Field of the Invention

The present invention relates to the oxidation of gas mixtures containing methane to a product comprising one or more C1-4 alcohol, preferably ethanol.

Background

Methane is a significant carbon source for chemical synthesis due to the large reserves of shale gas and natural gas hydrate, but traditional conversion processes require high-temperature and multi-step approaches (>700K). ^{1 5} Direct methane oxidation to carbon oxygenates under mild conditions is of great significance in order to gain both economic and ecological benefits. However, the control of selectivity to avoid over oxidation is challenging in such an energetically favourable reaction. Successful activation of methane at low temperatures has been achieved using a variety of catalytic species, such as Au-Pd colloids, ⁴ Cu/Rh-Na-ZSM-5, ⁵ and copper-exchanged mordenite zeolite. ³ Nevertheless, a low methane conversion rate of <0.1% is typically observed in such processes. Hydrophobic zeolites are also reported to confine the diffusion of H2O2 oxidant while allowing the desorption of generated methanol at 70 ^" C. ⁶ However, these catalysts were only able to control selectivity towards Ci carbon oxygenates (e.g. methanol), ^{3 11} while direct C-C coupling is much more complex and challenging under such conditions. ^{12 13} A new molecular tuning strategy has been applied to controlling the selectivity of CO2 reduction towards C2 products, ¹⁴ in which organic molecules are utilised to functionalise the metal catalysts to enhance the local concentration of adsorbed intermediates and to reduce the energy barrier. ¹⁴

Covalent Triazine Framework (CTF) materials are a range of polymeric organic species that are known to behave as catalysts in some reaction environments. For example in the photocatalytic splitting of water. ¹⁵ They are also known as support materials for some metallic catalysts in heterogeneous catalytic systems. ¹⁶

The present invention has been devised in light of the above considerations.

Summary of the Invention

The present proposals provide photocatalytic methods of direct methane transformation and C-C coupling to provide a product comprising one or more C1-4 alcohol by catalytic reaction in which methane is contacted with a catalyst comprising a CTF material.

In a first aspect the present proposals provide methods of photocatalytic formation of a product comprising one or more C1-4 alcohol, the method comprising contacting a gas mixture comprising methane and one or both of an oxygenation agent and water with a catalyst comprising a CTF material under irradiation including wavelengths in the range 200-1000 nm.

The present proposals also relate to the use of a CTF material in the photocatalytic formation of a product comprising one or more C1-4 alcohol by contacting a gas mixture comprising methane and one or both of an oxygenation agent and water with a catalyst comprising a CTF material under irradiation including wavelengths in the range 200-1000 nm; and CTF materials for such a use.

The methods described herein provide significant advantages over known methods of forming a product comprising one or more C _1-4 alcohol from methane sources. Some of the benefits associated with particular embodiments of the present methods include the following.

The methods may be run as a continuous process providing the associated industrial benefits, such as simpler apparatus and lack of “down time” in the product stream.

The methods are highly selective to C1-4 alcohol, particularly ethanol, in the product stream so avoiding further post-reaction processing and simplifying separation of the alcohol (e.g. ethanol) product.

The methods operate effectively over a wide range of reaction temperatures, notably including room temperature (i.e. without additional external heating). This is in contrast to a large portion of known processes which require significantly elevated temperature and incur large additional operating costs as a result.

The methods operate effectively over a wide range of reaction pressure, notably including atmospheric pressure (i.e. without reaction pressurisation). This is in contrast to a large portion of known processes which require significantly elevated pressure and incur large additional operating costs as a result.

The methods present a single step pathway to convert methane to a product comprising one or more C1-4 alcohol. This is simpler, and therefore cheaper to implement, than many known processes which require multiple reaction steps or significant post-reaction processing of the product.

The methods may, in preferred aspects, result in low levels of CO ₂ in the product stream. This is beneficial from an environmental viewpoint and contrasts with some known processes that result in high levels of CO2.

The methods can operate effectively using air as the oxygenation agent. Many known processes require powerful oxidising agents, such as hydrogen peroxide, as the oxygenation agent. Indeed the present methods can operate effectively with those more powerful oxygenation agents but the ability to use air leads to significant cost reduction and simplification of the commercial process due to the abundance and low cost of air.

The methods use catalysts that are relatively low cost and readily available or easily made. The catalysts do not include the precious metals used in some known methods. At a gas hourly space velocity (GHSV) of 2000 hr ¹ , extremely high selectivity of 80 % to ethanol with a yield of ethanol of ca. 100 mΐtioI g ^~1 hr ¹ and unprecedented apparent quantum efficiency of ca. 7% have been achieved by the methods herein with no decay over an extended long time run.

The invention includes the combination of the aspects and preferred features described except where such a combination is clearly impermissible or expressly avoided.

Summary of the Figures

Embodiments and experiments illustrating the principles of the invention will now be discussed with reference to the accompanying figures in which:

Figure 1. PXRD pattern for CTF-1 in vacuum. The signal drop before 3° is the response of the silica capillary which is the holder of powder samples. The first peak located at ca. 7.9° is associated with the hexagonal cages. The peak at ca. 24.8 ⁰ indicates a multi-layer structure with an interplanar stacking distance of 3.5 A tested experimentally.

Figure 2. Raman spectra of CTF-1 excited by 325nm laser in atmosphere. The sharp peak located at ca. 1609 crrr ¹ is assigned as G peak, which involves sp ² carbon in the polymer materials and likely indicates a high degree of conjugation. A small peak at ca. 1400 cm ¹ is assigned as D peak and related with the sp ³ carbon in the structure.

Figure 3. ATR-FTIR spectra of CTF-1 in atmosphere.

Figure 4. ¹³ C solid state NMR spectra of CTF-1.

Figure 5. Illustration of the reaction system of the photocatalytic methane oxidation to high-value chemicals, the reactor was sealed by a PTFE screw-type jacket, the photocatalyst is packed in a bed in the circular PTFE casing.

Figure 6. Mass spectral responses of ethanol (top line, m/z = 46), water (middle line, m/z = 18) and oxygen (bottom line, m/z = 32) of the outlet gas during the selective photocatalytic oxidation of methane. Reaction conditions: gas flow rate 40 ml min ¹ , room temperature and 365nm LED irradiation.

Figure 7. Time-on-line rate of methane conversion and product generation during photocatalytic methane transformation by CTF-1. Reaction conditions: 100 W 365 nm LED irradiation, GHSV = 2000 Sr ¹ , 16:1 methane (20% methane/argon) to oxygen (water situated simulated air, 20% oxygen/nitrogen) flow ratio.

Figure 8. Apparent quantum efficiency (AQE) (crossed circles) of photocatalytic methane transformation to ethanol by CTF-1 under 365 nm LED and 450 nm LED irradiation. The UV-Vis absorption spectrum (line) of CTF-1 is superimposed for comparison. The bandgap of CTF-1 is presented as the intersect of the tangent of the absorption spectrum (dash line) with the wavelength axis. Figure 9. Scheme of the proposed reaction pathway for photocatalytic methane oxidation to ethanol in the presence of humidified air by CTF-1 catalyst.

Figure 10. Comparison of photocatalytic activity of methane transformation on CTF-1 , P-C3N ₄ and T1O2.

Figure 11. Calculated water FTIR signals on CPU-saturated CTF-1 and P-C3N ₄

Figure 12. Calorimetric measurements of methanol and ethanol competitive adsorptions over CTF-1 & P-C3N ₄ . Adsorption heat flow of methanol on ethanol-saturated catalysts’ surface (top) and ethanol on methanol-saturated catalysts’ surface (bottom).

Figure 13. The lowest energy adsorption modes for methyl radical on CTF-1 (left) and P-C3N ₄

(Right).

Detailed Description of the invention

Aspects and embodiments of the present invention will now be discussed with reference to the accompanying figures. Further aspects and embodiments will be apparent to those skilled in the art. All documents mentioned in this text are incorporated herein by reference.

Product

The methods described herein produce a product comprising one or more C _1-4 alcohol. Preferably the product comprises one or more C _2-4 alcohols. Preferably the product comprises one or more C ₂ alcohol, preferably ethanol. In preferred aspects of the present methods the product comprises ethanol and is selective for C ₂ alcohols, in particular ethanol, over other reaction products, particularly over other alcohols. The present methods are particularly effective as production methods for ethanol.

Catalyst

The methods described herein utilise a catalyst that comprises a Covalent Triazine Framework (CTF) material. Such CTF materials include triazine rings that are covalently linked together in an extended framework. Typically the triazine groups are themselves arranged in larger ring arrangements. Typically the CTF materials form pores in the extended structure. In CTF materials, the triazine rings may be linked together via intervening groups such as cyclic or aromatic units arranged between the triazine rings. Preferably the CTF catalyst used herein comprises triazine rings and cyclic or aromatic ring units linked together in an alternating extended structure, i.e. the triazine rings each linked only to the cyclic or aromatic units and vice versa. Preferably the cyclic or aromatic units are aromatic. Preferably the cyclic or aromatic units comprise six membered rings. Preferably the cyclic or aromatic units are selected from phenylene, thiophenylene, pyridinylene, bipyridinylene, carbazolylene, furanylene, benzothiadiazolylene, N-ethylcarbazolylene The cyclic or aromatic units may be substituted, for example with one or more groups selected from -CN, -IMH2, -NO2, halogen, -OH, =0, C1-6 alkyl, and -0(Ci-e alkyl). Preferably the cyclic or aromatic units are unsubstituted.

Preferably the CTF material in the catalyst has formula (I)

Formula (I) wherein Ar is selected from

0 1-6 are known by the terminology CTF-n where n = 1-6. In this same nomenclature, the CTF in which Ar is is known as CTF-0.

Preferably Ar is selected from: n = 1-6 i.e. CTF-1 , CTF-2, CTF-3, CTF-4, CTF-5, or CTF-6.

More preferably wherein n = 1-3 (i.e. CTF-1 , CTF-2, or CTF-3), more preferably wherein n = 1 , i.e. CTF-1.

CTF materials in which Ar is phenylene are typically more readily available and cheaper to make or obtain than those having more exotic Ar units, with CTF-1 being one of the cheapest and easiest to obtain thus offering distinct commercial benefits over the more expensive CTF materials. The different CTF materials demonstrate varying catalytic activity in the methods described herein. CTF-1 demonstrates a high catalytic activity and therefore presents an excellent balance of low cost with high benefit.

The catalyst used in the present methods may contain other components in addition to the CTF material. These other components may be catalytically active or may be inactive. For example they may include inert support materials. In some preferred aspects, the catalyst may comprise a polymeric semiconductor in addition to the CTF material which may extend the light adsorption range or enhance the efficiency of charge transfer and separation. For example this polymeric semiconductor component may be comprise or be selected from melon (often called “carbon nitride”), covalent organic frameworks (COFs), and metal organic frameworks. In some preferred aspects, the catalyst may comprise a metal component in addition to the CTF material. For example this metal component may comprise a metal or metal compound (e.g. metal oxide), including mixed metals and metal compounds. In those cases, the metal or metal compound may be selected from platinum, platinum oxide, ruthenium, ruthenium oxide, palladium, palladium oxide, silver, silver oxide, iridium, iridium oxide, cobalt, cobalt oxide, vanadium, vanadium oxide, rhodium, rhodium oxide, silver phosphate, iron, iron oxide, nickel, nickel oxide, cerium, cerium oxide, gold, magnesium, magnesium oxide, titanium, titanium oxide, zinc, zinc oxide, strontium, strontium titanate, strontium carbonate, bismuth, bismuth oxide, bismuth vanadate, rhenium, rhenium oxide, cadmium sulfide, cadmium selenide, molybdenum, molybdenum carbide, molybdenum sulfide, and molybdenum nitride. Preferably the metal or metal compound oxide is a metal or metal oxide, preferably a metal oxide. Preferably the metal or metal oxide is selected from platinum, platinum oxide, ruthenium, ruthenium oxide, iron and iron oxide; more preferably selected from platinum oxide, ruthenium oxide, and iron oxide; most preferably iron oxide. In situations when the catalyst comprises a metal component, the metal component may be present as a mixture with the CTF material in the catalyst, e.g. in particulate or nanoparticulate form; or may be deposited on the CTF material, e.g. deposited as nanoparticles on the CTF material.

Reaction Parameters

The methods described herein are effective to form a product comprising one or more C1-4 alcohol from a gas mixture comprising methane under a wide range of reaction parameters. One of the specific benefits associated with the present methods is the effectiveness under a wide range of reaction conditions. This differs from many of the known methods of transforming methane into ethanol which require high reaction temperature, high reaction pressure, or a combination of both. As is well known, the requirements for elevated reaction temperature and/or pressure significantly increases the commercial cost and complexity of reactions. The present methods provide significant benefits in this regard.

The present methods provide effective transformation of a gas mixture comprising methane into a product comprising one or more C1-4 alcohol across a wide range of different reaction pressures. Preferably the reaction pressure is up to 10,132,500 Pa (100 atm). Preferably the reaction pressure is between 101 ,325 and 10,132,500 Pa (1-100 atm). Preferably the reaction pressure is between 101 ,325 and 1 ,013,250 Pa (1-10 atm). More preferably the reaction pressure is between 101 ,325 and 202,650 Pa (1-2 atm). Most preferably the reaction pressure is not elevated above ambient pressure, e.g. 101 ,325 Pa (1 atm), i.e. the reaction is not performed under elevated pressure. The ability of the present methods to effectively transform a gas mixture comprising methane into a product comprising one or more C1-4 alcohol across a wide range of pressures provides a distinct benefit in terms of the versatility of the methods. In particular the effectiveness at low pressures (close to ambient room pressure) is an advantage over known reactions and the effectiveness at ambient pressure (e.g. 101 ,325 Pa) is of particular benefit due to the avoidance of the requirement for costly pressurisation of reaction vessels or reagent streams.

The present methods provide effective transformation of a gas mixture comprising methane into a product comprising one or more C1-4 alcohol across a wide range of different reaction temperatures. Preferably the reaction temperature is in the range 0 to 200°C., preferably 0 to 100°C. As the reaction typically involves water in either the reagent or product stream, it is preferably that the reaction temperature is in the range 0 to 100°C (or more preferably greater than 0 to less than 100°C) so that the water is readily saturated into the gas feedstock and can be readily separated following the reaction. Preferably the reaction temperature is in the range greater than 0 to 50°C, more preferably 15 to 50°C, more preferably 15 to 40°C, more preferably 15 to 30°C. Most preferably the reaction temperature is room temperature, i.e. the reaction is performed without any external heating. The ability of the present methods to effectively transform a gas mixture comprising methane into a product comprising one or more C1-4 alcohol across a wide range of temperatures provides a distinct benefit in terms of the versatility of the methods. In particular the effectiveness at low temperatures (close to ambient room temperature) is an advantage over known reactions and the effectiveness at ambient temperature (e.g. about 20°C) is of particular benefit due to the avoidance of the requirement for costly heating of reaction vessels or reagent streams.

In the present methods, independent alteration of the temperature and/or pressure of the reaction may be effective to alter the distribution of the reaction products. Therefore specific combinations of temperature and pressure may be particularly preferred for their beneficial effect on the reaction product, e.g. providing a greater selectivity for the desired product comprising one or more C1-4 alcohol. However, for simplicity and cost considerations, the reaction is preferably conducted without external heating or pressurisation, i.e. at ambient temperature and pressure, such as 101 ,325 Pa and 20°C. The ability of the present methods to effectively transform a gas mixture comprising methane into a product comprising one or more C1-4 alcohol across a wide range of different reaction temperature/pressure combinations provides a particular advantage over known methods in which at least one of elevated temperature and pressure is typically required with the associated increase in both cost and complexity. In particular the ability of the present methods to operate without elevation of either temperature or pressure is a significant improvement on the known methods.

The present methods provide effective transformation of a wide range of feedstock gas mixtures into a product comprising one or more C1-4 alcohol with the requirement that the gas mixture comprises methane and one or both of an oxygenation agent and water. The gas feedstock may, in some cases, comprise methane and an oxygenation agent, i.e. a methane partial oxidation reaction to form the desired product. The gas feedstock may, in some other cases, comprise methane and water, i.e. a methane “steam reforming” reaction to form the desired product. However it is preferred that the gas feedstock comprise methane and both an oxygenation agent and water. Although the present methods are effective when just one of oxygenation agent and water is present alongside methane in the gas feedstock, they demonstrate notably improved production of a product comprising one or more C1-4 alcohol (e.g. ethanol) when the gas feedstock comprises both an oxygenation agent and water alongside methane.

In gas feedstocks comprising an oxygenation agent (either alone or the preferable feedstocks that also comprise water) the ratio of methane to oxygenation agent may have an effect on the effectiveness of the conversion of methane to a product comprising one or more C1-4 alcohol. In particular this ratio may influence the selectivity of the method for a desired alcohol, e.g. ethanol in the product. Preferably the ratio of methane to oxygenation agent in the gas feedstock is up to 200:1 . Preferably the ratio of methane to oxygenation agent in the gas feedstock is in the range 1 :10 to 200:1 , preferably 1 :1 to 200:1 , preferably 1 :1 to 100:1 , preferably 5:1 to 50:1 , preferably 5:1 to 30:1 , preferably 10:1 to 20:1 , for example about 16:1 . In the present methods, gas feedstocks that have a higher content of methane may be preferred as they may result in higher selectivity for ethanol in the reaction product. Lower levels of the oxygenation agent in the feedstock are also recognised to be preferable from a safety viewpoint. However, if the content of methane is too high, this can, depending on the catalyst contact time (e.g. in some cases influenced by gas feedstock flow rate), result in a reduced methane conversion rate, for example if the catalyst contact time is low due to a high gas flow rate, the conversion rate of methane into a product comprising one or more Ci-* alcohol can be impaired.

The amount of water in the gas feedstock is not particularly limited. Typically the gas feedstock is infused with water by bubbling part or all of the gas feedstock stream through water. Preferably the gas feedstock is saturated with water.

The versatility of the present methods in terms of gas feedstock is a notable benefit. As long as the feedstock comprises methane and one or both of oxygenation agent and water, the method is resilient to the presence of many other components. For example the methane component of the gas feedstock may, in some cases, be a methane-containing biogas, such as that generated from anaerobic breakdown of organic matter. In some cases the methane component of the gas feedstock may be biogas produced from sewage, green waste, food waste, or municipal solid waste (MSW).

The oxygenation agent component of the gas feedstock is any species that transfer oxygen atoms to a substrate, in the present methods that substrate being methane. Preferably the oxygenation agent in the present methods is oxygen. However other oxygenation agents such as nitrogen oxides, hydrogen peroxide, carbon oxides (such as carbon dioxide) and sulfur oxides may also be used as the oxygenation component. The oxygenation agent may be pure oxygen. However it is preferred, for economic reasons, that the oxygenation agent component of the gas feedstock is air. This ability of the present methods to use air as the oxidant is a significant benefit over some known methods which may use stronger oxidants such as nitrogen oxides (“NO _x ”) or hydrogen peroxide. The abundance and low cost of air provides significant commercial benefits over methods using other oxidant species.

The gas feedstock may, in some aspects, be a gas mixture containing little or no additional content beyond the methane and one or both of an oxygenation agent and water. In some aspects the gas feedstock consists of methane and one or both of an oxygenation agent and water. In some preferred aspects, the gas feedstock comprises, or consists of, a methane-containing component, and one or both of water and air. In some preferred aspects, the gas feedstock comprises, or consists of, a methane- containing component, and both water and air. In some preferred aspects, the gas feedstock comprises, or consists of, a methane-containing biogas, and both of water and air.

The methods described herein may be a continuous flow process in which gas feedstock is continuously supplied to the catalyst and products removed from the reactor. Alternatively the methods described herein may be a batch process. For example a batch process wherein the gas feedstock is contacted with the catalyst for an extended period to allow the reaction to proceed before removing products along with any un reacted reagents, prior to recharging with fresh gas feedstock. Another example of a batch process may be one in which a gas feedstock is flowed into a reactor where it contacts the catalyst and the product gas stream is recycled to be repeatedly contacted with the catalyst, prior to discharge of the product. The gas mixture in the methods described herein is typically contacted with the catalyst by flowing the gas mixture over or through the catalyst (e.g. a catalyst bed). In some aspects, the gas mixture may be contacted with the catalyst at a gas hourly space velocity (GHSV) of up to 100,000 hr ¹ . The gas hourly space velocity (GHSV) is a known measurement for catalytic reactions and represents the volumetric flow rate of gas (in this case the gas mixture) per unit of catalyst mass. Typically the GHSV value is provided at a specific temperature and pressure. The GHSV values herein are provided at the ambient temperature and pressure of the reaction, for example where the method is performed at atmospheric pressure and room temperature, the GHSV values are provided at the same. Preferably the gas mixture is provided at a GHSV of up to 300,000 fr ¹ , preferably up to 100,000 hr ¹ , preferably up to 50,000 hr ¹ , preferably up to 10,000 hr ¹ , preferably in the range 100 to 10,000 hr ¹ , preferably in the range 500 to 5,000 hr ¹ , preferably 500 to 2,500 hr ¹ , preferably about 2,000 hr ¹ or about 1 ,700 hr ¹ . Higher GHSV flow rates may, in some cases, result in a reduction in the level of conversion of methane, possibly as a result of the large volume of gas not being effectively contacted with the given amount of catalyst.

The methods described herein are photocatalytic methods. Therefore the reaction mixture requires irradiation with light containing the relevant wavelength to effect the conversion of the gas mixture into ethanol. In the present methods, the relevant wavelength is in the ultraviolet to visible region of the electromagnetic spectrum, i.e. in the range 100-1000 nm. In the methods described herein, the irradiation preferably includes wavelengths in the range 100-700 nm, preferably 300-700, preferably 350- 500 nm. Preferably the irradiation includes wavelengths in the UVA region of the electromagnetic spectrum, e.g. 350-450 nm, preferably in the range 350-370 nm, such as about 365 nm. The light used to provide the irradiation in the present methods may include a range of wavelengths, including in the visible region, as long as is includes wavelengths in the relevant region to effect the catalytic conversion as discussed above. The irradiation may be provided by any suitable light source, for example, in the present methods, the irradiation may be provided by an LED light source operating at the relevant wavelength as noted above, e.g. at about 365 nm (representing the most common commercially available UVA LED light source). The intensity of the irradiation is not particularly limited but is preferably in the range up to 200 mW cm ² , preferably in the range 2 to 200 mW cm ² , preferably in the range 5-100 mW cm ² , preferably in the range 5-50 mW cm ² , preferably in the range 5-25 mW cnr ² , preferably in the range 4 to 15 mW cm ² , preferably in the range 4 to 10 mW cm ² , preferably about 7 mW cm ² .

The methods described herein may preferably demonstrate a high selectivity for a product comprising one or more C1-4 alcohol, in particular ethanol, over other carbon-containing reaction products, as measured by comparison of the mass of carbon in each component of the reaction product. For example the methods described herein preferably demonstrate a selectivity for a product comprising one or more Ci - ₄ alcohol, e.g. ethanol, over other reaction products of at least 30 % by mass of carbon, preferably at least 40 %, preferably at least 50 %, preferably at least 60 %, preferably at least 65 %, preferably at least 70 %, preferably at least 75 % by mass of carbon.

The methods described herein may preferably maintain selectivity towards methanol at a low level in the reaction product, in favour of Ci alcohols, particularly ethanol. Preferably the percentage of methanol in the reaction product (as measured by mass percentage of carbon in the reaction components) is less than 20 %, preferably less than 15 %, preferably less than 10 %, preferably less than 5 %, preferably less than 2 %, preferably less than 1 %. Preferably the percentage of methanol in the reaction product (as measured by mass percentage of carbon in the reaction components) is negligible.

While the methods described herein may produce carbon dioxide as a secondary component of the reaction product, the methods preferably produce a low level of carbon dioxide, in favour of a product comprising one or more C1-4 alcohol, e.g. ethanol. Preferably the percentage of carbon dioxide in the reaction product (as measured by mass percentage of carbon in the reaction components) is less than 50 %, preferably less than 40 %, preferably less than 35 %, preferably less than 20 %, preferably less than 15 %, preferably less than 10 %.

The selectivity of the methods described herein for a product comprising one or more C1-4 alcohol, e.g. ethanol, over the other reaction products, and the relative levels of methanol and carbon dioxide in the reaction product, may be influenced, at least in part, by the nature of the catalyst (e.g. the nature of the CTF material), the GHSV flow rate of the gas mixture feedstock, and the relative amounts of methane and oxygenation agent in the gas mixture feedstock. In particular the ratio of methane to oxygenation agent in the gas mixture feedstock is preferably in the range discussed herein which may, in preferred methods, result in a high selectivity for ethanol in the reaction product.

The methods described herein preferably have a methane conversion level (the percentage by mass of the methane in the gas mixture that is converted into reaction product during the method in a continuous flow method) above 0.05 %, preferably above 0.1 %, preferably above 0.5 %, preferably above 0.75 %, preferably above 1 %, preferably above 1 .5 %, preferably above 2 %, preferably above 3 %, preferably above 5 %, preferably above 10 %, preferably above 15 %.

The methane conversion level in the present methods may be influenced by various reaction parameters, in particular the GHSV flow rate, and the methane to oxygenation agent ratio which are preferably as defined herein to elevate the methane conversion rate.

Without being bound by theory the mechanism of reaction is thought to be as shown in Fig. 9. ¹³ C isotopic labelling tests (for example as set out in Example 14) demonstrate that an ethanol product is generated from the methane in the feedstock and no mixed ¹² C ¹³ C ethanol product is observed indicating that all of the carbon in the ethanol product originates from the methane feedstock and not from the catalyst. ¹⁸ 0 isotopic labelling tests also confirmed the role of water as a promoter in the preferred reaction when the oxygenation agent is oxygen and water is present in the feedstock (as shown in the proposed mechanism in Fig. 9).

Elucidation of the mechanism of conversion using CTF-1 leads to some general insight into mechanistic features that may also be relevant for other CTF structures. For instance, Example 15 identifies the triazine component of the CTF-1 structure as being particularly relevant for activation of methane and for splitting of water molecules. The aromatic component of the CTF-1 structure is particularly relevant for methane absorption and C-C bond formation, particularly in the formation of C2-4 alcohol products such as ethanol. Both the triazine component and the aromatic framework are conserved across the CTF structures discussed herein. Therefore there is a reasonable expectation that these mechanistic features are conserved across the different CTF materials which would lead to similar activity.

In a similar way, the selectivity of CTF-1 catalysts for ethanol over methanol in the product stream is investigated in Example 16. It is identified that CTF-1 demonstrates significantly higher binding affinity for methanol than for ethanol whereas for the extended triazine structure P-C3N4 the binding affinity higher for ethanol than for methanol. This is thought to lead to a preference for further oxidation of more tightly bound ethanol on P-C3N4 whereas the CTF material CTF-1 allows the ethanol to be released as the product more easily. This is thought to contribute to the selectivity of CTF-1 for ethanol in the product stream over methanol. The relevant mechanistic features attributable to the presence of the aromatic unit in the CTF material as compared to P-C3N4 are thought to be conserved across the CTF materials discussed herein. Therefore a similar product selectivity might be expected.

The features disclosed in the foregoing description, or in the following claims, or in the accompanying drawings, expressed in their specific forms or in terms of a means for performing the disclosed function, or a method or process for obtaining the disclosed results, as appropriate, may, separately, or in any combination of such features, be utilised for realising the invention in diverse forms thereof.

While the invention has been described in conjunction with the exemplary embodiments described above, many equivalent modifications and variations will be apparent to those skilled in the art when given this disclosure. Accordingly, the exemplary embodiments of the invention set forth above are considered to be illustrative and not limiting. Various changes to the described embodiments may be made without departing from the spirit and scope of the invention.

For the avoidance of any doubt, any theoretical explanations provided herein are provided for the purposes of improving the understanding of a reader. The inventors do not wish to be bound by any of these theoretical explanations.

Any section headings used herein are for organizational purposes only and are not to be construed as limiting the subject matter described. Throughout this specification, including the claims which follow, unless the context requires otherwise, the word “comprise” and “include”, and variations such as “comprises”, “comprising”, and “including” will be understood to imply the inclusion of a stated integer or step or group of integers or steps but not the exclusion of any other integer or step or group of integers or steps.

It must be noted that, as used in the specification and the appended claims, the singular forms “a,” “an,” and “the” include plural referents unless the context clearly dictates otherwise. Ranges may be expressed herein as from “about” one particular value, and/or to “about” another particular value. When such a range is expressed, another embodiment includes from the one particular value and/or to the other particular value. Similarly, when values are expressed as approximations, by the use of the antecedent “about,” it will be understood that the particular value forms another embodiment. The term “about” in relation to a numerical value is optional and means for example +/- 10%.

Examples

Preparation Examples

Preparation of CTF-1

The photocatalyst covalent triazine based framework (CTF)-1 was synthesised by a modified microwave- assisted approach. 10 ml trifluoromethanesulfonic acid (Sigma-Aldrich, reagent grade 98%) and 3 g terephthalonitrile (Sigma-Aldrich, 98%) were mixed in a 100 ml PTFE liner (GEM). The liner was then protected by a sleeve, sealed by CEM standard frame support module and transferred into a microwave oven (MARS, CEM). The solvothermal reaction was temperature ramped over 25 min to 115 ^" C and held there for 60 min. The microwave output power was adjusted automatically to maintain the temperature and ramp rate. After cooling to room temperature, the bulk of yellow solid was ground into particles. Then particles were washed with acetonitrile at 70 ^° C three times to remove un reacted precursor and washed with deionised water a few times until neutral to remove the acid solvent. Finally, particles were dried in a vacuum oven in glass vials at 180 ^" C to remove residual solvent and excess trifluoromethanesulfonic acid overnight.

Ft deposition on CTF-1 or Ti02

Firstly, 200 mg F PtCle^F O (Sigma-Aldrich, ACS reagent, >37.50% Pt basis) was dissolved in 10 ml deionised water. Then, in each batch, 100 mg CTF-1 or T1O2 (Millennium PC 50) was suspended in 160 ml 10 vol.% methanol/water in a 450 ml gas-tight glass reactor. 400 pL prepared chloroplatinic acid aqueous solution was added to the suspension as the platinum precursor, which roughly contains 3 wt.% Pt to CTF-1 or T1O2. After one-hour 300 W Xenon lamp irradiation (Newport), hydrogen was detected by gas chromatography equipped with molecular 5A column and TCD detector (Varian GC430). The powder colour changed to light grey and was filtered by centrifuge. Such synthesised powder was washed with deionised water five times and dried in a vacuum oven at 70 ^° C overnight. Ru deposition on CTF-1

Firstly, 200 mg RuCb-xFfeO (Sigma-Aldrich, ACS reagent, 38-42% Ru basis) was dissolved in 10 ml deionised water. Then, in each batch, 100mg CTF-1 was suspended in 160 ml 10 vol.% methanol/water in a 450 ml gas-tight glass reactor. 400 pL prepared RuCh aqueous solution was added into the suspension as the Ruthenium precursor, which roughly contains 3 wt.% Ru to CTF-1 . After one-hour 300 W Xenon lamp irradiation (Newport), the powder colour changed to light grey. It was filtered by centrifuge. Such synthesised powder was wash by deionised water for five times and dried in a vacuum oven at 70 ^° C overnight.

Fe deposition on CTF-1

Firstly, 200 mg FeCb-xFUO (Sigma-Aldrich, ACS reagent, 28% Fe basis) was dissolved in 10 ml deionised water. Then, in each batch, 100mg CTF-1 was suspended in 160 ml water in a 450 ml gas-tight glass reactor. 540 pL prepared FeCb aqueous solution was added into the suspension as the iron precursor, which roughly contains 3 wt.% Fe to CTF-1 . After one-hour 300 W Xenon lamp irradiation (Newport), the powder colour changed to dark yellow. It was filtered by centrifuge. Such synthesised powder was wash by deionised water for five times and dried in a vacuum oven at 70 ^° C overnight.

Apparatus and Testing

Photocatalytic activity tests

The photocatalytic activation was carried out in a PTFE reactor with a quartz window, irradiated by a 365 nm LED source (Beijing Perfecting technology, PLS-LED 100, l=365 nm). Photocatalysts were packed between the quartz window and PTFE body. 20% CFU/Ar (BOC), water-saturated simulated air (20% O ₂ /N ₂ , BOC, zero grade no impurities) and argon (BOC, zero grade) were used as feedstock. The gas flow rates were controlled by Bronkhorst mass flow meters from the range of 1-500 seem, respectively. The outlet gases were monitored by an Agilent 7820 gas chromatograph equipped with online injection valves, a thermal conductivity detector (TCD) for Fb, O2, N2, CO, CO2, CFU detection and a flame ionisation detector (FID) for CFU, CFI3OH, C2FI5OH detection. On-line Mass spectrometry (MS) was performed by a Hiden Analytical mass spectrometer. Isotope labelling tests were carried out by a Shimadzu gas chromatography-mass spectrometer (GCMS-QP2010 SE). ¹³ C measurements were carried out in a 60 ml quartz reactor with ¹³ CFU (Sigma-Aldrich) and simulated air (20% O ₂ /N ₂ , BOC) on humidified CTF-1 catalysts. ¹⁸ 0 measurements were carried out in the same reactor with F½ ¹⁸ 0 (Sigma- Aldrich, 99%), CFU (20% CFU/Ar, BOC) and simulated air (20% O ₂ /N ₂ , BOC), where CFU:0 ₂ = 16:1.

Apparent quantum efficiency (AQE) calculation a x amount of ethanol generated

AQE(% ) = x 100%

Total photons incident ( N )

The proposed oxidation half-reaction: H ₂ O

2CH ₄ + 2 h ⁺ - > C ₂ H ₆ + 2 H ⁺

The proposed reduction half-reaction:

C ₂ H ₆ + 0 ₂ + 2e + 2H ⁺ ® C ₂ H _s OH + H ₂ 0

Thus, transferred electrons toward ethanol generation (a) = 2

Light intensity (/) = 100 mW-cnr ²

Irradiation area (A) = 3.14 cm ²

Wavelength of the LED light source (1) = 356 nm

Planck constant ( h ) = 6.63x 10 ³⁴ J-s

Speed of light (c) = 3x 10 ⁸ m-s ¹

2 x 2 pmol min x 10 ⁶ ÷ 60 x 6.02 x 10 ²³

AQE = X 100%

365 nm x 10

100 mW-crrv ² x 10 ^-3 x 3.14 cm ² x ⁹

6.63 X 10 ³⁴ J s X 3 X 10 ⁸ m s

AQE( 365nm) = 6.97%

Catalyst characterisation

Powder X-Ray Diffraction (PXRD) measurements were carried by a SAXSLAB Ganesha 300XL small- angle x-ray scattering (SAXS) system in Wide Angle X-ray Scattering mode with a range from 2q=2°-40° (wavelength 0.154 nm, Cu-ka radiation). Attenuated total reflection Fourier-transferred infrared spectroscopy (ATR-FTIR) was collected by a Shimadzu IRAffinity-1 s spectrometer with a Specac Quest (Germanium) ATR accessory at a range from 400-4000 cm ¹ . ¹³ C cross-polarization magic angle spinning (CP MAS) solid-state nuclear magnetic resonance (ssNMR) spectra were collected at ambient temperature on a BRUKER Advance 300 WB spectrometer (Bruker UK Ltd.) with 4 mm magic-angle spinning probe. Solution NMR spectra were measured using a Bruker Avance Neo (700 MHz) and ¹ H NMR spectra were referenced to residual protiated solvent at d 7.26 (CDCh). X-ray photoelectron spectroscopy (XPS) was conducted on a Thermo Scientific XPS K-alpha machine using monochromatic Al-Ka radiation. Survey scans were collected in the range of 0-1100 eV (binding energy) at pass energy of 160 eV. Higher resolution scans were recorded for the main core lines at pass energy of 20 eV. The analysis was performed on CasaXPS software. Raman spectra were measured on a Renishaw InVia Raman Microscope, using a 325 nm excitation laser, between 100 - 3500 cm ¹ . UV-Vis absorption spectra were obtained on an Agilent Carry 3500 UV-Vis-NIR spectrophotometer fitted with an integrating sphere. Reflectance measurements were performed on powdered samples, using a standard barium sulphate powder as a reference. The reflection measurements were converted to absorption spectra using the Kubelka-Mulk transformation. Thermogravimetric analyses (TGA) were carried out at ambient conditions (25°C, 1 bar) with Setsys from Seta ram Instrument. Calorimetric measurements were carried out at 293 K using a Setaram Sensys EVO 600 DSC microcalorimeter. 50 mg of catalysts with a bed thickness of 2 mm in the sample quartz tube was degassed at 293 K for 1 h in an Arflow of 30 mL/min, then the flow was switched to an Arflow bubbled through a saturator filled with liquid methanol (ethanol) at 298 K, and the measurements were stopped when a constant heat flow was acquired. Control experiments using empty quartz tube showed negligible heat flows for both methanol and ethanol. The powder XRD spectrum of the CTF-1 material produced as described above is shown in Fig. 1 in which the signal drop before 3° is the response of the silica capillary which is the holder of powder samples. The first peak located at ca. 7.9° is associated with the hexagonal cages. The peak at ca. 24.8° indicates a multi-layer structure with an interplanar stacking distance of 3.5 A tested experimentally.

The Raman spectrum (Fig. 2) showed a high degree of conjugation of the synthesised CTF-1 material indicating a well-ordered planar structure in each layer. Fourier-transform infrared spectroscopy (FTIR) (Fig. 3) and solid-state ¹³ C nuclear magnetic resonance (NMR) (Fig. 4) confirmed the polymeric structure and that the alternating aromatic units were synthesised successfully.

In the FTIR spectrum, the two strongest peaks at 1350 cm ¹ and 1513 cnr ¹ represent the in-plane rings stretching vibrations and C-N bonds stretching mode, respectively, which indicates the successful synthesis of triazine rings. Peaks at 1024 cnr ¹ , 1160 cnr ¹ and 1272 cnr ¹ are related to C-N bonds bending vibrations also confirms the polymerisation process. Peaks located at 810 cnr ¹ , 1408 cnr ¹ and 1623 cm ¹ are associated with benzene rings and can be assigned as the bending vibrations of out-of- plane C-H bonds, the stretching vibrations and C-C bonds stretching, respectively. The small peak at 2226 cnr ¹ is resulted by the stretching of terminal -CºN groups. Additional peaks at ca. 2360 cnr ¹ and 3500-4000 cnr ¹ are associated with the surface adsorbed CO2 and water in atmosphere.

In the NMR spectrum The Peaks assigned as 6 and 7 at ca. 170 ppm are related with the carbon atoms in triazine units. The peak at 139 ppm is associated with a-carbon atoms connected triazine rings. The strongest peak at 129 ppm represents carbon atoms in the aromatic ring. The weakest peaks at the highest field are characteristic of the two types of carbon atoms in the terminating cyanide groups.

Computational methods

Periodic Density Functional theory was used to assess the relative adsorption energies of various intermediate in the pathways for CO2 conversion to either methanol or ethanol. Models of CTF-1 were compared to models of P-C3N ₄ . The adsorption energies were calculated according to the following equation:

EA E _compiex — E _ads E _{sur †}

Whereby, the adsorption energy (EA) was determined by subtracting the energies of the adsorbent molecule in the gas phase from the energies of the neutral adsorbate(s) in a vacuum (Eads) and the energy of the pristine surface (Esun) of either CIT-1 org-CN from the adsorbed complex (Ecompiex). The resulting values would determine the desorption enthalpies of the neutral species and allow an assessment of whether the selectivity was a desorption driven phenomenon. Multiple binding sites were considered for each adsorption process.

All computational values were derived with the P red ew-B u rke-E rnze rh of f u n cti 0 n a I (PBE), as implemented via the Vienna Ab initio simulation (VASP) code. This methodology was previously applied to explain the bulk properties of P-C3N4, with the models produced for that study forming a basis for the present nitride component. Plane-wave basis sets were applied to the valence electrons of each element with core electrons described by the projected augmented wave method (PAW). Long range non-bonding interactions were assessed via the Grimme D3 empirical dispersion method. A fine Monkhorst-Pack grid with /(-point 5 x 5 x 5 and 5 * 5 * 1 was used to calculate bulk and surface wavefunctions, respectively. The electronic threshold for the convergence of the self-consistency cycles (SCF) was set to 10 ⁵ eV, with the convergence determined by the Blochl smearing method. No constraints were set in any of the systems reported here, with the ionic relaxation threshold of 0.01 eV A ¹ and a plane-wave cut-off of 520 eV being applied in all cases.

EXAMPLE 1

Two direct methane transformation reactions, methane partial oxidation (CFU + O2) and steam reforming (Ch + H ₂ O), were studied in a packed bed photocatalytic reactor with continuous reactant gas flow under irradiation by 100 W LED (365 nm) without additional heating or pressurisation. The experimental apparatus is shown in Fig. 5.

The activity was first monitored qualitatively by mass spectrometry as shown in Fig. 6. Background spectra were collected from 0 to 60 min in the presence of light irradiation and continuously argon flow through the fixed bed reactor. Then the methane partial oxidation reaction was carried out by feeding the reactor with premixed methane and oxygen (time 60 min-200 min in Fig. 6). A noticeable amount of ethanol was generated during the reaction (also proved by ¹ H NMR), suggesting a coupling process of methane occurred on the CTF-1 catalyst. Water was detected as a by-product indicating an overall redox reaction as shown below

2CH ₄ + 0 ₂ ® CH ₃ CH ₂ OH (g) + H ₂ 0 (g)

Then, the gas line was switched back to argon purge for around 100 min to remove all products until all signals stabilised. Next, the steam reforming (ChU + H2O) process was investigated by feeding methane with saturated water vapour (time 310 min- 490 min in Fig. 6). A small amount of ethanol was observed in the product stream, indicating that the reaction shown below was not very efficient.

2CH ₄ + H ₂ 0(g) ® CH ₃ CH ₂ OH (g) + H ₂

EXAMPLE 2

Using the same experimental set-up as in Example 1 , the gas feedstock was switched to contain both water and oxygen (time 490 min - 600 min in Fig. 6). The generation of ethanol was significantly increased over that seen in Example 1 by a factor of ca. 7 compared with that in the presence of oxygen alone, suggesting that water greatly promotes the reaction. A dramatic enhancement in the water signal indicates that it was also produced as a by-product. Thus, both water and oxygen are preferred in the present methods to drive the process efficiently.

EXAMPLES 3-6

Using the experimental set-up as in Example 1 , the photocatalytic activity of methane transformation on CTF-1 in a packed bed flow reactor under 365 nm LED light irradiation at different methane:oxygen ratios was investigated. Results are presented in Table 1 .

Methane Product selectivity based on a a GHSV (gas hourly space velocity) was calculated by the gas flow rate detected by flow meters over the reactor volume (1.6 cm ³ ). ^b CH ₄ source is 20% ChU/Ar and O ₂ source is simulated air (20% O2/N2). ^c The value is the average methane conversion rate during 4 h light irradiation. The error is the calculated standard deviation. ^d The value is the average product selectivity during 4 h light irradiation. The error is the calculated standard deviation over 3 hours (9 measurements).

Table 1

Examples 3-6 in Table 1 represent ratio change of methane to air with a fixed gas hourly space velocity (GHSV) of 2000 hr ¹ . The methane conversion rate decreased as the concentration of oxygen was reduced and the selectivity towards ethanol enhanced with increased methane concentration. The highest methane conversion rate was 6.1% (Example 3), continuously generated at a ratio of CH ₄ :0 ₂ =1 :1. However, under those conditions the major product was carbon dioxide, which was less desirable. When the CH ₄ :0 ₂ ratio increased to 4:1 (Example 4), the ethanol selectivity was improved to ca. 40%. Further increasing the methane concentration to CH ₄ :0 ₂ =16:1 (Example 5) led to the highest ethanol selectivity of 78.6% with a methane conversion rate of 1.7%, which is comparable to the reported benchmark conversion rate using known catalytic systems operated at high temperature (750-1000°C) and high pressure, and, notably, more than tenfold higher than that reported by heterogeneous catalysis operated at medium temperature and/or pressure. ^{3 5 17}

Further increasing the CH ₄ :0 ₂ ratio (Example 6) resulted in no significant change in ethanol selectivity but the methane conversion rate was more than halved, compared with the 16:1 (CH ₄ :0 ₂ ) of Example 5. Without being bound by theory, this may be due to the reduced mass transfer of oxygen to the catalyst surface. However, although the selectivity towards ethanol increased, the methane conversion decreased with reduced oxygen concentration. This may be because a higher oxygen amount accelerated the reaction rate and also resulted in the further oxidation process. EXAMPLES 7-9

Using the experimental set-up as in Example 1 , the photocatalytic activity of methane transformation on CTF-1 in a packed bed flow reactor under 365 nm LED light irradiation at different GHSV flow rates for the feedstock gas was investigated. Results are presented in Table 2.

Methane Product selectivity based on

GHSV ³

Q. conversion ⁰ carbon ^d / %

£ Photocatalyst I CH ₄ :0 ₂ ^b ra x I

LU h ¹ C2H5OH CH3OH CO2

%

1.7 78.6 8.2

5 CTF-1 1700 16:1

(±0.1) (±1.7) (±0.5)

2.5 60.4 35.1

7 CTF-1 850 16:1

(±0.3) (±4.0) (±6.9)

0.3 66.9 13.0

8 CTF-1 5080 16:1

(±0.02) (±7.5) ^"" (±5.4)

0.1 60.9 10.9

9 CTF-1 8500 16:1

(±0.01) (±28.6) ^” (±4.2) ^a GHSV (gas hourly space velocity) was calculated by the gas flow rate detected by flow meters over the reactor volume (1 .6 cm ³ ). ^b CH ₄ source is 20% CFU/Ar and O2 source is simulated air (20% O2/N2). ^c The value is the average methane conversion rate during 4 h light irradiation. The error is the calculated standard deviation. ^d The value is the average product selectivity during 4 h light irradiation. The error is the calculated standard deviation over 3 hours (9 measurements).

Table 2

The system was further evaluated using the 16:1 methane:oxygen ratio identified as preferred in Example 5. The effects of the flow rate at this ratio of reagents were investigated at different GHSV from 850 to

8500 hr ¹ as shown in Table 2, Examples 5 and 7-9. The methane conversion rate of 1 .7 generated under GHSV of 1700 hr ¹ (Example 5) increased slightly to 2.5 at lower GHSV of 850 h ¹ (Example 7). However under those conditions, the selectivity towards ethanol was slightly lower at 60.4% and the production of CO2 was notably higher.

The methane conversion rate dropped significantly to 0.3% and 0.1% when increasing the GHSV to 5080 and 8500 h ¹ , Examples 8 and 9, respectively. The selectivity towards ethanol remained almost constant at ca. 60-70%.

Therefore the GHSV of 1700 hr ¹ (Example 5) represented a good balance between methane conversion rate, ethanol selectivity, and low CO2 production.

COMPARATIVE EXAMPLES 1-4

Four comparative experiments were also conducted, as shown in Table 3. No methane transformation or ethanol generation was detected in the absence of either photocatalyst (Comparative Examples 1-3) or methane (Comparative Example 4). These data indicate that the ethanol product is derived from the methane feedstock and is driven by photocatalysis over CTF-1 .

Methane Product selectivity based on

GHSV ^a

CL conversion ⁰ carbon ^d / % E E Photocatalyst I CH ₄ :02 ^b

O ns / O LU X h- ¹ C2H5OH CH3OH CO2

%

0.8 86.8

1 Ti0 ₂ 1700 16:1

(±0.06) (±11.6)

0.6 46.1 45.67

2 P-C3N4 1700 16:1

(±0.2) (±0.7) (±0.4)

3 S1O2 1700 16:1 0

4 CTF-1 1700 100% Air - 0 0 0 ^a GHSV (gas hourly space velocity) was calculated by the gas flow rate detected by flow meters over the reactor volume (1 .6 cm ³ ). ^b CH ₄ source is 20% CFU/Ar and O2 source is simulated air (20% O2/N2). ^c The value is the average methane conversion rate during 4 h light irradiation. The error is the calculated standard deviation. ^d The value is the average product selectivity during 4 h light irradiation. The error is the calculated standard deviation over 3 hours (9 measurements).

Table 3

It can be seen that only CO2 was generated when using T1O2 as a photocatalyst (Comparative Example 1), consistent with previous reports and P-C3N4 only achieved 20 mhioI g ^~1 hr ¹ ethanol production rate with an equal selectivity of 46% to methanol and ethanol (Comparative Example 2). Contrast this with, CTF-1 as seen in Example 5 which represents a five times higher ethanol production rate with 79% selectivity.

EXAMPLES 10-12 AND COMPARATIVE EXAMPLE 5 Cocatalysts were decorated onto the photocatalysts to investigate the effect on the photocatalytic activity. Results are shown in Table 4. Platinum species as co-catalysts were first loaded onto T1O ₂ and CTF-1 photocatalysts. The chemical states of the platinum oxide species were analysed by XPS. Peaks located at 73 eV and 76.5 eV are from the Pt 4/7/ ₂ and Pt 4/5/2, respectively. On Ti0 ₂ most of surface Pt species being Pt(ll). On CTF-1 Pt species are Pt (II) and Pt (IV) states with the Pt (II) to Pt (IV) ratio being nearly 1 :1.

Methane Product selectivity based on

GHSV ^a . conversion ⁰ carbon ^d / % Photocatalyst I CH ₄ :0 ₂ ^b

/ h ¹ C2H5OH CH3OH CO2

%

5

3.3 96.1

(Comp 3wt.% PtOx/Ti0 ₂ 1700 16:1

(±0.1) ^{"" ""} (±3.5) Eg.)

2.3 79.6 11.8

10 3 wt.% PtOx/CTF-1 1700 16:1

(±0.1) (±7.0) ^"" (±1.2)

1.5 72.2 9.8

11 3 wt.% RuOx/CTF-1 1700 16:1

(±0.1) (±7.5) ^"" (±1.1)

2.0 83.1 12.8

12 3 wt.% FeOx/CTF-1 1700 16:1

(±0.1) (±2.5) ^~~ (±0.8) ^a GHSV ( ^" gas hourly space velocity) was calculated by the gas flow rate detected by flow meters over the reactor volume (1 .6 cm ³ ). ^b CH ₄ source is 20% ChU/Ar and O2 source is simulated air (20% O ₂ /N ₂ ). ^c The value is the average methane conversion rate during 4 h light irradiation. The error is the calculated standard deviation. ^d The value is the average product selectivity during 4 h light irradiation. The error is the calculated standard deviation over 3 hours (9 measurements).

Table 4

Platinum decorated anatase (T1O2) demonstrated more than four times higher methane conversion rate compared to the bare sample (Comparative Example 5 c.f. Comparative Example 1), but the major product was CO2 and no ethanol was produced. Example 10 demonstrates that the co-catalyst 3 wt.% PtOx/CTF-1 provided a conversion rate of methane that is increased by nearly 50% over the raw CTF-1 catalyst (Example 5) while the excellent selectivity to ethanol was maintained at ca. 80% compared with bare CTF-1 under identical conditions.

The major ruthenium species were confirmed as RuO _å by XPS. However, on testing under identical conditions, Example 11 shows that both the methane conversion rate and ethanol selectivity decreased slightly.

Iron decorated CTF-1 was also tested under identical conditions and Example 12 demonstrates excellent methane conversion (similar to Example 10 and superior to Example 5) and maintenance of excellent ethanol selectivity with only a minor increase in CO2 production.

EXAMPLE 13

The stability of the catalyst was studied and results are shown in Fig. 7 and Fig. 8.

The catalyst was first stabilised by absorption of water-saturated 16:1 CH4/O2 gas mixture with a GHSV of 2000 hr ¹ in the dark. Under 100 W 365 nm LED irradiation, methane was continuously converted, which was stable for around 12 h, resulting in 7200 mhhoΐbb methane converted per gram of catalyst. Only two products ethanol and CO2 were monitored and the ethanol selectivity was almost constant at ca. 80%.

On turning off the light source, CFU returned to initial concentration and the product amounts dropped to zero.

The photocatalyst presented similar chemical and crystalline structures after the 12 h test as demonstrated by FTIR and Raman spectroscopy. Therefore, the photocatalysts were shown to be stable under the reaction conditions for an extended reaction period.

The quantum efficiency was calculated as shown in Fig. 8 (see details above) assuming at least two electrons were required for one ethanol molecule to be produced from methane. The apparent quantum efficiency (AQE) was 7% at 365 nm, which might be underestimated as multi-electron could be involved. The UV-Vis absorption spectrum showed a band edge at E _g = 2.7 eV (ca. 450 nm) and a 450 nm LED light source was then used to drive the reaction with an AQE of ca. 0.5% at 450 nm, indicating a photocatalytic process.

EXAMPLE 14

The reaction mechanism and carbon source in the products were investigated by isotope labelling tests. ¹³ CH ₄ was first utilised to identify the carbon source for ethanol production. The most significant mass spectrometry peak at mass/charge (m/z) = 31 under ¹² C conditions which is assigned as - ¹² CH ₂ 0H ⁺ fragment ions shifted to m/z = 32 (- ¹³ CH ₂ 0H ⁺ ) when the feed source was switched to ¹³ CH ₄ . The second strongest peak at m/z = 45 represented - ¹² CH3 ¹² CH20 ⁺ and it shifted to m/z = 47 (- ¹³ CH3 ¹³ CH20 ⁺ ). All other peaks also shifted to the higher mass/charge ratio with constant relative intensities. Thus, ethanol was generated from methane during the photocatalytic process. More importantly, there was no fragment detected at m/z = 46 associated with the ¹² CH3 ¹² CH ₂ 0H ⁺ when using ¹³ C isotope labelled methane as a reactant, which strongly suggested that all the carbon in the ethanol was from methane, not from the polymer photocatalyst itself, providing further evidence for the stability of the polymer photocatalyst. In addition, the total ion chromatogram (TIC) spectrum showed one peak located at ca. 1.7 min assigned to CO2. The clear signal at ca. 6.5 min was assigned to water and the last peak at ca. 12.9 min to ethanol, confirmed by the associated mass spectrum. An extra peak at ca. 2.65 min was identified as ethane by the mass spectrum, which was attributed to the intermediates during the reaction.

To further confirm the function of water as a promoter of the reaction, ¹⁸ 0 labelled water was employed. Mass spectrometry indicated that all of the oxygen atoms in the produced ethanol with and without the presence of isotope labelled water were identical, detected as ¹⁶ 0, which came from the oxidant ¹⁶ C>2. This indicates that the ethanol was most likely generated via a radical reaction process with photocatalytically reduced O2 species. However, CO2 as the over oxidation product was different from ethanol. When using H ₂ ¹⁸ 0 to humidify the feed gas, the majority of oxygen atoms detected in the CO2 were ¹⁸ 0 atoms. The ratio of C ¹⁶ C>2: ¹⁸ 0=C= ¹⁶ 0: C ¹⁸ C>2 was ca. 1 : 3: 6. Thus, oxygen played a minor role in CO2 generation over the CTF-1 catalyst, which might be from overoxidation of intermediates (e.g. methanol). Indeed, when using H ₂ ¹⁸ 0 and ¹⁶ C>2 in the feed gas, H ₂ ¹⁶ 0 was also detected. This indicated that oxygen reduction occurred and provided further evidence that reduced oxygen species were present.

This process was also identified by gas adsorption. The mass change when using a water and methane mixture as the feed gas equalled to the sum of the mass changes when using either water or methane as the feed gas, respectively. Therefore, water molecules did not compete with methane for adsorption. Furthermore, the mass amount of water adsorbed nearly doubled that of the adsorbed methane, implying more water molecules were adsorbed on the polymer photocatalyst than methane, taking into account their similar molar masses. As shown in Fig. 9, the function of the water molecules is proposed to be to hydroxy late the catalyst surface via an oxidation reaction with photogenerated holes (H ₂ 0 + h ⁺ ® OH ^* + H ⁺ ). The presence of absorbed -OH radicals enhanced the ability of the C-H bond cleavage and the generation of -CH3 radicals. After that, further C-H bond cleavage occurs to generate CO2. On the other hand, two -CH3 radicals recombined to form a C2H6 molecule as observed by GC-MS. C2H6 then further reacts with reduced dioxygen species generated by photoelectrons in the CTF-1 , thus generating both ethanol and water. As a competitive reaction, methane is also likely to react with reduced oxygen species to form methanol that favours overoxidation to CO2 as indicated by the isotopic measurement.

EXAMPLE 15

To further investigate the mechanism on the high selectivity towards ethanol rather than Ci products (e.g. CO2 or methanol) over CTF-1 catalyst, the photocata lytic activity over two widely used photocatalysts, bare anatase T1O2 and polymeric carbon nitride (P-C3N ₄ ) were observed as shown in the Table 3 and Fig. 10. Both photocatalysts only achieved half of the methane conversion rate of CTF-1. Furthermore, only CO2 was generated when using T1O2 as a photocatalyst, consistent with previous reports. P-C3N ₄ presented a higher selectivity to methanol than ethanol, consistent with previous reports. The CTF-1 catalyst represented the highest conversion and more importantly 5 times yield of ethanol compared with p-CsIS The in-situ adsorption of feedstocks (ChU, O2 and H2O) and the desorption of products (methanol and ethanol) were analysed by diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) and electron spin resonance (ESR) under light irradiation to elucidate the reason for such high conversion and selectivity to ethanol. These were complemented by the isothermal desorption, calorimetric measurements and first-principles calculations. The groups of DRIFTS peaks located at ca. 3020 crrr ¹ were assigned as physisorbed CFU. The signal of ChU saturated P-C3N ₄ was much weaker than that of CTF-1 and T1O2, suggesting high surface adsorption of methane molecules on CTF-1 and T1O2. Meanwhile, CTF-1 sample presented a special peak located at ca. 1541 crrr ¹ , which was negligible on T1O2 and P-C3N ₄ . The additional methane adsorption peak on CTF-1 could be assigned as the degenerate deformation of methane molecules (chemical adsorbed), which was infrared inactive for a highly symmetric methane molecule. Such results were consistent with the calculated infrared signals, which indicate that benzene units in CTF-1 play a role in the enhanced physical and chemical adsorption of ChU. It further indicates that C-H bonds in methane were able to be activated by the hydrogen bonding interaction between benzene units and methane molecules. Both experiment and modelling results were consistent with the high conversion of methane over CTF-1 compared to the other two photocatalysts.

Isothermal desorption and calorimetric measurements of O2 over 50 mg Cksatu rated CTF-1 , P-C3N ₄ and T1O2 at 25°C show that CTF-1 adsorbs the smallest amount of O2 gas among all samples. When considering the calorimetric spectra, O2 desorption required the highest energy of ca. 27 kJ/mol over CTF-1 sample, while 9 kJ/mol and 6 kJ/mol were needed over P-C3N ₄ and T1O2, respectively. Thus, the activation of adsorbed O2 species was more favoured on CTF-1 than other samples. In-situ ESR under light irradiation indicates the smallest amount of O2" species on CTF-1 , agreement with the smallest adsorbed O2 molecules, which contributed to ethanol production while also facilitate overoxidation of methanol to CO2 as suggested by the isotopic labelling experiments. As both ^« Ch-h radicals could be efficiently generated on the CTF-1 surface while O _2' species concentration formed was lowest, there were more opportunities to couple ^« ChU to C2 products on the CTF-1 polymer photocatalyst compared to opportunities for overoxidation to the less preferred CO2 product.

DRIFTS spectra present the adsorption of H ₂ O molecules on different catalysts. CTF-1 had the highest capability for water adsorption and the majority was in the form of chemisorbed species as suggested by the DRIFTS peak located at ca. 3640 crrr ¹ of hydroxyl groups rather than molecular water. Water saturated T1O ₂ presented a broad peak at ca. 3200 crrr ¹ indicating strong physisorption of molecular water, while P-C3N ₄ showed the weakest adsorption of water.

Modelling results, as shown in Fig. 11 , indicate that one H atom in water molecule is bonded with the benzene unit and the other is just able to interact with the triazine unit due to hydrogen bonding. Thus, much stronger N-H infrared response is observed both experimentally and by modelling on water- saturated CTF-1 than that on water-saturated P-C3N ₄ . Therefore, the structure of CTF-1 leads to particularly enhanced water adsorption. Photogenerated holes are thought to be mainly accumulated on the triazine units, such that the interaction between triazine unit and water results in more efficient water oxidation, which is consistent with the acceleration of the methane conversion rate when adding water as the promoter. In in situ DRIFT of water adsorption under light irradiation, a broad peak on T1O ₂ sample shifted from ca. 3200 cm ¹ to ca. 3400-3500 cm ¹ , indicating the generation of -OH radicals adsorbed on the surface of T1O ₂ . CTF-1 sample showed a negative peak of at ca. 3640 cnr ¹ , suggesting the efficient desorption of ·OH radicals generated under light irradiation, which to some extent shed light on the activeness of ·OH radicals on CTF-1 . While P-C3N ₄ did not present a noticeable difference with and without light irradiation.

Such results were further confirmed by ESR. T1O ₂ had the highest generation rate of ·OH radicals, which was consistent with the high selectivity towards CO ₂ . Meanwhile, CTF-1 presented medium signal of -OH radicals among three samples, which indicates that water oxidation to ·OH radicals as a promoter on CTF-1 was much more efficient than P-C3N ₄ , but weaker than T1O ₂ , thus avoiding overoxidation observed over T1O2 .

EXAMPLE 16

To investigate the selectivity for ethanol over methanol, the relative binding energy between the catalyst surface and the two potential products (methanol & ethanol) were studied by calorimetric measurements and the first principles calculations. Calorimetric measurements when introducing methanol on ethanol- saturated CTF-1 and P-C3N4 measure the overall heat flows represented the heat released for methanol adsorption and ethanol desorption. The total energies were positive (exothermic) during methanol adsorption with simultaneous desorption of ethanol on both catalysts while it was much larger on CTF-1 (see Fig. 12). Whereas ethanol adsorption with methanol desorption was negative (endothermic) on CTF- 1 only. Therefore, methanol had stronger binding energy with the CTF-1 catalyst compared to ethanol, vice versa on P-C3N ₄ , indicating that ethanol can easily desorb after generation while methanol is strongly bound to the surface of the catalyst, which changes the selectivity of the products and might also be the reason of CO2 production as discussed below.

Modelling results also confirmed that the methane adsorption is stronger on CTF-1 (-116 kJ/mol) compared to a P-C3N ₄ surface (-61 kJ/mol) (see Fig. 13), while the ethanol adsorption was far more exothermic on p- C3N ₄ (-40 kJ/mol) than on CTF-1 (-18 kJ/mol). Thus, the greater selectivity towards ethanol over CTF-1 than P-C3N ₄ is likely driven by the desorption processes of products.

To experimentally observe the subsequent reaction of the strongly adsorbed methanol species on CTF-1 , excess methanol molecules were added into the feed gas source. The ethanol generation showed no obvious change, but the CO ₂ generation rate greatly increased. Thus, methanol was preferably overoxidised to CO2 rather than reacting with methyl radicals to form ethanol on CTF-1 catalyst. It also indicated that different activation sites exist for ethanol generation and methanol over-oxidation, which was consistent with the observation that the majority CO2 was due to the ·OH radicals from oxidised water and ethanol was formed by O2’ . Therefore, the photocatalytic transformation of methane to ethanol was likely via a multi-electron pathway. Herein CTF-1 is thought to be first excited by photons to generate a pair of electron and hole. Electron reacted with oxygen to form some reduced oxygen species. Hole reacted with water to form ·OH radicals, which had much stronger oxidation ability to break C-H bond, thus forming *CH3 radicals. Two methyl radicals could recombine to generate ethane and was further oxidised by reduced O2 species (e.g. O2 ) to form ethanol with a by-product of water. Due to the strong oxidation ability of ·OH radicals, methyl radicals could be further oxidised to methanol and then generate CO2. Therefore it is thought that the function of water may be to promote the yield of ^« Ch-h radicals and the function of O2 may be to accept electrons and then selectively convert ethane to ethanol together with regeneration of surface by a reaction with the proton to form water.

References

A number of publications are cited above in order to more fully describe and disclose the invention and the state of the art to which the invention pertains. Full citations for these references are provided below. The entirety of each of these references is incorporated herein.

1 . Guo, X. et al. Direct, nonoxidative conversion of methane to ethylene, aromatics, and hydrogen. Science 344, 616-619 (2014).

2. Meng, X. et al. Direct Methane Conversion under Mild Condition by Thermo-, Electro-, or Photocatalysis. Chem 5 2296-2325 (2019).

3. Sushkevich, V. L, Palagin, D., Ranocchiari, M. & Bokhoven, J. A. van. Selective anaerobic oxidation of methane enables direct synthesis of methanol. Science 356, 523-527 (2017).

4. Agarwal, N. et al. Aqueous Au-Pd colloids catalyze selective CH 4 oxidation to CH 3 OH with O 2 under mild conditions. Science 358, 223-227 (2017).

5. Shan, J., Li, M., Allard, L. F., Lee, S. & Flytzani-Stephanopoulos, M. Mild oxidation of methane to methanol or acetic acid on supported isolated rhodium catalysts. Nature 551 605-608 (2017).

6. Jin, Z. et al. Hydrophobic zeolite modification for in situ peroxide formation in methane oxidation to methanol. Science 367, 193-197 (2020).

7. Tomkins, P., Ranocchiari, M. & Van Bokhoven, J. A. Direct Conversion of Methane to Methanol under Mild Conditions over Cu-Zeolites and beyond. Acc. Chem. Res. 50, 418-425 (2017).

8. Williams, C. et al. Selective Oxidation of Methane to Methanol Using Supported AuPd Catalysts Prepared by Stabilizer-Free Sol-Immobilization. ACS Catal. 8, 2567-2576 (2018).

9. Cui, X. et al. Ro 0 m-Te m pe rat u re Methane Conversion by Graphene-Confined Single Iron Atoms. Chem 4, 1902-1910 (2018).

10. Xie, J. et al. Highly selective oxidation of methane to methanol at ambient conditions by titanium dioxide-supported iron species. Nat. Catal. 1 , 889-896 (2018).

11 . Yu, X., De Waele, V., Lofberg, A., Ordomsky, V. & Khodakov, A. Y. Selective photocatalytic conversion of methane into carbon monoxide over zinc-heteropolyacid-titania nanocomposites. Nat. Commun. 10, 700 (2019).

12. Zhou, Y., Zhang, L. & Wang, W. Direct functionalization of methane into ethanol over copper modified polymeric carbon nitride via photocatalysis. Nat. Commun. 10, 506 (2019).

13. Okolie, C. et al. Conversion of Methane into Methanol and Ethanol over Nickel Oxide on Ceria- Zirconia Catalysts in a Single Reactor. Angew. Chemie Int. Ed. 56, 13876-13881 (2017).

14. Li, F. et al. Molecular tuning of CO 2 -to-ethylene conversion. 577, 509-514 (2020).

15. Tang et al., ACS Catal., 2019, 9, 9, 7697-7707

16. Tahir et al., Polymers (Basel) 2019, 11 (8), 1326 17. de Klerk, A. Engineering evaluation of direct methane to methanol conversion. Energy Sci. Eng. 3 60-70 (2015).