A particulate photocatalyst and photocatalytic method

Abstract

The invention provides a particulate photocatalyst comprising: a plasmonic core comprising a metallic nanoparticle; a shell at least partially surrounding the plasmonic core, the shell comprising nitrogen-doped carbon; and a transition metal immobilised on the shell, wherein at least a portion of the transition metal is atomically dispersed.

Description

[0001] A particulate photocatalyst and photocatalytic method

[0002] Technical Field

[0003] [1] The invention relates to a particulate photocatalyst comprising: a plasmonic core, a shell comprising nitrogen-doped carbon which at least partially surrounds the plasmonic core, and an atomically dispersed transition metal immobilised on the shell. The invention further relates to a use of, and a method of producing, the particulate photocatalyst. The invention further relates to a photocatalytic method comprising contacting a hydrogen-bearing compound with the particulate photocatalyst and irradiating the particulate photocatalyst with light, thereby photocatalytically dehydrogenating at least a portion of the hydrogen-bearing compound with the particulate photocatalyst to form dihydrogen.

[0004] Background of Invention

[0005] [2] There is an urgent need to decarbonise the production of dihydrogen (H2) for current industrial uses, and a large opportunity for sustainably produced H2 as an energy carrier and chemical feedstock in a decarbonised world economy.

[0006] [3] Within the contemporary landscape of energy conversion technologies, photocatalysis stands out for its potential to utilize abundant resources, such as sunlight and water, to generate H2. Ideally, an untreated water source would be used as feedstock to a photocatalytic H2 production process. Seawater is particularly attractive as it is a virtually limitless resource. However, developing photocatalysts capable of providing high H2 production rates and stability in saline environments is required for the development of scalable photocatalytic systems.

[0007] [4] Some reported approaches to seawater splitting rely on the use of noble metals and / or sacrificial reagents. However, the scarcity of noble metals and the additional costs associated with sacrificial agents have limited the widespread application of such approaches. There is therefore an unmet need for low-cost, high-performance photocatalysts without noble metals and / or which can operate effectively without sacrificial reagents.

[0008] [5] While the above discussion relates specifically to water splitting, particularly of seawater, there is also a need for effective photocatalysts in other photocatalytic redox processes such as reforming of organic feedstocks, carbon dioxide reduction, organic waste oxidation or dehydrogenation, dinitrogen reduction, methane oxidation, and ammonia oxidation.

[0009] [6] There is therefore an ongoing need for photocatalysts which at least partially address one or more of the above-mentioned short-comings, or provide a useful alternative. [7] 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.

[0010] Summary of Invention

[0011] [8] It has now been discovered that seawater splitting and dehydrogenation of alcohols may be conducted with exceptionally high and sustained H2 production rates and solar-to- hydrogen efficiencies by using a particulate photocatalyst comprising atomically dispersed transition metals, such as cobalt, immobilised on nitrogen-doped carbon. The particulate photocatalyst is configured in a core-shell morphology, with the nitrogen-doped carbon present as the shell on a plasmonic nanoparticle core and the transition metal atomically dispersed on the outer surface of the shell.

[0012] [9] The nanoparticle core may be either a metallic nanoparticle or a metal oxide nanoparticle, but is preferably a metallic nanoparticle such as copper for some applications. Such nanoparticle cores have been shown to exhibit broad-spectrum light absorption with generation of plasmonic resonance, and thus to enhance the rate of photocatalytic water splitting. Without wishing to be limited by any theory, the nanoparticle cores are believed to concentrate light energy through a localized surface plasmon resonance (LSPR) effect, generating and providing hot plasmonic charge carriers to the catalytically active sites on the transition metal-decorated nitrogen-doped carbon shell. The electric field induced by the LSPR may also reduce the energy barriers to the redox reactions involved in water splitting, thereby enhancing the overall rate of H2 production.

[0013]

[0010] Advantageously, the core-shell morphology allows the plasmonic nanoparticles to enhance photocatalytic water splitting across the entire shell surface, which may be near- uniformly separated from the core by only a thin shell. This contrasts with other photocatalyst morphologies where deposition of plasmonic nanoparticles on the photocatalytic surface of a functional semiconductor blocks a portion of the active photocatalytic sites and / or restricts the plasmonic enhancement to a small interface region between nanoparticle and semiconductor. In addition, the shell may protect the metallic plasmonic core from oxidation during extended water splitting reactions, thus limiting photocatalyst degradation.

[0014]

[0011] It has further been demonstrated that the photocatalytic H2 production rate in water splitting is enhanced under conditions of elevated salinity approximating that of seawater. It is proposed, on the basis of theoretical calculations, that this beneficial effect is due to salt ions absorbing onto the photocatalyst surface, inducing charge polarization and suppressing the recombination of photo-generated charge carriers.

[0012] In accordance with a first aspect, disclosed herein is a particulate photocatalyst comprising: a plasmonic core comprising a metallic nanoparticle; a shell at least partially surrounding the plasmonic core, the shell comprising nitrogen-doped carbon; and a transition metal immobilised on the shell, wherein at least a portion of the transition metal is atomically dispersed.

[0015]

[0013] In some embodiments, the metallic nanoparticle comprises a metal selected from the group consisting of copper, aluminium, bismuth, gold, silver and alloys thereof. In some embodiments, the metallic nanoparticle comprises metallic copper.

[0016]

[0014] In some embodiments, the transition metal comprises one or more group 7 to 11 transition metals. In some embodiments, the transition metal is selected from the group consisting of cobalt, nickel, copper and combinations thereof.

[0017]

[0015] In some embodiments, the transition metal is cobalt.

[0018]

[0016] In some embodiments, at least a portion of the transition metal is co-ordinated to nitrogen atoms of the nitrogen-doped carbon.

[0019]

[0017] In some embodiments, the particulate photocatalyst comprises the transition metal in an amount of between 0.2 wt.% and 1 wt.%, such as between 0.25 wt.% and 0.6 wt.%.

[0020]

[0018] In some embodiments, the shell completely encapsulates the plasmonic core, and the shell has a thickness in the range of 1 nm to 15 nm. In some embodiments, the shell has a thickness of less than 5 nm, such as less than 3 nm, for example less than 2 nm.

[0021]

[0019] In some embodiments, the nitrogen-doped carbon comprises graphitic carbon.

[0022]

[0020] In some embodiments, the particulate photocatalyst comprises particles with a distribution of particle sizes. The particles sizes may, for example, range from less than 40 nm to greater than 80 nm.

[0023]

[0021] In accordance with a second aspect, disclosed herein is the use of the particulate photocatalyst according to the first aspect in a photocatalytic process for oxidising or reducing a reactant.

[0024]

[0022] In some embodiments, the reactant is a hydrogen-bearing compound and the hydrogen-bearing compound is photocatalytically dehydrogenated to produce dihydrogen.

[0025]

[0023] In accordance with a third aspect, disclosed herein is a method of producing a particulate photocatalyst according to any embodiment of the first aspect. The method comprises: providing a dispersion of nanoparticles comprising a metal oxide in a liquid comprising a salt of a transition metal and a nitrogen-bearing organic compound; and thermally treating the dispersion at a temperature sufficient to (i) decompose at least a portion of the nitrogen-bearing organic compound, thereby producing nitrogen-doped carbon in shells which at least partially surround the nanoparticles, (ii) reduce the metal oxide to form metallic nanoparticles, and (iii) immobilise at least a portion of the transition metal, as atomically dispersed transition metal, on the shells.

[0026]

[0024] In some embodiments, the metal oxide is copper oxide.

[0027]

[0025] In some embodiments, the nitrogen-bearing organic compound comprises at least one selected from an amide, an aromatic amine and a N-heteroaromatic. In some embodiments, the nitrogen-bearing organic compound comprises formamide.

[0028]

[0026] In some embodiments, the method comprises acid treating the particulate photocatalyst, for example with nitric acid. In some embodiments, the method comprises thermally annealing the particulate photocatalyst under an inert atmosphere, such as under argon.

[0029]

[0027] In accordance with a fourth aspect, disclosed herein is a photocatalytic method comprising: contacting a composition comprising a hydrogen-bearing compound with a particulate photocatalyst; and irradiating the particulate photocatalyst with light, thereby photocatalytically dehydrogenating at least a portion of the hydrogen-bearing compound with the particulate photocatalyst to form dihydrogen, wherein the particulate photocatalyst comprises: a plasmonic core; a shell at least partially surrounding the plasmonic core, the shell comprising nitrogen-doped carbon; and a transition metal immobilised on the shell, wherein at least a portion of the transition metal is atomically dispersed.

[0030]

[0028] In some embodiments, the plasmonic core comprises a metallic or metal oxide nanoparticle.

[0031]

[0029] In some embodiments, the plasmonic core comprises a metallic nanoparticle. In some embodiments, the metallic nanoparticle comprises metallic copper.

[0032]

[0030] In other embodiments, the plasmonic core comprises a metal oxide, such as an oxygen-deficient metal oxide. Suitably, the metal oxide may be a titanium oxide.

[0033]

[0031] In some embodiments, the hydrogen-bearing compound is water.

[0034]

[0032] In some embodiments, the composition comprising a hydrogen-bearing compound is an aqueous solution comprising NaCI, wherein water is photocatalytically dehydrogenated to produce dihydrogen. In some embodiments, the composition is seawater.

[0035]

[0033] In some embodiments, the composition comprising a hydrogen-bearing compound is an aqueous composition comprising one or more organic compounds, wherein at least one organic compound is photocatalytically dehydrogenated to produce dihydrogen. In some such embodiments, the composition is wastewater.

[0034] In some embodiments, the transition metal comprises one or more group 7 to 11 transition metals. In some embodiments, the transition metal is selected from the group consisting of cobalt, nickel and copper.

[0036]

[0035] In some embodiments, the transition metal is cobalt.

[0037]

[0036] In some embodiments, the shell has a thickness of less than 5 nm, such as less than 3 nm, for example less than 2 nm.

[0038]

[0037] In some embodiments, the particulate photocatalyst comprises particles with a distribution of particle sizes. The particles sizes may, for example, range from less than 40 nm to greater than 80 nm.

[0039]

[0038] In some embodiments, the light comprises visible light in the range of 450-700 nm.

[0040]

[0039] In some embodiments, the solar-to-hydrogen (STH) efficiency is at least 4%, such as at least 4.5%, for at least 1 hour of irradiation. In some embodiments, the solar-to-hydrogen (STH) efficiency is at least 4%, such as at least 4.5%, for at least 100 hours of irradiation.

[0041]

[0040] In accordance with a fifth aspect the invention provides a particulate photocatalyst comprising: a plasmonic core comprising a metallic or metal oxide nanoparticle; a shell at least partially surrounding the plasmonic core, the shell comprising nitrogen-doped carbon; and cobalt immobilised on the shell, wherein at least a portion of the cobalt is atomically dispersed.

[0042]

[0041] Other features of the particulate photocatalyst may generally be as disclosed herein in the context of the fourth aspect.

[0043]

[0042] 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.

[0044]

[0043] Further aspects of the invention appear below in the detailed description of the invention.

[0045] Brief Description of Drawings

[0046]

[0044] Embodiments of the invention will herein be illustrated by way of example only with reference to the accompanying drawings in which:

[0047]

[0045] Figure 1 is an X-Ray diffraction (XRD) pattern of NC@Cu and Co-NC@Cu photocatalysts produced in Example 1 , with comparison against the known XRD patterns of CuO, Cu metal and graphite.

[0048]

[0046] Figure 2 is a histogram showing the particle size distribution of Co-NC@Cu photocatalyst produced in Example 1 , as determined from TEM images.

[0047] Figure 3 depicts Fourier-transform extended X-ray absorption fine structure (FT- EXAFS) spectra of Co-NC@Cu photocatalyst produced in Example 1 , with comparison against spectra for metallic cobalt foil and cobalt(ll) phthalocyanine (CoPC).

[0049]

[0048] Figure 4 depicts spatially resolved cathodoluminescence (CL) spectra for edge sites on various differently sized particles of Co-NC@Cu photocatalyst produced in Example 1.

[0050]

[0049] Figure 5 depicts electron energy loss spectroscopy (EELS) spectra at different imaged locations on a Co-NC@Cu photocatalyst particle, as produced in Example 1.

[0051]

[0050] Figure 6 is a graph of the H2 production rate in two-hour photocatalytic saline water splitting reactions using NC@Cu and Co-NC@Cu photocatalysts produced in Example 1 , with comparison against Cu, CuO and CoPc.

[0052]

[0051] Figure 7 is a graph of the H2 production rate in two-hour photocatalytic saline water splitting reactions using Co-NC@Cu photocatalysts as produced in Example 1 with differing amounts of C0CI2 precursor.

[0053]

[0052] Figure 8 is a graph of the H2 production rate in two-hour photocatalytic saline water splitting reactions using Co-NC@Cu photocatalysts as produced in Example 1 , at different controlled reaction temperatures.

[0054]

[0053] Figure 9 is a graph of the H2 production rate in two-hour photocatalytic saline water splitting reactions using Co-NC@Cu photocatalysts as produced in Example 1 , when irradiated with light of differing wavelength.

[0055]

[0054] Figure 10 is a graph of the H2 production rate, and resultant reaction temperature, in two-hour photocatalytic saline water splitting reactions using Co-NC@Cu photocatalysts as produced in Example 1 , with differing concentrations of NaCI in the aqueous reaction medium.

[0056]

[0055] Figure 11 is a graph of the solar-to-thermal efficiency over time during an extended photocatalytic saline water splitting reaction using a Co-NC@Cu photocatalyst as produced in Example 1 .

[0057]

[0056] Figure 12 is a graph of H2 production rate over time during an extended photocatalytic saline water splitting reaction using a Co-NC@Cu photocatalyst as produced in Example 1 .

[0058]

[0057] Figure 13 is a graph showing time-resolved photoluminescence (TRPL) measurements for Co-NC@Cu photocatalyst as produced in Example 1 in pure water vs concentrated aqueous NaCI solutions.

[0059]

[0058] Figure 14 is a simulated spectrum of the local E-field distribution of a Co-NC@Cu photocatalyst with Cu nanoparticle cores of differing sizes and 10 nm NC shell.

[0059] Figure 15 is a DFT model of intermediates involved in the reaction pathway for the hydrogen evolution reaction conducted on NC@Cu and Co-NC@Cu, with and without applied electric field perturbation.

[0060]

[0060] Figure 16 is a DFT model of intermediates involved in the reaction pathway for the oxygen evolution reaction conducted on NC@Cu and Co-NC@Cu, with and without applied electric field perturbation.

[0061]

[0061] Figure 17 is a graph of H2 production rate in dehydrogenation of methanol-water mixtures using a Co-NC@TiC>2 photocatalyst, in different catalyst concentrations, as performed in Example 12.

[0062]

[0062] Figure 18 is a graph of H2 production rate in dehydrogenation of methanol-water mixtures, containing different methanol concentrations, using a Co-NC@TiC>2 photocatalyst as performed in Example 12.

[0063]

[0063] Figure 19 is a graph of H2 production rate in dehydrogenation of different alcohol- water mixtures, using a Co-NC@TiC>2 photocatalyst as performed in Example 12.

[0064]

[0064] Figure 20 is a graph of H2 production rate in dehydrogenation of methanol-water mixtures using Co-NC@TiC>2 photocatalyst, as produced in solvothermal reactions with different reaction times, as performed in Example 13.

[0065] Detailed Description

[0066] Particulate photocatalyst

[0067]

[0065] The present disclosure relates to a particulate photocatalyst. The particulate photocatalyst comprises a plasmonic core comprising a metallic nanoparticle and a shell comprising nitrogen-doped carbon which at least partially surrounds the plasmonic core. A transition metal is immobilised on the shell, with at least a portion of the transition metal being atomically dispersed.

[0068]

[0066] The particulate photocatalyst thus comprises a plurality of particles having a coreshell morphology. The core-shell particles may be predominantly nanoparticles, i.e. particles having at least one dimension in the nanometre range (<100nm), although the presence of a fraction of larger particles is not excluded. In some embodiments, the particles are substantially spheroidal. In some embodiments, the particles have an aspect ratio of less than 3:1 , such as less than 2:1.

[0069]

[0067] In some embodiments, at least 50%, such as at least 70%, for example at least 80%, of the particles (by number) have a particle size of less than 100 nm. As used herein, the particle size refers to the maximum dimension (diameter) of a particle. In some embodiments, the particulate photocatalyst has an average particle size of less than about 80 nm. As used herein, the average particle size refers to the number average particle size (where each observed particle has an equal contribution to the particle size distribution subjected to averaging, regardless of size). Individual particle sizes, and the average particle size of the particulate photocatalyst, may be measured or estimated, for example, by using microscopy techniques (e.g. transmission electron microscopy) with image analysis software used to assign each observed particle a size.

[0070]

[0068] Preferably, the particles of the particulate photocatalyst have a distribution of particle sizes, for example ranging from particles of less than 40 nm to particles of greater than 80 nm (and thus also including particles of various sizes between 40 and 80 nm). A broad distribution of particle sizes in the nanometre range (<100nm) may advantageously facilitate broadband absorption of light in the visible and near-infrared regions, associated with the generation of multiple resonances at different wavelengths by the differently sized plasmonic cores.

[0071] Plasmonic core

[0072]

[0069] The particulate photocatalyst comprises a plasmonic core. The plasmonic core may be a metallic nanoparticle. As used herein, a metallic nanoparticle refers to a nanoparticle comprising and preferably consisting of metal elements in their metallic form, i.e. metal(O), including both pure metals and alloys.

[0073]

[0070] Plasmonic metallic nanoparticles are metallic nanoparticles which interact strongly with incident light having a wavelength longer than the particle size, causing strong scattering and / or absorption of the light. The interaction occurs when conduction electrons on the nanoparticle surface oscillate when excited by light of a specific wavelength. The oscillation is known as a localised surface plasmon resonance (LSPR). The plasmonic properties of metallic nanoparticles depend on the wavelength-dependent dielectric function (polarizability) of the nanoparticle (as well as the dielectric function of the surrounding medium) and are thus affected by the nanoparticle composition and size.

[0074]

[0071] It was observed by cathodoluminescence spectroscopy that particulate core-shell photocatalysts, comprising plasmonic copper nanoparticle cores and nitrogen-doped carbon (NC) shells, can absorb a wide range of wavelengths in the range of 400 to 900 nm, with differently sized plasmonic cores providing stronger absorption in different portions of this spectrum. Advantageously, this allows efficient use of the solar spectrum in embodiments where the light used to induce photocatalysis is solar radiation. Without wishing to be bound by any theory, it is proposed that the resultant localised surface plasmon resonance (LSPR) perturbs the surface of the NC-shell by strong E-field effects and transfer of energetic charge carriers, thus modulating photocatalytic activities and selectivities.

[0072] Metallic copper nanoparticle cores have been found particularly effective for plasmonic enhancement of photocatalytic dehydrogenation reactions taking place on the adjacent shell surface. The inventors have shown by experiment that the dihydrogen production rate is enhanced by about an order of magnitude during photocatalytic saline water splitting with core-shell catalysts as disclosed herein, in comparison to otherwise similar photocatalysts lacking the plasmonic core. Surprisingly, the copper nanoparticle cores were found to remain metallic with no observable oxidation to copper oxides even after extended photocatalytic water splitting reactions. Moreover, copper has the advantage of being relatively earth abundant and lower cost compared to some other metals used in plasmonic nanoparticles (e.g. the noble metals).

[0075]

[0073] Other base metal nanoparticles with suitable plasmonic properties may include aluminium and bismuth. Noble metal nanoparticles also have good plasmonic properties and may also be used as plasmonic cores of the particulate photocatalysts disclosed herein. Nobel metal nanoparticles have the advantage of intrinsic stability against oxidation and can therefore be expected to retain their plasmonic properties over extended periods of use.

[0076]

[0074] In some embodiments, therefore, the plasmonic metallic nanoparticles comprise a metal selected from the group consisting of copper, aluminium, bismuth, gold, silver and alloys thereof. In some embodiments, the plasmonic metallic nanoparticles comprise, and preferably consist of, a metal selected from the group consisting of copper, gold, silver and alloys thereof. In some embodiments, the plasmonic metallic nanoparticles comprise, or consist of, copper or a copper alloy. Suitable copper alloys may include AgCu, for example. In some embodiments, the plasmonic cores are copper metal nanoparticles.

[0077] Shell

[0078]

[0075] The particulate photocatalyst comprises a shell which at least partially surrounds the plasmonic core. The shell may thus partially or completely surround the plasmonic core. A shell which partially surrounds the core may cover at least 50% of the shell surface, or at least 60%, at least 70%, at least 80%, or at least 90% of the shell surface. In some embodiments, the shell completely encapsulates the plasmonic core, typically as a thin continuous layer which completely surrounds the plasmonic core. The shell may have a thickness in the range of 1 nm to 15 nm, such as in the range of 1.5 nm to 10 nm. In some embodiments, the shell is less than 3nm, or less than 2 nm, thick. An encapsulating shell having a thickness in such ranges may advantageously: (i) maintain the active catalytic sites on the shell surface in close proximity to and at near-uniform distances from the plasmonic core, thus allowing plasmonic enhancement of the photocatalytic reaction, and / or (ii) protect the plasmonic core from exposure to the reagents, intermediates and products in the photocatalytic reaction, thus preventing e.g. undesired oxidation of the core.

[0079]

[0076] The thickness of the shell may be varied, to a degree, by controlling the reaction conditions, e.g. the reaction time, of the solvothermal procedure by which the nitrogen-doped carbon is deposited on the nanoparticle cores. The inventors have found that coherent shells with a thickness of less than 2 nm may be produced.

[0080]

[0077] The shell comprises nitrogen-doped carbon. Nitrogen-doped carbon is a solid carbon material which is doped with nitrogen atoms, and may thus consist essentially of carbon and nitrogen atoms (e.g. at least 95 wt.% carbon and nitrogen). In some embodiments, the nitrogen-doped carbon comprises graphitic carbon, with the material thus arranged in planar sheets having a honeycomb arrangements of carbon atoms. In such materials, nitrogen atoms may substitute a fraction of the carbon atoms at doping sites on the sheet edges and within the sheet, with the nitrogen atoms present as pyridinic, pyrrolic and graphitic nitrogen atoms. Graphitic carbon within the nitrogen-doped carbon shell may be detected via Raman spectroscopy.

[0081]

[0078] Nitrogen atoms within the nitrogen-doped carbon shell are believed to coordinate to and thus immobilise the atomically dispersed transition metal, thereby providing active sites for photocatalysis.

[0082] Transition metal dopant

[0083]

[0079] The particulate photocatalyst comprises a transition metal immobilised on the shell. At least a portion of the transition metal is atomically dispersed. The atomically dispersed transition metal may be coordinated to one or more nitrogen atoms of the nitrogen-doped carbon. Without wishing to be limited by any theory, it is proposed that M-NC sites on the shell surface, where M is the atomically dispersed transition metal, are active sites for photocatalytic reactions including dehydrogenation reactions such as water splitting.

[0084]

[0080] In some embodiments, the transition metal comprises one or more group 7 to 11 transition metals. The inventors have demonstrated enhanced water splitting activity of coreshell catalysts doped with monoatomic cobalt, copper, nickel, iron, gold, and platinum, in comparison to the undoped analogue. In some embodiments, the transition metal comprises one or more group 9 to 11 transition metals. In some embodiments, the transition metal is selected from the group consisting of cobalt, nickel, copper and combinations thereof. Base metals dopants capable of providing high photocatalytic activities may be preferred over less earth-abundant noble metal dopants. In some embodiments, the transition metal is cobalt. Cobalt-doped core shell catalysts, as disclosed herein, have been found particularly active for photocatalytic dehydrogenation reactions such as (sea)water splitting.

[0081] The transition metal may be present in any amount that can be immobilised, predominantly as atomically dispersed atoms, on the shell. In some embodiments, the particulate photocatalyst comprises the transition metal in an amount of between 0.2 wt.% and 1 wt.%, such as between 0.25 wt.% and 0.6 wt.%, based on the total weight of the particulate photocatalyst.

[0085]

[0082] Preferably, substantially all of the immobilised transition metal is atomically dispersed. In particular, the particulate photocatalyst may be substantially free of surfacebound clusters or nanoparticles of the doped transition metal. Acid treatment (e.g. nitric acid washing) of the photocatalyst during its preparation may remove any clusters of transition metal, and the absence of clusters can be inferred from microscopic imaging (e.g. HAADF- STEM images) where the transition metal can be seen only in atomically dispersed form.

[0086] Method of producing a particulate photocatalyst

[0087]

[0083] The present disclosure further relates to methods of producing a particulate photocatalyst comprising metallic nanoparticle plasmonic cores, as disclosed herein.

[0088]

[0084] The method may comprise combining metallic or metal oxide nanoparticles with a salt of a transition metal and a nitrogen-bearing organic compound. In some embodiments, the nanoparticles are dispersed in a liquid comprising the salt of a transition metal and a nitrogenbearing organic compound. The combination, such as a dispersion, is thermally treated at a temperature sufficient to (i) decompose at least a portion of the nitrogen-bearing organic compound, thereby producing nitrogen-doped carbon in shells which at least partially surround the nanoparticles, (ii) reduce the metal oxide of the nanoparticles, when metal oxide nanoparticles are present, to form metallic nanoparticles; and (iii) immobilise at least a portion of the transition metal, as atomically dispersed transition metal, on the shells.

[0089]

[0085] In some embodiments, the method comprises providing a dispersion of nanoparticles comprising a metal oxide in a liquid comprising a salt of a transition metal and a nitrogen-bearing organic compound. The dispersion is thermally treated at a temperature sufficient to (i) decompose at least a portion of the nitrogen-bearing organic compound, thereby producing nitrogen-doped carbon in shells which at least partially surround the nanoparticles, (ii) reduce the metal oxide to form metallic nanoparticles; and (iii) immobilise at least a portion of the transition metal salt, as atomically dispersed transition metal, on the shells.

[0090]

[0086] The method may thus involve a solvothermal treatment which produces, typically in a single reaction step, the particulate photocatalyst comprising plasmonic metallic nanoparticle core, shell comprising nitrogen-doped carbon surrounding the plasmonic core, and atomically dispersed transition metal immobilised on the shell.

[0087] The liquid comprises a nitrogen-bearing organic compound which thermally decomposes in the solvothermal process to produce nitrogen-doped carbon. In some embodiments, the nitrogen-bearing organic compound is itself a liquid at the temperature of the reaction, and may therefore be the primary or only liquid component present. Inert liquid solvents may therefore not be required, and in some embodiments are absent. In other embodiments, the liquid may further comprise a solvent, such as ethanol, in combination with the nitrogen-bearing organic compound. Suitable nitrogen-bearing organic compounds may include amides, such as formamide (melting point c.a. 3°C), dimethylformamide (melting point c.a. -61 °C), acetamide (melting point c.a. 80°C) and urea (melting point c.a. 140°C), aromatic amines, such as aniline (melting point c.a. -6°C), and N-heteroaromatic compounds, such as pyridine (melting point c.a. -42°C), pyrrole (melting point c.a. -23°C) and melamine. In some embodiments, the nitrogen-bearing organic compound comprises formamide.

[0091]

[0088] The temperature at which the synthesis is conducted is high enough to decompose the nitrogen-bearing organic compound. It will be appreciated that the required temperature may thus depend on the choice of nitrogen-bearing organic compound. In some embodiments, the temperature is above 120°C, such as above 150°C, for example between 160°C and 200°C. A temperature of 180°C was found to be suitable when formamide is used. High temperature calcination (e.g. 400-600°C) may be required to form a nitrogen-doped carbon shell from some precursors such as melamine.

[0092]

[0089] Under the high temperature, e.g. solvothermal, reaction conditions, the nitrogenbearing organic compound decomposes to form nitrogen-doped carbon, which has been found to encapsulate nanoparticles dispersed in the liquid, thus forming the shell of the core-shell particle morphology. The thickness of the shell may be controlled by varying the solvothermal reaction conditions, for example the reaction time.

[0093]

[0090] The liquid subjected to heating in the photocatalyst synthesis may comprise dispersed nanoparticles comprising a metal oxide and a salt of a transition metal. Under the solvothermal conditions where the nitrogen-bearing organic compound is decomposed, the transition metal may be immobilised, predominantly in atomically dispersed form, on the shells. It is proposed that the transition metal coordinates to nitrogen atoms of the nitrogen-bearing organic compound and / or its decomposition products during the synthesis, thus facilitating the atomic dispersion of the transition metal in the final product.

[0094]

[0091] The transition metal salt may comprise any salt or salts of the required transition metal dopant that can be dispersed and preferably dissolved in the liquid, and which can be immobilised in situ to form atomically dispersed transition metal on the shell. In some embodiments, the transition metal salt is a halide salt, such as a chloride salt. Non-limiting examples of such salts include FeCh, COCI2.6H2O, NiCh, CuCh, PdCh, AgCI, ^PtCle x W, and HAUCI4' 3H2O.

[0095]

[0092] Surprisingly, it has now been found that the metal oxide composition of certain nanoparticle precursors can be reduced in situ to form metallic nanoparticles as the plasmonic cores of core-shell particles. Without wishing to be limited by any theory, it is proposed that reductants such as carbon monoxide (CO) and ammonia (NH3) are produced during thermal decomposition of the nitrogen-bearing organic compound, and that these reductants then reduce the metal oxide nanoparticles. The inventors have found that solvothermal treatment of a dispersion of copper (II) oxide (CuO) nanoparticles in formamide completely reduces the CuO to form metallic copper nanoparticle cores encapsulated by a thin shell of nitrogen-doped carbon.

[0096]

[0093] After the synthesis of the particulate photocatalyst with core-shell morphology, the method may comprise one or more post-processing steps. In some embodiments, the particulate photocatalyst is treated with an acid, for example dilute nitric acid. The acid treatment may remove any transition metal which is not atomically dispersed and immobilised on the shell. In particular, the acid may remove any transition metal clusters, thus ensuring that the transition metal remaining in the photocatalyst is predominantly or substantially entirely atomically dispersed.

[0097]

[0094] In some embodiments, the particulate photocatalyst (e.g. after acid treatment) is thermally annealed under an inert atmosphere, for example under argon gas. The annealing may be conducted to a maximum temperature greater than the temperature of solvothermal synthesis, for example a maximum annealing temperature of greater than 300°C, such as about 400°C. Without wishing to be limited by any theory, the annealing may strengthen and condition the nitrogen-doped carbon shell, remove volatile components, and assist to crystallise the metallic core.

[0098] Uses

[0099]

[0095] The present disclosure further relates to use of the particulate photocatalyst disclosed herein (containing a plasmonic metallic nanoparticle core) in a photocatalytic process for oxidising or reducing a reactant.

[0100]

[0096] A wide range of photocatalytic reactions is contemplated, including but not limited to water splitting, reforming of organic feedstocks, carbon dioxide reduction, organic waste oxidation, including in aqueous systems, dinitrogen reduction, methane oxidation, ammonia oxidation, and reduction (e.g. hydrogenation) or oxidation (e.g. dehydrogenation) of a range of other small molecules.

[0097] In some embodiments, the reactant is a hydrogen-bearing compound and the hydrogen-bearing compound is photocatalytically dehydrogenated (an oxidation reaction of the precursor compound) to produce dihydrogen. In some embodiments, the hydrogen-bearing compound is water and the photocatalytic process is thus a photocatalytic water splitting process. In other embodiments, the hydrogen-bearing compound is an organic compound, such as an alcohol, for example a C1-C4 alcohol, e.g. methanol, ethanol, isopropanol, 1 ,2- propanediol or glycerol. The photocatalytic process is thus a photocatalytic reforming process.

[0101] Photocatalytic method

[0102]

[0098] The invention further relates to a photocatalytic method. The photocatalytic method comprises contacting a composition comprising a hydrogen-bearing compound with a particulate photocatalyst, and irradiating the particulate photocatalyst with light, thereby photocatalytically dehydrogenating at least a portion of the hydrogen-bearing compound with the particulate photocatalyst to form dihydrogen. The particulate photocatalyst comprises a plasmonic core and a shell comprising nitrogen-doped carbon which at least partially surrounds the plasmonic core. A transition metal is immobilised on the shell. At least a portion of the transition metal is atomically dispersed.

[0103]

[0099] In the broadest form of the methods disclosed herein, the plasmonic core of the particulate photocatalyst may generally be any nanoparticles with plasmonic properties. Plasmonic properties of nanoparticle cores may be ascertained by cathodoluminescence spectroscopy or electron energy loss spectroscopy.

[0104] Particulate photocatalyst used in the method

[0105]

[0100] The particulate photocatalyst used in the method is generally as already described herein, except that in the broadest form the plasmonic core may comprise any nanoparticle with plasmonic properties. In some embodiments, the plasmonic core comprises a metallic or metal oxide nanoparticle.

[0106]

[0101] The shell comprising nitrogen-doped carbon, and the atomically dispersed transition metal immobilised on the shell, may be according to any of the embodiments already disclosed herein in the context of the particulate photocatalysts with metallic nanoparticle plasmonic cores.

[0107]

[0102] In some embodiments, the plasmonic core comprises a metallic nanoparticle. The metallic nanoparticle plasmonic core may generally be according to any of the embodiments already disclosed herein. In some embodiments, the metallic nanoparticle comprises metallic copper.

[0103] In other embodiments, the plasmonic core comprises a metal oxide nanoparticle. Certain oxygen-deficient (defect-rich) or doped metal oxide nanoparticles have plasmonic properties similar to metallic nanoparticles, and may thus enhance photocatalytic reactions taking place on the shell surface, such as (sea)water splitting, according to the principles disclosed herein.

[0108]

[0104] In some embodiments, the metal oxide nanoparticle core is selected from titanium dioxide nanoparticles, tungsten oxide nanoparticles and molybdenum oxide nanoparticles. Defect-rich forms of such metal oxide nanoparticles, which may have suitable plasmonic properties, are denoted as TiC>2-x, MoCh-x, and WCh-x.

[0109]

[0105] Particulate photocatalysts comprising plasmonic metal oxide nanoparticle cores may generally be produced by the same solvothermal procedure used to produce the particulate photocatalysts with plasmonic metallic nanoparticle cores, except that the metal oxide nanoparticles are not reduced in situ to metallic form. Unlike with CuO nanoparticles, the inventors have found that certain metal oxide nanoparticle precursors, such as TiC>2 and WO3 nanoparticles, are not fully reduced to metallic form during the solvothermal procedure. Without wishing to be limited by any theory, it is believed that the reductive conditions instead produce oxygen-deficient oxidic nanoparticles having defect sites which confer plasmonic properties on the resultant metal oxide core in the core-shell particles.

[0110]

[0106] Thus, suitable methods for producing particulate photocatalysts with plasmonic metal oxide cores may comprise providing a dispersion of nanoparticles comprising a suitable metal oxide in a liquid comprising a salt of a transition metal and a nitrogen-bearing organic compound. The dispersion is thermally treated at a temperature sufficient to (i) decompose at least a portion of the nitrogen-bearing organic compound, thereby producing nitrogen-doped carbon in shells which at least partially surround the nanoparticles, and (ii) immobilise at least a portion of the transition metal, as atomically dispersed transition metal, on the shells.

[0111]

[0107] In some embodiments, the thermal treatment renders the nanoparticles plasmonic by partially reducing the metal oxide of the nanoparticles to an oxygen-deficient metal oxide form.

[0112] Composition comprising a hydrogen-bearing compound

[0113]

[0108] The photocatalytic method involves contacting a composition comprising a hydrogen-bearing compound with the particulate photocatalyst. Typically, the composition is a liquid. The particulate photocatalyst may thus be dispersed in the liquid during the photocatalysis. However, it is not excluded that the hydrogen-bearing compound is in gaseous form when contacted with the particulate photocatalyst.

[0109] In some embodiments, the hydrogen-bearing compound is water. The method may thus be a photocatalytic method of splitting water to produce H2. O2 may be produced as a byproduct. In such embodiments, a sacrificial reagent is not needed to produce H2.

[0114]

[0110] In some embodiments, the composition comprising a hydrogen-bearing compound is an aqueous solution comprising a salt. The salt may comprise NaCI, optionally in combination with other inorganic cations and anions. In some embodiments, the aqueous solution comprises salt, optionally NaCI, in an amount of at least 0.1 M, such as at least 0.2 M, or at least 0.3 M, or at least 0.4 M, such as at least 0.5 M. The inventors have found that the photocatalytic H2 production rate is enhanced under conditions of elevated salinity. It is proposed, on the basis of theoretical calculations, that this beneficial effect is due to salt ions absorbing onto the photocatalyst surface, inducing charge polarization and suppressing the recombination of photo-generated charge carriers.

[0115]

[0111] In some embodiments, the composition comprising a hydrogen-bearing compound is seawater.

[0116]

[0112] In other embodiments, the hydrogen-bearing compound is an organic compound such as an alcohol. The photocatalytic process may thus be a photocatalytic reforming process for H2 production.

[0117]

[0113] In some embodiments, the hydrogen-bearing compound is an alcohol, for example having from 1 to 4 carbon atoms (i.e. C1-C4 alcohols). In some embodiments, the C1-C4 alcohol is selected from the group consisting of methanol, ethanol, 2-propanol (isopropanol) and combinations thereof. Photocatalytic reforming of such alcohols is expected to produce hydrogen without co-producing substantial quantities of carbon dioxide, based on photocatalytic dehydrogenation selectivities seen with other systems (e.g. Uddin et al, Applied Catalysis B: Environmental 2021 292 120212). Other suitable C1-C4 alcohols may include 1 ,2- ethanediol (ethylene glycol), 1 ,2-propanediol, 1 ,3-propanediol, glycerol, 1-propanol, and 1- butanol. Other suitable alcohols may include benzyl alcohol.

[0118]

[0114] In some embodiments, the C1-C4 alcohol is selected from methanol, ethanol and combinations thereof. In some embodiments, the C1-C4 alcohol comprises, or consists of methanol. Methanol is particularly attractive as a hydrogen carrier due to its high gravimetric content of H2 and transportability.

[0119]

[0115] The composition may be composed predominantly, or entirely, of the organic compound to be dehydrogenated. For example, a C1-C4 alcohol may be present in an amount of at least 80 wt.%, or at least 90 wt.%, or at least 95 wt.%, such at least 99 wt.% of the liquid composition. However, it is not excluded that the composition may include other liquid components, for example less reactive or inert components such as hydrocarbons and the like.

[0116] In some embodiments where an organic compound is to be dehydrogenated, the composition may be substantially free of water. As used herein, “substantially free of water” means that water is either entirely absent or present in only trace amounts, such as below 0.5 wt.%, preferably below 0.1 wt.% and more preferably below 0.05 wt.%. Water may influence the activity and / or selectivity of photocatalytic alcohol reforming, for example by increasing the carbon dioxide selectivity (via formation of carbonyl species and water gas shift reaction).

[0120]

[0117] In other embodiments, the composition comprising a hydrogen-bearing compound is present in an aqueous mixture, such as a wastewater. The aqueous mixture may comprise one or more organic compounds, such as one or more alcohols, which are susceptible to photocatalytic dehydrogenation with the particulate photocatalyst. The method may thus simultaneously degrade organic pollutants present in the wastewater and produce dihydrogen.

[0121]

[0118] The organic compounds may be present as minor components of the aqueous mixture, such as less than 30 wt.%, or less than 20 wt.%.

[0122]

[0119] In some such embodiments, the organic pollutants in the wastewater are selectively dehydrogenated relative to water. H2 may thus be produced as the primary gaseous product, since the dehydrogenated by-products of the organic compound remain in solution. In particular, H2 may advantageously be produced in large excess relative to O2, thus mitigating the challenges of H2 separation from O2 in the process.

[0123] Irradiating the particulate photocatalyst

[0124]

[0120] The method comprises irradiating the particulate photocatalyst with light while the particulate photocatalyst is in contact with the composition comprising a hydrogen-bearing compound. At least a portion of the hydrogen-bearing compound is thus photocatalytically dehydrogenated to form dihydrogen.

[0125]

[0121] The particulate photocatalyst may be dispersed in the liquid composition during the irradiation, for example by stirring, sonication or other methods of agitation. However, it is not excluded that the particulate photocatalyst is supported on a solid substate. The particulate photocatalyst may be contacted with the composition in any suitable amount, for example at least 0.01 g / L, or at least 0.1 g / L, such as in the range of 0.01 to 10 g / L, or in the range of 0.05 to 1 g / L, or in the range of 0.1 to 0.5 g / L. The inventors have found that concentrations of 0.2 g / L are suitable to achieve excellent photocatalytic performance.

[0126]

[0122] The dehydrogenation reaction is driven photocatalytically, and therefore does not necessarily require elevated reaction temperatures. However, elevated reaction temperatures have been found to enhance the rate of photocatalytic seawater splitting, demonstrating that photothermal effects may be in operation. In some embodiments, the reaction medium is heated to the required reaction temperature due to the irradiation with light and high photothermal conversion efficiency, i.e. there is no need for externally applied heating other than via the irradiation with light. In some embodiments, the liquid composition is maintained at a temperature of above 50°C, such as above 80°C, for example above 100°C. In some embodiments, the liquid composition is maintained at a temperature of below 200°C, for example below 150°C, such as in the range of 20°C to 150°C, or in the range of 50°C to 140°C during the reaction.

[0127]

[0123] In some embodiments, the light comprises a broad spectrum of visible and nearinfrared radiation across a wavelength range of at least between 400 and 1000 nm, and particularly visible light in the range of 450-700 nm. For example, the light may comprise solar radiation. Advantageously, it has been shown that the particulate photocatalyst provides an excellent photocatalytic response when stimulated by light of broadly varying wavelengths.

[0128]

[0124] Optionally, the photocatalytic reaction is performed in a reactor equipped with a condenser lens or other means of concentrating solar radiation into the reactor. The resulting higher power input of irradiation into the reactor may thus increase the photocatalytic reaction rate.

[0129]

[0125] The hydrogen-bearing compound may be dehydrogenated to produce dihydrogen at a rate of at least 1 mmol. g-1. IT1, such as at least 2 mmol. g-1. IT1, or at least 5 mmol.g’1.h’1over a reaction period of at least one hour, for example at least 2 hours. In some embodiments, the solar-to-hydrogen (STH) efficiency is at least 4%, such as at least 4.5%, over a reaction period of at least 1 hour, or at least 2 hours, for example at least 100 hours.

[0130]

[0126] The particulate photocatalyst may advantageously be used to dehydrogenate the hydrogen-bearing compound for extended periods, for example at least 100 hours, or at least 300 hours. It was found by experiment that particulate photocatalysts according to some embodiments could be used in saline water splitting for over 340 hours with little observed deactivation.

[0131]

[0127] In embodiments where the hydrogen-bearing compound is water, dioxygen (O2) is formed as a by-product of the reaction. The H2 and O2 products may be removed continuously or intermittently from the reactor to avoid pressure build-up, and separated downstream of the reactor by known methods.

[0132] EXAMPLES

[0133]

[0128] 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 and methods

[0134]

[0129] The following chemicals were sourced: Copper (II) oxide (CuO; Sigma-Aldrich); cobalt (II) chloride hexahydrate (COCI2 6H2O; Sigma-Aldrich); cobalt(ll) phthalocyanine (“CoPC”; Sigma-Aldrich-ACS reagent, p-form, Dye content 97 %); titanium(IV) oxide (Sigma- Aldrich, nanopowder, <25 nm particle size, 99.7% trace metals basis); tungsten(VI) oxide (Sigma-Aldrich, nanopowder, <100 nm particle size (TEM)); copper(ll) chloride (Sigma-Aldrich, powder, 99%); nickel(ll) chloride (Sigma-Aldrich, anhydrous for synthesis); gold(lll) chloride trihydrate (ACS reagent, 5 9.0% Au basis); iron(lll) chloride (Sigma-Aldrich, anhydrous, powder, 5^99.99% trace metals basis); chloroplatinic acid hydrate (Sigma-Aldrich, 2599.9% trace metals basis); nitric acid (Sigma-Aldrich-ACS reagent, 70%); formamide (Sigma-Aldrich- ACS reagent, 5^99.5%);; deionised (DI) water. All the chemical reagents were of analytical grade and used without any further purification.

[0135]

[0130] X-ray absorption spectroscopy (XAS) experiments were performed on the Medium Energy X-ray Absorption Spectroscopy (MEX-1) beamline stationed at the Australian Synchrotron (AS). The X-ray absorption near edge structure (XANES) and extended X-ray absorption fine structure (EXAFS) data for Co K-edge were acquired under ambient conditions. The subsequent interpretation of XAS data was facilitated using the ATHENA software suite.

[0136]

[0131] The surface analysis of X-ray photoelectron spectroscopy (XPS) was collected by an Al Ka X-ray source, 1486.6 eV (Kratos AXIS Ultra DLD).

[0137]

[0132] A FEI Verios scanning electron microscope (SEM) equipped with Everhart-Thornley and immersion in-lens detectors for secondary electrons detection was used to image the morphology of the samples. The same system was used to acquire cathodoluminescence (CL) signals by using a Gatan MonoCL4 Elite system. Full spectrum CL maps were acquired using a charge-coupled device (CCD).

[0138]

[0133] XRD patterns were acquired by Bruker X-ray diffractometer (Cu Ka).

[0139]

[0134] EPR spectra were acquired by a Bruker E500 spectrometer equipped with a Bruker ER 4122 SHQE resonator. Light source is 300 W Xenon lamp from PerfectLight.

[0140]

[0135] Gas products were analyzed by Nexis GC-2030 (Shimadzu Scientific instruments) which is equipped with a thermal conductivity detector (TCD), and a flame ionization detector (FID) with the ShinCarbon column.

[0141]

[0136] Time-resolved photoluminescence (TRPL) curves are captured at room temperature using an Horiba iHR 320 mm spectrophotometer equipped with a compact singlephoton silicon detector (PPD-900, detection range between 350-920nm). A 485-nm picosecond pulse laser (Horiba DeltaDiode) is used as an excitation source. The laser light is focused on the samples through an Olympus BX53 microscope equipped with a 50x objective lens and an X-Y micropositioner.

[0142]

[0137] Micro- Raman spectroscopy measurements are conducted using a Horiba Labram system, which is equipped with confocal optics, a 532 nm diode-pumped solid-state (DPSS) laser, and a charge-coupled device (CCD) Si detector with detection range of 540-1000 nm. The laser light is focused onto the samples through a PL FLUOTAR microscope objective (50x magnification and a 0.55 numerical aperture). The excitation power directed onto the sample remains consistent across all measurements unless specified otherwise in the manuscript.

[0143]

[0138] The Co content in different samples was measured by an Agilent ICP-OES 5110. The powder samples were digested with 10% HNO3 to obtain the accurate metal compositions.

[0144] Example 1. Preparation of Co-NC@Cu

[0145]

[0139] Initially, 1g of CuO is dispersed in 30 mL of formamide (HCONH2) and stirred for 30 min until the solution is uniformly coloured. Subsequently, the required amount of C0CI2 (e.g. 100 mg of C0CI2, corresponding to 183.3 mg of COCI2 6H2O) is added to the solution and stirred for 2 h. Then, the mixture of 30 mL solution is transferred into a 100 mL autoclave for solvothermal treatment at 180°C for 12 hours. Under these reaction conditions, part of the formamide converts into amorphous nitrogen-doped carbon (NC) material. After the hydrothermal reaction, the solid was recovered by centrifugation at 7800 revolutions per minute (rpm), and washed 4 times with 1 :1 ethanol and water, with solids recovery by centrifugation. The washed material is then treated with 0.1 mM dilute nitric acid 3 times to remove any reduced cobalt (e.g. cobalt nanoparticles) which is not atomically dispersed and immobilised on the NC-shell, followed by water rinsing to remove the nitric acid. Subsequently, the material is dried in a vacuum oven at 60°C for 12 hours. The dried powder sample is ground and placed into a quartz boat, followed by annealing in an Ar environment in a tube furnace to condition the nitrogen-doped carbon shell and remove volatiles. The reaction temperature is set at 400°C for 4 hours with a heating rate of 10°C per min. Finally, the resulting sample is ground again to complete the synthesis process, thus producing Co-NC@Cu.

[0146]

[0140] The photocatalysts produced using this methodology are shown in Table 1. Inductively coupled plasma optical emission spectrometry (ICP-OES) analysis was used to determine the actual Co loading, which is also shown in the Table. Table 1.

[0147] Example 2 (comparative). Preparation of NC@Cu

[0148]

[0141] The synthesis steps for NC@Cu are same as Co-NC@Cu (Example 1), except that C0CI2 was not added.

[0149] Example 4. Physical characterisation of Co-NC@Cu and NC@Cu photocatalysts

[0150]

[0142] Characterisation of the Co-NC@Cu-100 and NC@Cu photocatalysts indicated a core-shell morphology comprising a nanoparticle core of reduced metallic copper and a shell of amorphous nitrogen-doped carbon. Monoatomic cobalt atoms are dispersed on the shell, coordinated to nitrogen atoms, in Co-NC@Cu-100. It is proposed that CO and NH3 generated from formamide in the hydrothermal synthesis reduced the CuO to form Cu metal nanoparticle cores.

[0151]

[0143] X-ray diffraction (XRD) results, shown in Figure 1 reveal that the main diffraction peaks of the core component for both NC@Cu and Co-NC@Cu samples match with face centered cubic Cu, without any apparent peaks corresponding to CuO substances. The initial CuO nanoparticles were thus successfully reduced to metallic Cu nanoparticles. Furthermore, the XRD patterns of NC@Cu and Co-NC@Cu showed a weak C peak, which is associated with the NC shell layer.

[0152]

[0144] Transmission electron microscopy (TEM) images showed that the surface of the Cu nanoparticles was covered by thin NC layer. To investigate the dispersion and location of cobalt, the Co-NC@Cu photocatalyst was characterized on an aberration-corrected TEM (AC- TEM), using high-angle annular dark-field scanning TEM (HAADF-STEM) coupled with energy- dispersive X-ray spectroscopy (EDS) elemental mapping and atomic-resolution HAADF-STEM imaging. The results confirmed the monoatomic dispersion of Co atoms on the NC layer, with the Co atoms appearing as isolated bright spots in the HAADF-STEM images. No Co nanoparticles or clusters were observed.

[0145] Atomic-resolution STEM images also showed highly crystalline Cu lattices with a face-centered cubic (fee) structure.

[0153]

[0146] The particle size distribution of Co-NC@Cu, as determined from TEM images, is shown in Figure 2. A wide range of particles sizes, mostly below 100 nm, were present in the sample. The thickness of the NC shell was in the range of about 5 nm to about 10 nm.

[0154]

[0147] X-ray absorption spectroscopy (XAS) analysis was conducted to investigate the coordination environment of Co, providing further confirmation of the monoatomic dispersion of Co in Co-NC@Cu. The acquired data were compared with Co foil and CoPc as reference materials. The normalized X-ray absorption near-edge structure (XANES) spectra of the Co K- edge indicates that the Co K-edge energy of Co-NC@Cu is comparable with that of CoPc, suggesting similar chemical states of Co in these materials.

[0155]

[0148] In addition, Co-NC@Cu exhibits similar pre-edge profile to that of CoPc for Fourier- transform extended X-ray absorption fine structure (FT-EXAFS) spectra, as seen in Figure 3. In the FT-EXAFS Co K-edge, a prominent peak at 1.43 A corresponds to the first-shell coordination of the Co-N bond, mirroring the contribution observed in the CoPc reference value. Notably, no discernible Co-Co peak was detected at 2.17 A, indicating the negligible presence of metallic Co species. These findings corroborated the monoatomic dispersion of the Co-N coordination in Co-NC@Cu.

[0156]

[0149] X-ray photoelectron spectroscopy (XPS) was employed to investigate the valence states of elements. The Co 2p spectra of Co-NC@Cu exhibited two characteristic main peaks at 796.0 eV and 780.3 eV corresponding to the Co 2p3 / 2 and Co 2p1 / 2 orbitals. The peaks of N1s corresponding to graphite N (401.1 eV), pyridinic N (398.6 eV), pyrrolic N (400.1 eV), and N-metal species (399.4 eV, corresponding to Co-N interaction) can be resolved in N 1s XPS spectra.

[0157]

[0150] Raman spectroscopy confirmed the presence of a graphitic carbon structure in the NC-shell. The D band at 1433 cm-1is attributed to in-plane vibrations of sp2 carbon atoms in graphite structure, while the G band at 1678 cm-1originates from defects in graphene.

[0158]

[0151] The electron paramagnetic resonance (EPR) spectrum revealed the presence of N defects (i.e., the doped N in the shell) in Co-NC@Cu. The observed signal at g = 2.003 confirmed these N defects in the sample, suggesting atomic-level coordination of Co atoms with N within NC@Cu. Example 5. Cathodoluminescence (CL) spectroscopy and EELS spectroscopy of Co- NC@Cu photocatalysts

[0159]

[0152] The spatial and energy distribution of the resonances in Co-NC@Cu-100 was investigated using spatially resolved cathodoluminescence (CL) spectroscopy. Intensity maps of the Co-NC@Cu particles imaged by SEM were obtained by integrating the CL emission across different wavelength ranges from 400 to 900 nm. The core-shell nanoparticles with varied particle sizes (Figure 2) resulted in broadband absorption in the visible and near-infrared regions. This phenomenon arises from the existence of multiple plasmon modes excited within the differently sized plasmonic nanoparticles. CL measurements were conducted in raster scanning mode, locally exciting the nanostructure, and acquiring the full spectra of multiple pixels of the scanned SEM images. Selected CL spectra from different regions of the sample are shown in Figure 4. The analysis of the CL signals at different edge sites revealed one or more peaks in the spectral region of 500-750 nm, where the signals exhibit relatively stronger intensity. This demonstrates the generation of multiple resonances at different wavelengths by the different sized Cu cores in the Co-NC@Cu catalyst sample. The multiple resonances contribute to an overall broadband absorption in the visible and near-infrared spectra.

[0160]

[0153] To gain further insights into optical absorption by the plasmonic Cu nanoparticle cores, electron energy loss spectroscopy (EELS) was coupled with STEM-HAADF imaging to measure the local optical response. In Figure 5, the EELS spectra of Co-NC@Cu particles is depicted, exhibiting a distinctive absorption shoulder. Optical absorption in the range of 1.8-2.7 eV corresponds to wavelengths between 459-689 nm. This range matches the complete sum of spectra of CL presented in Figure 4. The intensity at positions I and II, near the edge and within the NC shell, respectively, are significantly higher than that the intensity at position III which is within the copper core. These results confirm the LSPR effect which is well-distributed around the Co-NC@Cu nanoparticles without quenching. These results highlight the LSPR phenomenon in Co-NC@Cu and demonstrate that Co-NC@Cu nanoparticles with core-shell structure can harvest light across a wide wavelength range from the visible to near-infrared regions.

[0161] Example 4. Preparation of other photocatalysts

[0162]

[0154] Core-shell photocatalysts with different nanoparticle cores and doping atoms on the shell, were prepared by the same methodology as Example 1. 1g of CuO, TiC>2 or WO3 nanoparticles was used, and 100 mg of the metal salt (doping metal precursor). The photocatalysts thus prepared are shown in Table 2. Table 2.

[0163]

[0155] XRD analysis indicated that the CuO nanoparticles were reduced to metallic Cu in the core of all photocatalysts. However, the WO3and TiC>2 nanoparticles remained in oxidic form in the nanoparticle cores. A core-shell morphology, with atomically dispersed doping metals on the NC shell, was present as for Co-NC@Cu.

[0164] Example 5. Photocatalytic experiments with Co-NC@Cu photocatalysts

[0165]

[0156] The photocatalytic water splitting reaction was carried out in a liquid-solid phase reaction in a quartz reactor tightly sealed by a septum. The reaction was conducted under light irradiation of a 300 W Xe lamp (PLS-SXE300D, Perfect Light) with AM 1.5G filter without the use of any noble metal, sacrificial agents or external heat source. The xenon light source was equipped with a condenser lens which generated a spot with a diameter of 1.6 cm and a light intensity of 600 mW cm"2(AM 1.5G).

[0166]

[0157] Typically, 2 mg of the sample was dispersed in 10 ml artificial seawater (aqueous NaCI solution) and was sonicated (B2500R-DTH, Branson) for 10 min. The solution was then transferred into a reaction vessel and stirred. Subsequently, the reactor was sealed and purged with pure Ar for 15 minutes to remove air. Finally, the xenon lamp was turned on to initiate the photocatalytic reaction. Following completion of the photocatalytic reaction, the reactor was allowed to cool, and gas chromatography (GO) measurements were conducted at room temperature to quantify the results.

[0158] Photocatalytic seawater splitting performance (1 M NaCI artificial seawater) with different photocatalysts is compared in Figure 6 (2 hour reactions). Cu, CuO and CoPc provided poor hydrogen production rates. In contrast, the Co-NC@Cu-100 photocatalyst with Co single atoms anchored on the NC shell outperforms NC@Cu without Co single atoms by over 10 times. A hydrogen production rate of 9080 pmol / g / h was obtained for the Co-NC@Cu- 100 photocatalyst. A reaction temperature of about 110°C (measured by thermocouple) was obtained due to the concentration of input light and the high photothermal conversion efficiency of the catalyst.

[0167]

[0159] Figure 7 shows the photocatalytic seawater splitting performance for Co-NC@Cu photocatalysts with different Co loadings. The optimal performance was obtained for Co- NC@Cu-100, followed by Co-NC@Cu-50 and Co-NC@Cu-250. These photocatalysts has a Co loading of between 0.3 and 0.5 wt.% (Table 1).

[0168]

[0160] Only a low rate of water splitting was obtained with Co-NC@Cu-100 under unfocused normal solar illumination (1 sun with 100 mW cm-2; c.a. 320 pmol / g / h H2 production), while no H2 or O2 products were detected when Co-NC@Cu-100 was heated in artificial seawater at temperatures up to 150°C in the dark.

[0169]

[0161] Figure 8 shows the effect of temperature on photocatalytic performance, under the same irradiation conditions. The reaction temperature (measured by thermocouple) was controlled using a water bath for the reactor. The results show that thermal effects contribute to H2 production.

[0170]

[0162] Without wishing to be limited by any theory, it is proposed that the photocatalytic reaction is driven by a combination of broad-spectrum solar illumination and photothermal effects (enhanced by focusing the light). The photothermal effect may include both photoinduced hot carrier generation and reaction environment heating. Reaction-environment heating will result in an increase in the thermal conductivity of seawater, enhancing heat transport and thereby promoting photothermal conversion efficiency.

[0171]

[0163] Figure 9 shows the relationship between the H2 yield and the wavelength of light. A bandpass filter on the xenon lamp was used to select the target wavelength for each experiment. As this resulted in a different light intensity at each wavelength, the results shown in Figure 9 were normalised to allow comparison by the following equation:

[0172] Normalised H2 yield = (1 sun intensity / single wavelength intensity) x H2 yield under single wavelength light

[0173]

[0164] The results correlate with the results from CL spectroscopy and EELS. The highest optical absorption and H2 production rates were obtained in the range of 459-689 nm. While the Co-NC@Cu photocatalyst can still absorb a significant amount of 700 nm photons, the lower energy of these photons is believed to produce photo-excited electron-hole pairs with lower energy and which therefore have relatively lower reduction / oxidation potentials. Consequently, light at this wavelength contributes less to the photocatalytic process.

[0174]

[0165] Figure 10 shows the effect of NaCI concentration on H2 production rate and the resultant reaction temperature. As NaCI concentration increases from 0 M to 1 M, the H2 production rate increases to 9080 pmol g-1h'1and the temperature rises to ~110°C. Further increasing NaCI concentration slightly increases the H2 production rate and temperature, eventually reaching saturation.

[0175]

[0166] The results show that the saline content in seawater can significantly enhance the performance of photocatalytic seawater splitting using the Co-NC@Cu photocatalyst.

[0176] Example 6. Long term photocatalytic experiments

[0177]

[0167] Long-term photocatalytic experiments with the Co-NC@Cu-100 photocatalyst were conducted as described in Example 5. After each cycle of 23 hours of reaction, the light source was turned off. The reactor was then degassed by purging with pure Ar for 15 minutes to remove gases from the previous reaction cycle, and the light source was reactivated to initiate the next reaction cycle. The photocatalytic performance was determined by GC measurements every two hours.

[0178]

[0168] As seen in Figure 11 , the solar-to-hydrogen (STH) efficiency was stable over 44 hours of reaction, with the average STH efficiency being 4.78%. Figure 12 shows that remarkable long-term stability and consistent photoactivity was observed over 340 hours, with the yield of H2 and O2 remaining close to the stoichiometric 2:1 ratio. The observed cycling of activity is due to the daily degassing to release significant pressure build-up of H2 and O2.

[0179]

[0169] After the long-term (340 hour) stability tests, the Co-NC@Cu-100 photocatalyst was recovered and characterised by XRD. The XRD spectrum, including characteristic peaks of Cu and C, remained comparable to the pristine Co-NC@Cu sample. The Cu core thus remained unoxidized after prolonged seawater splitting, indicating the effective protective attributes of the NC shell. EDS elemental mapping confirmed that the photocatalyst maintained the presence of constituent elements, notably Co, which remains well-dispersed throughout the material without any observable signs of aggregation.

[0180] Example 1. Time-resolved photoluminescence

[0181]

[0170] Time-resolved photoluminescence (TRPL) measurements, as shown in Figure 13, and density functional theory (DFT) calculations were used to study the effect of NaCI content on photocatalytic seawater splitting. The TRPL analysis was conducted with 1mg photocatalyst / 1 ml of NaCI solution. Figure 13 shows the normalized photoluminescence, meaning that the photoluminescence (PL) intensity values were adjusted so that they can be compared on the same scale, regardless of differences in overall intensity. Fitting of the TRPL spectra revealed an increase in charge carrier lifetime (from 12 ns for pure water, to 15 ns in 1 M NaCI, to 21 ns in 5 M NaCI), which can be attributed to the suppression of charge recombination due to the local polarization effect introduced near the surface by the adsorbed ionic species.

[0182]

[0171] The DFT analysis shows that traditionally inert cations / anions in seawater, such as Cl", Na+, as well as their hydrated forms, adsorb onto the surface of Co-NC@Cu, inducing charge polarization.

[0183]

[0172] The results suggest that the saline content in seawater enhances the photocatalytic performance (Example 5) by inducing strong charge polarization and thereby prolonging carrier lifetimes.

[0184] Example 8. Photocatalytic experiments with metal-doped NC@Cu photocatalysts

[0185]

[0173] Photocatalysts similar to Co-NC@Cu but with different metals doped on the NC shell, as prepared in Example 4, were evaluated in photocatalytic experiments as described in Example 5. The results are shown in Table 3.

[0186] Table 3.

[0187]

[0174] A range of other transition metals all provided enhanced photocatalytic activity for seawater splitting, relative to the undoped analogue NC@Cu.

[0188] Example 9. Photocatalytic experiments with metal-doped NC@W03 and NC@TiO2 photocatalysts

[0189]

[0175] Photocatalysts similar to Co-NC@Cu but with oxidic cores (either WO3 or TiC>2) and various metals doped on the NC shell, as prepared in Example 4, were evaluated in photocatalytic experiments as described in Example 5. The results are shown in Table 4. Table 4.

[0190]

[0176] Photocatalysts comprising oxidic nanoparticles cores and NC-shells doped with various metal atoms were also found to be active for photocatalytic seawater splitting.

[0191] Example 10. Seawater splitting with Co-NC@Cu photocatalysts

[0192]

[0177] The Co-NC@Cu-100 photocatalyst was evaluated for splitting of actual seawater using the procedure described in Example 5. The H2 yield rate over 2 hours was determined to be 5.3 mmol / g / h (c.f. 9.1 mmol / g / h with 1 M NaCI).

[0193] Example 11. Theoretical calculations

[0194]

[0178] To gain deeper insights into the local E-field distribution associated with the plasmonic properties of Co-NC@Cu, DFT calculations were conducted to predict optical properties and compared to Finite Difference Time Domain (FDTD) simulations conducted by structural modelling based on the STEM-HAADF results. FDTD simulation offers information of the local E-field enhancement induced by the LSPR effect across the entire Co-NC@Cu system.

[0195]

[0179] The local E-field distribution of Co-NC@Cu with a 50 nm Cu nanoparticle core and 10 nm NC shell was simulated at five different wavelengths (400, 500, 600, 700 and 800 nm). The E-field enhancement was found to be particularly prominent in the visible and near-infrared range, with negligible quenching phenomenon observed on the thin Co-NC shell. As evidenced by the spectra in Figure 14, enlarging the size of the Cu NPs core results in a red shift of the dominated resonant peaks. The increase in the core size of Cu NPs forms a larger plasmon resonator, thereby reducing the plasmon frequency and energy.

[0180] Increasing the Co-NC shell thickness (5 nm vs 10 nm) also lead to a minor red shift in peak energy and a slight decrease in E-field intensity, while still maintaining over *10 local E-field enhancement when the Cu nanoparticle size exceeds 30 nm under full-spectrum solar illumination. Therefore, the enhanced local E-field of the Cu nanoparticle core effectively transfers to the Co-NC shell surface without significant quenching by the shell.

[0196]

[0181] The local E-field profoundly influences the desorption and adsorption of intermediates during the reaction steps, thereby activating the chemical bonds and energy of the intermediates and modulating the Gibbs free energy and reaction activity. The effect of the E-field on the Gibbs free energy of each reaction step was thus simulated for NC shells with or without Co atoms as catalytic reaction sites.

[0197]

[0182] The HER (hydrogen evolution reaction) is typically described as a three-step process: the initial state H++ e; the intermediate adsorbed H*, and the free energy of the final product %H2, and the AGH* is considered to describe the activity. As seen in Figure 15, Co-NC were found to be more thermodynamically favourable for H2 production compared to the undoped NC. The applied E-field further reduces the GH* for Co-NC under E-field perturbation.

[0198]

[0183] In the DFT model depicted in Figure 16, the OER (oxygen evolution reaction) mechanism involves four concerted proton-electron transfer steps. Four transition state (TS) processes include TS1 as *H2O — *OH + *H, TS2 as *OH — *0 + *H, TS3 as *0 + *H2O — *OOH + *H, and TS4 as *OOH — *02 + *H. The impact of the E-field on the Gibbs free energy of each reaction step in the OER was thus simulated. The reaction barriers and ratedetermining steps (RDS) are crucial descriptors for assessing OER performance, so the intermediate adsorption structures for all potential sites were modelled. The free energy reaction pathway analysis indicates that Co-NC exhibits better oxidation reaction activity compared to NC alone.

[0199]

[0184] Furthermore, the transition state calculations reveal that the *H2O — *OH+*H step is the RDS (TS1). The energy barrier of this RDS decreases from 1.96 eV under no E-field to 1.83 eV under *10 E-field perturbation, favouring the activity of the oxidation reaction. The calculations show perturbing E-fields play a crucial role in achieving ideal photocatalytic water splitting performance, particularly for the LSPR-active Co-NC@Cu core-shell.

[0200]

[0185] Although FDTD simulations reveal that Co-NC nanoparticles maintain relatively high E-field intensity at a wavelength of 700 nm (Figure 14), the photon energy at this wavelength is relatively low. Therefore, it is proposed that the E-field perturbation alone does not determine H2performance. Instead, under the local electric field perturbation induced by LSPR, the energy barriers for reduction and oxidation reactions are reduced, and their enhancement degree on the photocatalytic activity is affected by the incident photon energy (i.e. , light wavelength). This accounts for the observed experimental data where the relatively high E-field intensity at 700 nm does not translate into enhanced H2production due to lower photon energy and decreased effectiveness of photo-excited carriers in driving the redox reactions.

[0201] Example 12. Photocatalytic alcohol dehydrogenation experiments with Co- NC@TiO2 photocatalysts

[0202]

[0186] Core-shell photocatalysts with titanium dioxide cores and cobalt atoms immobilised on the nitrogen-doped carbon shell were prepared by a similar methodology as Example 1. The reactant composition included 1g of TiC>2 nanoparticles, 30 ml of formamide and 50 mg of C0CI2. This mixture was heated to 180°C for 7 hours.

[0203]

[0187] TEM imaging confirmed that the nitrogen-doped carbon shell produced under these conditions completely encapsulated the core and had a thickness of less than 2 nm. The reduced shell thickness was ascribed to the shorter reaction time (7 hours vs 12 hours in Example 1). XANES and EXAFS analysis confirmed that cobalt dispersed on the shell was not present in metallic clusters but had properties more akin to CoPC, indicating monoatomic dispersion and likely coordination to nitrogen.

[0204]

[0188] This Co-NC@TiC>2 photocatalyst was evaluated for photocatalytic dehydrogenation of methanol (60%) in water, using a similar procedure as used in Example 5. Experiments were done with different photocatalyst loadings (between 0.15 g / L and 1 g / L) and the results are shown in Figure 17.

[0205]

[0189] The hydrogen evolution rate was maximum at a catalyst concentration of 0.2 g / L. At this optimal concentration, the hydrogen evolution rate is approximately 55 mmol g-1h’1, suggesting a good balance between sufficient active surface area and light penetration. At higher catalyst concentrations, the hydrogen evolution rate began to decrease, possibly due to: 1) light shielding effects at higher catalyst concentrations, where excess catalyst particles block light from reaching other particles, and / or 2) agglomeration of catalyst particles, reducing the effective surface area available for the reaction.

[0206]

[0190] Experiments were also done with different methanol concentrations (between 20 and 90%) at constant photocatalyst loading of 0.2 g / L and the results are shown in Figure 18. The dihydrogen production rate increases as the methanol concentration rises, indicating that methanol acts as an efficient sacrificial agent, donating electrons and enhancing hydrogen production. Little or no O2 was produced, confirming that methanol was selectively dehydrogenated in preference to water, forming products such as formic acid and formaldehyde.

[0191] Experiments were also done with different alcohols commonly found in wastewater, at 20% concentration in water and at constant photocatalyst loading of 0.2 g / L, and the results for methanol (MeOH), ethanol (EtOH), isopropanol (IPA), 1 ,2-propsane-diol (1 ,2 Pr-diol) glycerol (GLY) are shown in Figure 19. Methanol and ethanol provided the highest hydrogen production rates, but good activity was obtained with all alcohols tested.

[0207] Example 13. Photocatalytic methanol dehydrogenation experiments with Co- NC@TiO2 photocatalysts - effect of solvothermal reaction time

[0208]

[0192] Co-NC@TiC>2 photocatalysts, were prepared as described in Example 12, but with varying solvothermal reaction time. The resulting photocatalysts were evaluated for photocatalytic dehydrogenation of methanol (60%) in water, using a similar procedure as used in Example 12 and a catalyst loading of 0.2 g / L. The results are shown in Figure 20. Reducing the solvothermal reaction time from 12 hours to 7 hours increased the photocatalytic activity to about 85 mmol / g / h. Without wishing to be limited by any theory, this is ascribed to the thinner nitrogen-doped carbon shell which allows greater plasmonic enhancement of photocatalytic activity on the shell surface.

[0209]

[0193] 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.

Claims

Claims1. A particulate photocatalyst comprising: a plasmonic core comprising a metallic nanoparticle; a shell at least partially surrounding the plasmonic core, the shell comprising nitrogen-doped carbon; and a transition metal immobilised on the shell, wherein at least a portion of the transition metal is atomically dispersed.

2. The particulate photocatalyst according to claim 1 , wherein the metallic nanoparticle comprises a metal selected from the group consisting of copper, aluminium, bismuth, gold, silver and alloys thereof.

3. The particulate photocatalyst according to claim 1 or claim 2, wherein the metallic nanoparticle comprises metallic copper.

4. The particulate photocatalyst according to any one of claims 1 to 3, wherein the transition metal comprises one or more group 7 to 11 transition metals.

5. The particulate photocatalyst according to any one of claims 1 to 4, wherein the transition metal is selected from the group consisting of cobalt, nickel, copper and combinations thereof.

6. The particulate photocatalyst according to any one of claims 1 to 5, wherein the transition metal is cobalt.

7. The particulate photocatalyst according to any one of claims 1 to 6, wherein at least a portion of the transition metal is co-ordinated to nitrogen atoms of the nitrogen-doped carbon.

8. The particulate photocatalyst according to any one of claims 1 to 7, comprising the transition metal in an amount of between 0.2 wt.% and 1 wt.%.

9. The particulate photocatalyst according to any one of claims 1 to 8, wherein the shell completely encapsulates the plasmonic core, and wherein the shell has a thickness in the range of 1 nm to 15 nm.

10. The particulate photocatalyst according to any one of claims 1 to 9, wherein the nitrogen- doped carbon comprises graphitic carbon.

11. The particulate photocatalyst according to any one of claims 1 to 10, comprising particles with a distribution of particle sizes ranging from less than 40 nm to greater than 80 nm.

12. Use of the particulate photocatalyst according to any one of claims 1 to 11 in a photocatalytic process for oxidising or reducing a reactant.

13. Use according to claim 12, wherein the reactant is a hydrogen-bearing compound and the hydrogen-bearing compound is photocatalytically dehydrogenated to produce dihydrogen.

14. A method of producing a particulate photocatalyst according to any one of claims 1 to 11 , the method comprising: providing a dispersion of nanoparticles comprising a metal oxide in a liquid comprising a salt of a transition metal and a nitrogen-bearing organic compound; and thermally treating the dispersion at a temperature sufficient to (i) decompose at least a portion of the nitrogen-bearing organic compound, thereby producing nitrogen-doped carbon in shells which at least partially surround the nanoparticles, (ii) reduce the metal oxide to form metallic nanoparticles, and (iii) immobilise at least a portion of the transition metal, as atomically dispersed transition metal, on the shells.

15. The method according to claim 14, wherein the metal oxide is copper oxide.

16. The method according to claim 14 or claim 15, wherein the nitrogen-bearing organic compound comprises at least one selected from an amide, an aromatic amine and a N- heteroaromatic.

17. The method according to any one of claims 14 to 16, comprising at least one of (i) acid treating the particulate photocatalyst and (ii) thermally annealing the particulate photocatalyst under an inert atmosphere.

18. A photocatalytic method comprising: contacting a composition comprising a hydrogen-bearing compound with a particulate photocatalyst; andirradiating the particulate photocatalyst with light, thereby photocatalytically dehydrogenating at least a portion of the hydrogen-bearing compound with the particulate photocatalyst to form dihydrogen, wherein the particulate photocatalyst comprises: a plasmonic core; a shell at least partially surrounding the plasmonic core, the shell comprising nitrogen-doped carbon; and a transition metal immobilised on the shell, wherein at least a portion of the transition metal is atomically dispersed.

19. The photocatalytic method according to claim 18, wherein the plasmonic core comprises a metallic or metal oxide nanoparticle.

20. The photocatalytic method according to claim 18, wherein the plasmonic core comprises a metallic nanoparticle.

21. The photocatalytic method according to claim 20, wherein the metallic nanoparticle comprises metallic copper.

22. The photocatalytic method according to any one of claims 18 to 21, wherein the hydrogen-bearing compound is water.

23. The photocatalytic method according to any one of claims 18 to 22, wherein the composition comprising a hydrogen-bearing compound is an aqueous solution comprising NaCI, wherein water is photocatalytically dehydrogenated to produce dihydrogen.

24. The photocatalytic method according to any one of claims 18 to 22, wherein the composition comprising a hydrogen-bearing compound is an aqueous composition comprising one or more organic compounds, wherein at least one organic compound is photocatalytically dehydrogenated to produce dihydrogen.

25. The photocatalytic method according to any one of claims 18 to 24, wherein the transition metal comprises one or more group 7 to 11 transition metals.

26. The photocatalytic method according to any one of claims 18 to 25, wherein the transition metal is cobalt.

27. The photocatalytic method according to any one of claims 18 to 26, wherein the shell has a thickness of less than 3 nm.

28. The photocatalytic method according to any one of claims 18 to 27, wherein the particulate photocatalyst comprises particles with a distribution of particle sizes ranging from less than 40 nm to greater than 80 nm.

29. The photocatalytic method according to any one of claims 18 to 28, wherein the light comprises visible light in the range of 450-700 nm.

30. The photocatalytic method according to any one of claims 18 to 29, wherein the solar-to- hydrogen (STH) efficiency is at least 4% for at least 1 hour of irradiation.

Images