the presence of Ar sufficiently only to effect a minor change in the phase composition (“under-treated in Ar TiO ₂ ” in Table 6) showed a photocatalytic performance and product selectivity analogous to the untreated TiO ₂ . Table 6. Starting composition: NH4 ⁺ 1 mM, PBS 4 mM (pH 8) ^b [NO ₂ -] _t /[NH ₃ ] _T,0 x 100; dash in table indicates below Limit of Detection (0.5 μM). ^c [NO ₃ -] _t /[NH ₃ ] _T,0 x 100. ^d ([NO ₂ -] _t + [NO ₃ -] _t + [NH ₃ ] _T,t )/[NH ₃ ] _T,0 x 100. Example 9. Effect of the gas atmosphere [146] The effect of the gas atmosphere during ammonia photooxidation in the presence of phosphate was investigated by performing comparative photooxidation experiments (100 µM ammonia (NH ₄ OH) concentration, 1 mM phosphate concentration, initial pH 8.0, TiO ₂ heat-treated in Ar, 4 hour duration) under (i) a static air atmosphere (as in previous Examples) vs (ii) under an atmosphere of flowing argon gas (4.2mL.min ^-1 ). The results are shown in Figure 13 (NH ₄ + conversion) and Figure 14 (NO ₃ - formation). The reaction rate is higher in the presence of O ₂ , which can be ascribed to the better ability of O ₂ , compared to water, to be photocatalytically reduced by accepting a valence band electron from TiO ₂ . Example 10. TiO ₂ -catalyzed photooxidation of ammonia in the presence of phosphate to produce solutions suitable as fertilizers [147] An experiment was performed to determine if TiO ₂ -catalyzed photooxidation of ammonia in the presence of phosphate could be used to produce solutions with commercially relevant concentrations of nitrate, phosphate (and optionally potassium), for use as fertilizers. Relevant targets were: • Ammonia conversion to ammonium nitrate (preferably c.a.50% conversion) • Nitrogen concentrations of at least 1 mM and preferably more than 5 mM • N:P ratios of between 5:1 and 20:1, preferably around 10:1 • Near-zero nitrite concentration • Potassium levels variable, depending on the fertilizer application. [148] An aqueous solution containing 10 mM ammonium (added as NH ₄ OH) and 1 mM phosphate (added as phosphoric acid; H ₃ PO ₄ ) was prepared. The initial pH of the solution was about 9.9. The solution was then pre-purged with O ₂ gas (2 mL·min ^-1 ) for 40 min prior to photooxidation (under static O ₂ atmosphere) for 24 hours in the presence of the TiO ₂ heat-treated in Ar (per Example 1). The results are shown in Figure 15. After 20 hours, the ammonia conversion was about 40%, with only nitrate present in solution, and the pH had dropped to about 6.3. [149] The solution was then spiked with potassium hydroxide (KOH) such that the N:P:K ratio in solution was 10:1:3, resulting in an increase of the pH to about 8.3. Photooxidation was then continued under the same conditions as before for another 6 hours (26 hours total). Analyses conducted over the final three hours confirmed that the ammonia oxidation reaction had reached an end point at a pH of about 7.5, with final ammonia concentration of about 3.9 mmol (62% conversion), final nitrate concentration of about 5.3 mM and undetectable nitrite. [150] This experiment demonstrates that the phosphate additive enhances ammonia conversion even when present in sub-stoichiometric quantities relative to ammonia (N:P ratio = 10:1). Furthermore, ammonia conversion continues after the pH drops to values (<6.5) where further conversion is inhibited in the absence of the phosphate. The conversion can be further enhanced by the addition of KOH in relatively small amounts, as required to give N:P:K ratios in a range useful for fertilizers. Example 11: TiO ₂ -catalyzed photooxidation of ammonia by monochromic light UV 365nm in the presence of phosphate [151] TiO ₂ -catalyzed photooxidation of ammonia was investigated using a single monochromic LED diode. The LED diode emits UV light with wavelength 365nm at radiated power of 13.2 mW. Following the method of Example 4, an aqueous 100 µM NH ₄ OH solution containing 1 mM phosphate (as PBS buffer) was mixed with TiO ₂ heat- treated in Ar. The initial pH of the solution was 8.0. After 4 hours of irradiation by the LED, the ammonia conversion was about 37%, with only nitrate and ammonium present in solution. Example 12: TiO ₂ Coated on a solid substrate [152] To assist in separation and recycling of the TiO ₂ , the TiO ₂ prepared as described in Example 1 was combined with ethyl cellulose binder in terpinol carrier solvent to prepare a screen-printing ink. The material was then screen printed on to a flat piece of glass of dimensions 10 by 10 cm. The screen-printed film, in disc shape with area around 38.5 cm ² (d = 7 cm), was calcined to remove all organic binder without affecting the crystal structure of TiO ₂ photocatalyst. Example 13: Flow through photo conversion cell [153] The TiO ₂ coated glass prepared as in Example 12 was used as the front surface in a flow through cell of dimensions 12×12×3 cm (H×W×D). The internal depth of reactor is 1.2 cm. The whole TiO ₂ film was illuminated under 1 Sun AM 1.5G. Following the method of Example 8, an aqueous 1.5 mM NH ₄ OH solution containing 4 mM phosphate (as PBS buffer) was transferred into glass reservoir. The reactant solution (150 mL) was pumped into the reactor at a flow rate of 50 mL/min. The initial pH of the solution was 8.0. After 6 hours of irradiation, the ammonia conversion was approximately 24%. Example 14: Effect of Co-catalyst materials decorated onto TiO ₂ in ammonia photooxidation in the presence of phosphate [154] The effect of co-catalyst materials (Au, Ag, Pt, Ni) on ammonia oxidation under phosphate-mediated photooxidation conditions was investigated using the TiO ₂ heat-treated in Ar as a base photocatalyst (as described in Example 1). Deposition of co-catalysts was achieved by photoreduction of metal precursors onto the surface of the TiO ₂ , using the method reported by Penumaka et al (Scientific Reports, 2021, 11:8084). In general, H ₂ PtCl ₆ ·xH ₂ O, AgNO ₃ , and HAuCl ₄ ·3H ₂ O were used as precursors for Pt, Ag and Au co-catalyst preparation respectively. A ratio of 2 wt.% (M:TiO ₂ ) was chosen as the loading amount of the co-catalyst on the TiO ₂ base photocatalyst. Photoreduction of the salt precursor from aqueous solution to produce metallic nanoparticles occurred by external irradiation of a 500W Xe lamp under vigorous stirring and bubbling of inert N ₂ gas. The resulting cocatalysts were present as nanoparticles, with average particle size below 5 nm, decorated on the surface of the nanoparticulate TiO ₂ . This can be seen in Figure 16 which shows a transmission electron micrograph of the TiO ₂ photocatalyst decorated with Ag co-catalyst (Ag/TiO ₂ ). [155] An aqueous solution containing 1 mM ammonia concentration and 4 mM phosphate concentration by the addition of PBS was thus prepared, and the solution was subjected to photooxidation in the presence of the different decorated TiO ₂ photocatalyst samples over a period of 8 hours. Results are shown in Table 7. [156] The TiO ₂ photocatalyst decorated with Pt co-catalyst (Pt/TiO ₂ ) gave the best initial rate in a short period of time. The Ag/TiO ₂ system gave the second-best initial rate. [157] Without wishing to be limited by any theory, the observed effects may be the result of the better utilization of incident irradiation. Metal particles when in contact with semiconductors create Schottky barriers (φ ^B ) by the different work function of metal and the electron affinity of the semiconductor conduction band. This barrier may promote enhanced charge separation by efficient e- transfer between TiO ₂ and metal. Moreover, plasmonic materials (e.g. Pt, Au, Ag) could further facilitate reactions by plasmon resonance effects. The Nickel-TiO ₂ system showed no activity towards ammonia oxidation. This suggests that the effect seen with Pt, Au and Ag may be at least in part a plasmonic effect, since Ni is not known to be plasmonically active. Table 7. Starting composition: NH ₄ ⁺ 1 mM, PBS _{4 mM} (pH 8) Example 15. Ag/TiO ₂ -catalyzed photooxidation of ammonia under air atmosphere in the presence of phosphate to produce solutions suitable as fertilizers [158] An experiment was performed to determine if Ag/TiO ₂ -catalyzed photooxidation of ammonia in the presence of phosphate could be used to produce solutions with commercially relevant concentrations of nitrate, phosphate (and optionally potassium) for use as fertilizers. Relevant targets were: • Ammonia conversion to ammonium nitrate (preferably c.a.50% conversion) • Nitrogen concentrations of at least 1 mM and preferably more than 5 mM • Near-zero nitrite concentration • Potassium levels variable, depending on the fertilizer application. [159] An aqueous solution containing ~10 mM ammonium (9.1 mM, added as NH ₄ OH) and 4 mM phosphate (added as PBS) was prepared. The initial pH of the solution was about 10.1. The solution was stirred for 30 min to reach adsorption/desorption equilibrium and then connected to an air gas-bag prior to photooxidation (under a static air atmosphere) for 24 hours in the presence of the Ag/TiO ₂ photocatalyst (as per Example 14). The results are shown in Figure 16. After 12 hours, the ammonia conversion was about 55%, with 75% of the product being nitrate, and the pH had dropped to about 8.3. [160] Photooxidation was then continued under the same conditions as before for another 12 hours (24 hours total). The ammonia oxidation reaction reached an end point at a pH of about 7.3, with final ammonia concentration of about 3.0 mmol (80.5% conversion), final nitrate concentration of about 5.6 mM and undetectable nitrite. Total mass N-recovery was 96%. [161] This experiment demonstrates that the Ag/TiO ₂ photocatalyst enhances ammonia conversion, relative to an unpromoted TiO ₂ photocatalyst (c.f. Example 10). This enhancement was achieved despite the use of an air atmosphere, in contrast with the O ₂ gas atmosphere used in Example 10, thus demonstrating that commercially relevant nitrate-based fertilizer compositions can be produced by photocatalytic oxidation of ammonia solutions even without a O ₂ -rich environment. Example 16. TiO ₂ -catalyzed photooxidation of ammonia in the presence of phosphonate to produce solutions suitable as fertilizers [162] An experiment was performed to determine if TiO ₂ -catalyzed photooxidation of ammonia in the presence of phosphonate (in contrast to phosphate as used in Examples 10 and 15) could be used to produce solutions with commercially relevant concentrations of nitrate, phosphate (and optionally potassium), for use as fertilizers. Relevant targets were: • Ammonia conversion to ammonium nitrate (preferably c.a.50% conversion) • Nitrogen concentrations of at least 1 mM and preferably more than 5 mM • N:P ratios of between 5:1 and 20:1, preferably around 10:1 • Near-zero nitrite concentration • Potassium levels variable, depending on the fertilizer application. [163] An aqueous solution containing ~10 mM ammonium (9.3 mM, added as NH ₄ OH) and 1 mM phosphonate (added as phosphonic acid; H ₃ PO ₃ ) was prepared. The initial pH of the solution was about 9.7. The solution was then pre-purged with O ₂ gas (10 mL·min ^-1 ) for 30 min and then connected to an O ₂ gas-bag prior to photooxidation for 24 hours prior to photooxidation (under static O ₂ atmosphere) for 24 hours in the presence of the TiO ₂ heat-treated in Ar (per Example 1). The results are shown in Figure 17. After 24 hours, the ammonia conversion was about 49% (with final ammonia concentration of about 4.7 mmol), with only nitrate (4.4 mM) present in solution, and the pH had dropped to about 5.3. [164] This experiment demonstrates that the phosphonate additive also enhances ammonia conversion even when present in sub-stoichiometric quantities relative to ammonia (N:P ratio = 10:1). Furthermore, ammonia conversion continues after the pH drops to values (<6.5) where further conversion is inhibited in the absence of phosphorous-based species. The conversion can be further enhanced by the addition of KOH in relatively small amounts, as required to give N:P:K ratios in a range useful for fertilizers as described in Example 10. [165] Those skilled in the art will appreciate that the invention described herein is susceptible to variations and modifications other than those specifically described. It is understood that the invention includes all such variations and modifications which fall within the spirit and scope of the present invention.