FERTILISER COMPOSITION

FIELD OF THE INVENTION

The invention relates to fertiliser compositions and in particular to controlled release fertiliser compositions.

BACKGROUND OF THE INVENTION

Fertiliser compositions are used to provide a plant growing medium with fertiliser. A fertiliser is a substance that comprises one or more plant nutrients and, when in use, is capable of providing the plant growth medium with such plant nutrients. A plant nutrient is a chemical element essential for plant growth. Essential chemical elements must be present in the soil in appropriate amount and proportions to provide a balanced contribution to the plant growth.

Basic forms of fertiliser compositions consist of 'naked' or 'pure' fertilisers, usually highly water-soluble salts, provided directly into the plant growth medium.

However, the excessive solubility of the pure fertilisers in water limits their use in a pure form. In the case of highly moist soils, for example, pure fertilisers may provide an oversupply of chemical elements which can be toxic to the plant and compromise plant growth. On the other hand, flowing water (e.g. rain water or irrigation water) filtering down through the plant growth medium can leach the nutrients away from the plant roots, effectively depriving the plant of their nutrients.

To minimise leaching of nutrients away from the plant roots, fertilisers may be provided as fertiliser compositions in the form of pellets, prills, granules or single crystals encapsulated within a coating layer. The coating layer may be made of elemental sulphur, 'sulphur plus polymer' multilayers, wax polymeric materials or polymeric/polyolefin materials. Examples of fertiliser compositions of this kind are given in KR946937 or WO2012048276.

The coating layer reduces water penetration through the layer, thereby reducing the dissolution rate of the fertiliser and consequently the leaching of chemical elements away from the plant roots. However, coated fertilisers are limited by the poor mechanical stability of the coating layer. Mechanical mixing of the coated pellets, prills, granules or single crystals during production, handling and/or storage operations may damage the coating layer, effectively exposing the fertiliser to leaching when the fertiliser composition is in use. In addition, conventional fertiliser compositions are inherently incapable of meeting a variable plant nutrient demand because the release rate of the fertiliser depends on a multiplicity of environmental factors (e.g. soil moisture, temperate, pH, etc.), resulting in unpredictable release rate of the fertiliser. There remains therefore an opportunity to provide fertiliser compositions that address one or more problems of prior art compositions, or provide a useful alternative.

SUMMARY OF THE INVENTION The present invention provides a controlled release fertiliser composition comprising a Metal Organic Framework (MOF), the MOF incorporating within its framework fertiliser.

Without wishing to be limited by theory, the fertiliser compositions according to the present invention are believed to release fertiliser into a plant growth medium (e.g. soil or soil-less media such as sphagnum peat moss, bark, sand, vermiculite, perlite, or any combinations thereof) by a different mechanism than conventional fertiliser compositions. Specifically, the compositions according to the invention are believed to offer control over the diffusion rate of fertiliser from within the MOF framework into the growth medium. As a result, the release of fertiliser can be advantageously more predictable and tailorable than in conventional fertiliser compositions because it is predominantly determined by fewer factors, principally the nature of the MOF.

MOFs are hybrid coordination structures formed by metal clusters comprising metal ions, e.g. metal ions or metal oxides, coordinated by multi-functional organic ligands. This results in the formation of one-, two- or three-dimensional structures that can be highly porous. The porosity of MOFs can be visualised as a spatial arrangement of cavities in the form of cages connected by channels. Depending on the particular choice of metal ions and organic ligands, MOFs having cavities in the form of open micro- and meso-pores are available.

Fertiliser incorporated within the framework of the MOF can diffuse to the outside of the MOF according to diffusion rates that strictly depend on the specific porosity characteristics of each MOF. The possibility to modify the chemistry of the pores advantageously adds a further parameter that may be tuned to determine specific diffusion rate of species incorporated within a MOF framework.

As each MOF presents a unique porosity structure, fertiliser compositions having a specific release rate of fertiliser can be obtained by associating a particular MOF with a given fertiliser.

As diffusion is a concentration-driven process, low concentration of nutrients would be a primary factor regulating the release of fertiliser from the inside of the MOF framework with the presence of moisture. This is in contrast with conventional fertiliser compositions, which release of fertiliser is less predictable because it is driven by a variety of concurring environmental factors. A particular advantage offered by the fertiliser compositions of the present invention is that variable (and mostly unpredictable) rainfall levels do not substantially affect the release rate of fertiliser, thus reducing loss of fertiliser due to leaching of nutrients away from the plant roots. Accordingly, the possibility to tune the release of fertiliser almost independently from a variety of concurring environmental factors allows for a less frequent re-fertilisation of the growth medium.

Further, a given fertiliser incorporated within the framework of different MOFs can be released at different rates, thereby providing for a fertiliser composition which release of a given fertiliser can be tailored to meet a changing demand of plant nutrients.

Providing a mixture of different MOFs each containing a given fertiliser can also allow for fertiliser compositions that can be tailored to release different fertilisers at predetermined stages of plant growth.

The present invention also provides a method for producing a controlled release fertiliser composition, the method comprising incorporating fertiliser into MOF. This may be achieved, for example, by either infiltrating fertiliser into the framework of pre-formed MOF, or by providing MOF precursors and fertiliser together in a solution and subsequently promote formation of MOF that encapsulates the fertiliser.

Further advantages offered by the fertiliser compositions of the present invention reside in their chemical and mechanical stability under moist conditions. The high porosity of MOFs also ensures a high loading capacity. In addition, the MOF component of the present invention is cheap to produce.

In one embodiment, the MOF is made to be biodegradable. A biodegradable MOF can itself deliver trace metal nutrients and sugars that make up the MOF framework.

Also provided is a method of fertilising a plant growth medium, the method comprising adding a fertiliser composition according to the present invention to the plant growth medium.

The invention also provides a method of increasing plant growth rate in a growth medium, the method comprising adding a controlled release fertiliser composition according to the present invention to the plant growth medium.

Further aspects and embodiments on the invention are outlined below. BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments of the invention will be now described with reference to the following non- limiting drawings, in which: Figure 1 shows the equilibrium concentration of nitrogen in an aqueous solution of ammonium nitrate fertiliser containing MOF;

Figure 2 shows the equilibrium concentration of nitrogen in an aqueous solution of urea fertiliser containing MOF;

Figure 3 shows nitrogen release profiles measured for pure ammonium nitrate fertiliser and MOFs incorporating ammonium nitrate fertiliser. MOFs are CAU-10-OH, CAU-IO-NO ₂ , UiO-66, UiO-66-NH ₂ ; Figure 4 shows nitrogen release profiles measured for pure urea fertiliser and MOFs incorporating urea fertiliser. MOFs are CAU-10-OH and U1O-66-NH ₂ ;

Figure 5 shows nitrogen release profiles of pure urea fertiliser, pure ammonium nitrate fertiliser, urea and ammonium nitrate fertilisers incorporated within CAU-IO-NO ₂ and CAU-10-NH ₂ MOFs; Figure 6 shows the nitrogen balance for pure fertiliser and fertiliser incorporated within MOFs, as measured in wheat growth experiments in term of amount of nitrogen (i) remaining in the soil after plant growth, (ii) taken up by wheat plants, (iii) leached out of the soil sample;

Figure 7 shows the differential amount of nitrogen leached from pure ammonium and nitrate fertilisers and ammonium nitrate fertilisers incorporated within MOFs. Nitrate- nitrogen is the top portion of the bars (green) and ammonium-nitrogen is the bottom portion (red);

Figure 8 shows the ammonium nitrate-nitrogen and urea-nitrogen balance for the pure fertilisers and the fertilisers incorporated within MOFs, as measured in wheat growth experiments in terms of amount of nitrogen (i) remaining in the soil after plant growth, (ii) taken up by wheat plants, (iii) leached out of the soil (A.N. abbreviates ammonium nitrate);

Figure 9 shows the differential amount of nitrogen leached from pure ammonium nitrate and urea fertilisers and ammonium nitrate and urea fertilisers incorporated within MOFs (red and green respectively), and total nitrogen leached (purple) based on the applied amount of nitrogen; and

Figure 10 visualises the plant nitrogen uptake relative to applied nitrogen of pure fertilisers and experimental fertiliser compositions according to embodiments of the invention. The blue (upper) diagonal line represents 100% efficiency, while the red (lower) line confines the performance of conventional fertiliser.

Some Figures contain colour representations or entities. Coloured versions of the Figures are available upon request. DETAILED DESCRIPTION OF THE INVENTION

The controlled release fertiliser composition of the present invention comprises a MOF. Provided the MOF can incorporate fertiliser, there is no particular restriction on the composition of MOF that can be used according to the invention.

A variety of different MOFs may be used. For example, the MOF may be amorphous or crystalline. The MOF may also be micro-MOF or meso-MOF.

MOFs according to the present invention include those having at least two metal clusters coordinated by at least one organic ligand.

As used herein, the expression 'metal cluster' is intended to mean a chemical moiety that contains at least one atom or ion of at least one metal or metalloid. This definition embraces single atoms or ions and groups of atoms or ions that optionally include organic ligands or covalently bonded groups. Accordingly, the expression 'metal ion' includes, for example, metal ions, metalloid ions and metal oxides.

Suitable metal ions that form part of a MOF structure can be selected from Group 1 through 16 metals of the IUPAC Periodic Table of the Elements including actinides, and lanthanides, and combinations thereof. The metal ion may be selected from Li ⁺ , Na ⁺ , K ⁺ Rb ⁺ , Be ²⁺ , Mg ²⁺ , Ca ²⁺ , Sr ²⁺ , Ba ²⁺ , Sc ³⁺ , Y ³⁺ , Ti ⁴⁺ , Zr ⁴⁺ , Hf ⁴ *, V ⁵⁺ , V ⁴⁺ , V ³⁺ , V ²⁺ , Nb ³⁺ Nb ⁵⁺ , Ta ⁵⁺ , Cr ⁶⁺ , Cr ³⁺ , Mo ⁶⁺ , Mo ³⁺ , W ⁶⁺ , W ³⁺ , Mn ⁴⁺ , Mn ³⁺ , MN ²⁺ , Re ⁷⁺ , Re ²⁺ , Fe ³⁺ , Fe ²⁺ Ru 4 ⁴ + ⁺ , r R> .u 3 ³ + ⁺ , n R.u 2 ² + ⁺ , _n R,h3 ³ + ⁺ , r R>ih-2 ² + ⁺ , r R,h+ ⁺ _Ύ

, _T I_r4 ⁴ + ⁺ , Ir2 ² + ⁺ , _T Ir + ⁺ , N _τ i· 2 ² + ⁺ , _n Pdj ⁴ + ⁺ , _n Pdj + Pt ⁴⁺ , Pt ²⁺ , Cu ²⁺ , Cu ⁺ , Ag ⁺ , Au ⁺ , Zn ²⁺ , Cd ²⁺ , Hg ²⁺ , B ³⁺ , Al ³⁺ , Ga ³⁺ , In ³⁺ , Tl ³⁺ , Si ⁴⁺ , Si 2+

Ge ⁴⁺ , Ge ²⁺ , Sn ⁴⁺ , Sn ²⁺ , Pb ⁴⁺ , Pb ²⁺ , As ⁵⁺ , As ³⁺ , Sb ⁵⁺ , Sb ³⁺ , Bi ⁵⁺ , Bi ³⁺ , La ³⁺ , Ce ³⁺ , Ce ⁴⁺ , Pr ³⁺ Pr ⁴⁺ , Nd ³⁺ , Sm ³⁺ , Sm ²⁺ , Eu ³⁺ , Eu ²⁺ , Gd ³⁺ , Tb ³⁺ , Tb ⁴⁺ , Dy ³⁺ , Ho ³⁺ , Er ³⁺ , Tm ³⁺ , Tm ²⁺ , Yb ³⁺

Yb ²⁺ , Lu ³⁺ , Th ⁴⁺ , U ⁶⁺ , U ⁵⁺ , U ⁴⁺ , U ³⁺ , and combinations thereof.

Suitable metal ion multidentate-coordinating organic ligands can be derived from oxalic acid, malonic acid, fumaric acid, succinic acid, glutaric acid, adipic acid, phthalic acid, isophthalic acid, terephthalic acid, citric acid, trimesic acid, 1,2,3-triazole, 1,2,4-triazole, 1,2,3,4-tetrazole, pyrrodiazole, squaric acid, sulfonic acids or phosphonic acids . Organic ligands suitable for the purpose of the invention comprise organic ligands listed in WO 2010/075610 and Filipe A. Almeida Paz, Jacek Klinowski, Sergio M. F. Vilela, Joao P. C. Tome, Jose A. S. Cavaleiro, Joao Rocha, 'Ligand design for functional metal-organic frameworks', Chemical Society Reviews, 2012, 41, 1088, the content of which is included herein in its entirety.

In some embodiments, the MOF is a Zr-based MOF. Examples of Zr-based MOFs that can be used in the invention include MOFs of the UiO-66 or UiO-67 type. A detailed characterisation of these MOFs and description of their synthesis is reported in J. H. Cavka et al., Journal of the American Chemical Society, 2008, 130, 13850, the content of which is incorporated herein in its entirety.

Advantageously, MOFs of the UiO-66 or UiO-67 type have remarkable mechanical stability under moist conditions and/or in the presence of different solvents, acids and bases. In addition, their production cost is low due to the simplicity of their method of synthesis.

MOFs of the UiO-66 type comprise zirconium secondary building units with preferably a 1,4-benzenedicarboxylate as the organic ligand. The inorganic centres are the key to the exceptional mechanical and chemical stability possessed by these structures. These centres are formed by an inner Zr ₆ 0 ₄ (OH) ₄ core in which each zirconium is linked to four oxygen atoms and each oxygen to three zirconium atoms resulting in a highly symmetric structure shaped like a Maltese star.

The structure of MOFs of the UiO-66 type is highly stable in various solvents such as water, methanol, acetone and dimethylformamide (DMF). Mechanical stability is reported for a pressure of 981 MPa. Thermal decomposition occurs at 540 °C, though structural - Si - changes are already reported for temperatures around 250 °C.

In some embodiments, the MOF is functionalised MOF. In these embodiments, the organic ligands of the MOF are functionalised organic ligands. For example, any one of the organic ligands listed herein may be additionally characterised by the presence of amino-, such as 2-aminoterephthalic acid, urethane-, acetamide-, or amide- sulfonate-moieties. The organic ligand can be functionalised before being used as precursor for MOF formation or alternatively pre-formed MOF can be chemically treated to functionalise its bridging ligands.

A skilled person would be aware of suitable chemical protocols that would allow functionalising MOFs with functional groups, either by pre-functionalising ligands used to synthesise MOFs or by post- functionalising pre-formed MOFs. Accordingly, suitable functional groups that may be provided on the MOFs include -NHR, -N(R) ₂ , -NH ₂ , -N0 ₂ , -NH(aryl), halides, aryl, aralkyl, alkenyl, alkynyl, azobenzene, pyridyl, bipyridyl, terpyridyl, anilino, -O(alkyl), cycloalkyl, cycloalkenyl, cycloalkynyl, sulfonamido, hydroxyl, cyano, -(CO)R, -(S0 ₂ )R, -(C0 ₂ )R, -SH, -S(alkyl), -S0 ₃ H, -S0 ₃ ^" M ⁺ , -COOH, COO ^" M ⁺ , -P0 ₃ H ₂ , -P0 ₃ H ^" M ⁺ , -P0 ₃ ^2" M ²⁺ , silyl derivatives, borane derivatives, ferrocenes and other metallocenes. In the chemical formulas listed above, M is a metal cation, and R is Cl-10 alkyl or aromatic ring.

Examples of functionalised MOFs include UiO-66-NH ₂ (amino functional group), UiO-66- OH (hydroxyl functional group) and UiO-66-N0 ₂ (nitroxide functional group). The cavity surface can be functionalised by exchanging the 1,4-benzenedicarboxylate with a functionalised 1 ,4-benzenedicarboxylate.

In some embodiments, the MOF is a zinc imidazolate framework (ZIF). ZIFs are water stable and non-toxic MOFs of straightforward and cheap synthesis. ZIFs show remarkable stability under alkaline conditions, a characteristic that makes ZIFs particularly suited to use in alkaline soil. Further advantageously ZIFs can be synthesised in water and are chemically stable in water even at high temperatures (e.g. at boiling point) for prolonged periods of time (e.g. several weeks).

ZIF frameworks feature tetrahedrally-coordinated transition metal ions (e.g. Fe, Co, Cu, Zn) connected by organic imidazolate organic ligands, resulting in three-dimensional porous solids. Similarly to zeolites, ZIFs have great thermal and chemical stability. Depending on the choice of precursors, many ZIF topologies can be synthesised.

Examples of ZIFs include ZIF-8, in which zinc in located in the metal node, bound to four nitrogen atoms of four different 2-methylimidazole molecules forming four coordinated zincs. The ZIF-8 framework is reported to be highly temperature stable (> 500 °C) and stable in various organic solvents as well as water up to their boiling points.

Specific examples of ZIF and specific synthesis methods to produce them are reported in Rahul Banerjee, et al., 'High-Throughput Synthesis of Zeolitic Imidazolate Frameworks and Application to C0 ₂ Capture', Science 2008, 319, 939 and corresponding Supporting Information. The whole content of those documents are incorporated herein in its entirety.

In some embodiments, the MOF is an aluminium-based MOF. Examples of aluminium- based MOFs include MIL-lOO(Al) and CAU-10. CAU-10 is made up of helical chains of cis-connected, corner sharing A10 ₆ -polyhedra linked by the V-shaped 1,3-benzene dicarboxylic acid or isophtalic acid. Examples of aluminium-based MOFs and their respective synthesis is are described in Reinsch, Stock, et al., Chemistry of Materials, 2013, 25, 17, the content of which is incorporated herein in its entirety.

In some embodiments, the MOF is biodegradable. As used herein, the term 'biodegradable' in the context of MOF refers to the ability of the MOF to be degraded by microorganisms and/or natural environmental factors into components that form the framework of the MOF, i.e. metal elements and compounds derived from the organic linkers forming the MOF framework. In a biodegradable MOF the organic linker molecules may be naturally occurring biodegradable molecules. Examples of such molecules include cyclodextrins, for example alkali metal cyclodextrins, sugars, amino acids and nucleobases. In one embodiment, the MOF is amorphous. In an amorphous MOF (aMOF), metal clusters and organic ligands form a framework that does not have detectable spatial order. The cavities of an aMOF result from an aperiodic spatial distribution of atoms, and are spatially distributed in a random fashion within the MOF framework. Aperiodic arrangements of atoms result in aMOFs generating X-ray diffraction patterns dominated by broad 'humps' caused by diffuse scattering and thus they are largely indistinguishable from one another by means of X-ray diffraction (XRD) measurements.

Any of the MOFs listed herein may be an aMOF. Characteristics and properties of aMOFs are described, for example, in Thomas D. Bennett, Anthony K. Cheetham, 'Amorphous Metal-Organic Frameworks', Accounts of Chemical Research, 2014, 47, 1555, the content of which is incorporated herein in its entirety.

The size distribution of the cavities of aMOFs can be determined by techniques that would be known to the skilled person. For example, measurements based on the use of Brunauer Emmett and Teller method (BET) are proposed in Brunauer, S., Emmett, P., and Teller, E. 'Adsorption of gases in multimolecular layers', Journal of the American Chemical Society, 1938, 60, 309. Whilst different gases can be used as probes (such as nitrogen, hydrogen, argon, helium, carbon dioxide, H ₂ 0 and methane) nitrogen as the gas probe is the most common.

Depending on the kind of aMOF, the cavities of the resulting framework may, for example, have a size measured with BET of up to 500 A.

In one embodiment, the MOF is crystalline.

In a crystalline MOF the metal clusters are coordinated by the organic ligands to form a geometrically regular network made of repeating units of cluster/organic ligand arrangements.

A crystalline MOF generates diffraction patterns when characterised by commonly known crystallographic characterisation techniques. These include, for example, powder X-ray diffraction (PXPD), grazing incidence X-ray diffraction, small angle X-ray scattering (SAXS), single crystal X-Ray diffraction, electron diffraction, neutron diffraction and other techniques that would be known to the skilled person in the field of crystallography of materials.

The crystalline nature of MOFs arises from regular and spatially ordered distribution of intrinsic cavities forming the framework.

As used herein the expression 'intrinsic cavities' is intended to mean the ordered network of interconnected voids that is specific to a crystalline MOF by the very nature of the MOF. As it is known in the art, the intrinsic cavity network of a MOF results from the specific spatial arrangement of the MOF's metal clusters and organic ligands and is unique to any pristine crystalline MOF. The intrinsic cavities of crystalline MOFs can be visualised as being formed by regularly distributed cages interconnected by windows or channels. The specific shape of cages and window/channels in crystalline MOFs is determined by the spatial arrangement of the chemical species forming the MOF framework. Accordingly, the expression 'intrinsic cavities' specifically identifies the overall ordered network of cages and window/channels of the native MOF framework.

According to the present invention, the dimensions of the intrinsic cavities of a crystalline MOF are to be quantified by mathematical models. As it is known in the art, the three- dimensional chemical structure of a crystalline MOF can be reproduced by mathematical models on the basis of the specific spatial distribution of the atoms constituting the MOF framework. The models allow extrapolating a parameter that is indicative of the dimensions of the intrinsic cavities, namely the 'largest cavity diameter' (LCD), which indicates the diameter of the largest spherical probe that can be inserted at some point of space within the MOF intrinsic cavities without overlapping with any framework atoms. Values of the LCD of intrinsic cavities of crystalline MOFs are intended herein as being those calculated according to the procedure described in E. Haldoupis, S. Nair and D. S. Sholl, Journal of the American Chemical Society, 2010, 132, 7528, which is herein incorporated by reference in its entirety. Depending on the kind of crystalline MOF, the intrinsic cavities may be characterised by values of LCD within the range of between about 5 A and about 500 A, between about 5 A and about 100 A, between about 5 A and about 50 A, between about 5 A and about 40 A, between about 5 A and about 30 A, between about 5 A and about 20 A, between about 5 A and about 15 A, between about 5 A and about 12 A, between about 5 A and about 10 A, between about 5 A and about 9 A, between about 5 A and about 8 A, between about 5 A and about 7 A, or between about 5 A and 6 about A.

The present invention is applicable to both micro-MOFs and meso-MOFs. As used herein, the term 'micro-MOFs' refers to MOFs in which the measured size of the cavities (for aMOFs) or the LCD of the intrinsic cavities (for crystalline MOFs) is smaller than 2 nm. The term 'meso-MOFs' include those MOFs in which the measured size of the cavities (for aMOFs) or the LCD of the intrinsic cavities (for crystalline MOFs) is between 2 nm and 50 nm. In the present invention the MOF incorporates fertiliser within its framework .

As described above, a 'fertiliser' is a substance that is added to a plant growth medium in order to increase fertility of the medium. Specifically, the term 'fertiliser' as used herein means a substance that comprises one or more plant nutrients, which are chemical elements essential to plant growth. A person skilled in the agricultural and horticultural arts would understand that in the context of the present invention the term 'fertiliser' is intended to embrace only those substances that are manufactured, isolated, represented, sold or used specifically as a means for supplying nutriment to enhance the development, productivity, quality or reproductive capacity of plants.

Examples of fertilisers that can be used in the context of the present invention therefore include organic or inorganic synthetic fertilisers that are commercially produced from petroleum or natural gas sources.

A fertiliser improves fertility of a plant growth medium by providing the medium with essential plant nutrients. Typically, essential plant nutrients provided by a fertiliser include any one or all of the following: (1) Three primary elements (nitrogen, phosphorus, and potassium);

(2) Three secondary elements (calcium, magnesium, and sulphur);

(3) Micro-nutrients (copper, iron, manganese, molybdenum, zinc, nickel, boron, chlorine, silicon, cobalt, iodine, selenium, sodium, vanadium). Elements in (1) and (2) above may also be referred to as 'macro-nutrients'. Elements in (3) above may also be referred to as 'trace elements' . The expression 'trace elements' is used to identify chemical elements found in small quantities in plants and/or the soil and which are used by organisms, including plants and animals, and are essential or beneficial to their physiology.

There is no limitation on the amount of fertiliser that may be present in the fertiliser composition of the present invention. For example, fertiliser may be present in an amount of least about 1 mg/g, at least about 25 mg/g, at least about 50 mg/g, at least about 150 mg/g, at least about 250 mg/g, or at least about 500 mg/g relative to the weight of MOF. In some embodiments, the fertiliser composition of the present invention comprises fertiliser in an amount of up to about 750 mg/g, up to about 500 mg/g, up to about 350 mg/g, or up to about 250 mg/g relative to the weight of MOF. Accordingly, in some embodiments the fertiliser composition of the present invention comprises from about 1 mg/g to about 750 mg/g, from about 25 mg/g to about 500 mg/g, from about 50 mg/g to about 350 mg/g, or from about 150 mg/g to about 250 mg/g of fertiliser relative to the weight of MOF.

Nitrogen is the main nutrient for strong, vigorous growth, good leaf color, and photosynthesis. Examples of fertilisers providing nitrogen that may be used in a fertiliser composition according to the present invention include ammonium nitrate (NH ₄ N0 ₃ ), ammonium sulphate ((NH ₄ ) ₂ S0 ₄ ), ammonium nitrate- sulfate, urea (CO(NH ₂ ) ₂ ), urea- ammonium nitrate, amino acids, ammonia, calcium nitrate, calcium ammonium- nitrate and mixtures thereof.

There is no limitation on the amount of nitrogen that may be present in the fertiliser composition of the present invention. For example, nitrogen may be present in an amount of least about 15 mg/g, at least about 40 mg/g, at least about 55 mg/g, at least 80 mg/g relative to the weight of MOF. In some embodiments, nitrogen is present in an amount of up to about 350 mg/g, up to about 250 mg/g, or up to 150 mg/g relative to the weight of MOF. Accordingly, in some embodiments the fertiliser composition of the present invention comprises from about 15 mg/g to about 350 mg/g, from about 40 mg/g to about 250 mg/g, or from about 55 mg/g to 150 mg/g of nitrogen relative to the weight of MOF.

Phosphorus promotes root development which helps strengthen plants. It also increases blooms on flowers and the ripening of seeds and fruit. Phosphorus supports formation of bulbs, perennials, and newly planted trees and shrubs. Examples of fertilisers providing phosphorus that may be used in a fertiliser composition according to the present invention include phosphates (P0 ₄ ^" ), metal phosphates, calcium phosphates, magnesium phosphates, potassium phosphates, phosphoric acid, monoammonium phosphate (MAP), diammonium phosphate (DAP), and ammonium polyphosphates (APP). There is no limitation on the amount of phosphorus that may be present in the fertiliser composition of the present invention. For example, phosphorus may be present in an amount of least about 100 mg/g, at least about 150 mg/g, at least about 200 mg/g, or at least about 250 mg/g relative to the weight of MOF. In some embodiments, the fertiliser composition of the present invention comprises phosphorus in an amount of up to about 500 mg/g, up to about 400 mg/g, or up to about 350 mg/g. Accordingly, in some embodiments the fertiliser composition of the present invention comprises from about 50 mg/g to about 500 mg/g, from about 100 mg/g to about 400 mg/g, or from about 150 mg/g to about 350 mg/g of phosphorus relative to the weight of MOF.

Potassium improves the overall health of plants. It helps them withstand very hot or cold weather, defend against diseases, helps fruit formation, photosynthesis, and the uptake of other nutrients. Examples of fertilisers providing potassium that may be used in a fertiliser composition according to the present invention include potassium nitrate, potassium chloride, potassium sulphate, potassium carbonate, and potassium phosphates, potassium magnesium sulphate, potassium-organic ligand complexes.

There is no limitation on the amount of potassium that may be present in the fertiliser composition of the present invention. For example, potassium may be present in an amount of least about 10 mg/g, at least about 20 mg/g, at least about 30 mg/g, at least about 40 mg/g. In some embodiments, the fertiliser composition of the present invention comprises potassium in an amount of up to about 100 mg/g, up to about 80 mg/g, or up to about 60 mg/g. Accordingly, in some embodiments the fertiliser composition of the present invention comprises from about 10 mg/g to about 100 mg/g, from about 20 mg/g to about 80 mg/g, from about 30 mg/g to 60 mg/g of potassium relative to the weight of MOF. In some embodiments, the fertiliser in the fertiliser composition of the present invention is a binary fertiliser, such as a NP, NK, or PK fertiliser, which contains at least two of the primary elements. Examples of binary fertilisers that may be used in the invention include combinations of aforementioned N, P, and K fertilisers, monoammonium phosphate (MAP) and diammonium phosphate (DAP). The active ingredient in MAP is NH ₄ H ₂ P0 ₄ . The active ingredient is DAP is (NH ₄ ) ₂ HP0 ₄ . There is no limitation on the amount of binary fertiliser that may be present in the fertiliser composition of the present invention. For example, binary fertiliser may be present in an amount of least about 10 mg/g, at least about 50 mg/g, at least about 100 mg/g, or at least about 200 mg/g relative to the weight of MOF. In some embodiments, the fertiliser composition of the present invention comprises binary fertiliser in an amount of up to about 500 mg/g, up to about 400 mg/g, or up to about 350 mg/g. Accordingly, in some embodiments the fertiliser composition of the present invention comprises from about 10 mg/g to about 500 mg/g, from about 50 mg/g to about 400 mg/g, from about 100 mg/g to about 350 mg/g, or from about 100 mg/g to about 350 mg/g of binary fertiliser relative to the weight of MOF.

In some embodiments, the fertiliser in the fertiliser composition of the invention is a NPK fertiliser, which includes all of the primary elements. Examples of NPK fertilisers that may be used in the invention include combinations of aforementioned N, P, and K fertilisers, and compounds capable to leach into the soil all three primary elements, for example ammonium sulphate, diammonium phosphate (DAP), monoammonium phosphate (MAP), potassium sulphate, calcium phosphate monobasic, or combinations thereof. In some embodiments, ammonium phosphate may be substituted for ammonium sulphate and calcium phosphate monobasic to provide the desired nitrogen and phosphorus levels.

There is no limitation on the amount of NPK fertiliser that may be present in the fertiliser composition of the present invention. For example, NPK fertiliser may be present in an amount of least about 10 mg/g, at least about 50 mg/g, at least about 100 mg/g, or at least about 200 mg/g relative to the weight of MOF. In some embodiments, the fertiliser composition of the present invention comprises NPK fertiliser in an amount of up to about 500 mg/g, up to about 400 mg/g, or up to about 350 mg/g. Accordingly, in some embodiments the fertiliser composition of the present invention comprises from about 10 mg/g to about 500 mg/g, from about 50 mg/g to about 400 mg/g, from about 100 mg/g to about 350 mg/g, or from about 100 mg/g to about 350 mg/g of NPK fertiliser relative to the weight of MOF. In some embodiments, the amount of nitrogen, phosphorus and potassium sources in NPK fertiliser that are used in the invention include ammonium sulphate in an amount of between about 10 to 40% by weight of the fertiliser composition, potassium sulphate in an amount of between 0 to 40% by weight of the fertiliser composition and calcium phosphate monobasic in an amount of between 0 to 40% by weight of the fertiliser composition.

In some embodiments, the fertiliser composition of the present invention is provided in form of pellets, granules or powder. Traditional methods to form fertiliser compositions into pellets, granules or powder would be known to a skilled person. Examples include milling, spraying, granulation, suspensions, prilling and hydraulic or mechanical compression of dry or moist fertiliser compositions.

The fertiliser composition of the present invention is a controlled release fertiliser composition.

As used herein, the expression 'controlled release fertiliser composition' means a fertiliser composition that is capable of releasing fertiliser within a predetermined period of time and according to a predetermined release profile. In consequence of being a 'controlled' release, plant nutrient provided by fertiliser released from the fertiliser composition of the present invention will accumulate in plant growth medium at a lower or reduced rate relative to the rate plant nutrients would accumulate if only pure fertiliser were used (which may be referred as an 'uncontrolled release').

In this context, the 'release rate' of fertiliser means the amount of fertiliser that is released from the fertiliser composition of the present invention per unit of time.

A parameter that gives a quantitative indication of the 'release rate' of fertiliser is the coefficient of nutrient release as determined by the method described by Milani N, McLaughlin MJ, Stacey SP, Kirby JK, Hettiarachchi GM, Beak DG, Cornells G 'Dissolution kinetics of macronutrient fertilizers coated with manufactured zinc oxide nanoparticles', Journal of Agricultural and Food Chemistry 2012, 60(16), 3991-3998. According to this method, a target sample (i.e. the fertiliser composition or the pure fertiliser) is packed in the form of particles inside a leaching column containing acid- washed sand. A 0.01 M CaCl ₂ leaching solution having pH 6, imitating soil conditions, is subsequently pumped through the column at a velocity of 10 mL/hr using peristaltic pump. A leachate solution is then collected at the opposite end of the column. The amount of nutrient collected in the leachate solution is measured at predetermined time intervals over a 48 h period. Nitrogen concentrations in each fraction can be measured using Inductively Coupled Plasma-Optical Emission Spectrometry (ICP-OES) following 0.2 μιη filtration. A [nutrient concentration] vs [time] plot allows visualisation of the release profile for the target sample. The nutrient concentration typically follows a first-order rate law, which rate constant is the coefficient of nutrient release.

The MOF of the fertiliser composition of the present invention effectively inhibits (in a controlled manner) the release rate of fertiliser into plant growth medium. As a result, the accumulation of plant nutrient provided by the fertiliser is inhibited compared to the accumulation of nutrients from the pure fertiliser.

Without being bound to a specific theory, it is believed that the release of fertiliser from within the MOF is a diffusion-limited process controlled by the specific porosity characteristics of the MOF. Thus, fertiliser compositions according to the present invention comprising different MOFs incorporating the same fertiliser within their framework can advantageously provide for a different release rate of that fertiliser. It will therefore be understood that the release rate of fertiliser from a fertiliser composition of the present invention can be directly determined by the specific combination of MOF with a specific fertiliser. The high number of MOFs available for the purpose of this invention combined with the high number of fertilisers that can be incorporated within the MOF framework advantageously provide for a vast amount of available fertiliser compositions having specific, tuneable, tailorable and customisable fertiliser release rates. The fertiliser composition of the present invention is not limited to any particular combination of MOF and fertiliser, provided that the fertiliser can be incorporated within the MOF.

The present invention also provides a method for producing a controlled release fertiliser composition, the method comprising incorporating fertiliser into Metal Organic Framework (MOF). As used herein, the term 'incorporating' will be understood as encompassing (i) synthesis methods in which the fertiliser is infiltrated into pre-formed MOF as well as (ii) methods in which the fertiliser is provided together with MOF precursors in a solution, and is trapped/encapsulated within the MOF framework during formation of the framework. The skilled person will be aware of synthesis methods that can be employed to produce a MOF that is suitable for the purpose of the present invention. Those synthesis methods traditionally require the combination of suitable MOF precursors in a solvent.

MOF precursors include those compounds known in the art that provide the metal ions listed herein in solution within a suitable solvent. Those compounds may be salts of the relevant metal ions, including metal-chlorides, -nitrates, -acetates -sulphates, -hydrogen sulphates, -bromides, -carbonates, -phosphates, and derivatives thereof, including mono- and poly- hydrate derivatives. MOF precursors also include organic ligands of the kind described herein that coordinate the metal ion clusters in the MOF framework. The organic ligands include molecules that have at least two chemical moieties capable of coordinating a metal ion. In some embodiments, these groups comprise carboxylates, phosphonates, sulphonates, N- heterocyclic groups, and combinations thereof.

In one embodiment, MOF is first synthesised and a fertiliser is subsequently infiltrated into the MOF framework.

Examples of procedures that can be adopted to infiltrate fertiliser into MOF are 'equilibrium absorption' and 'two solvents absorption'.

Equilibrium absorption is performed by immersing the MOF into a solution containing dissolved fertiliser. The concentration gradient of fertiliser promotes its diffusion into the MOF framework until concentration equilibrium is reached. There is no particular limitation with regard to the solvent that can be used in this procedure, provided that the fertiliser is soluble in the solvent and that the solvent can permeate the cavities of the MOF framework and then be evaporated.

Example of solvents suitable for fertiliser infiltration by equilibrium absorption include water, ethanol, methanol, isopropanol, tetrahydrofuran, acetone, ethyl acetate, acetonitrile and other polar solvents.

In the two solvents absorption MOFs are suspended in a hydrophobic solvent (the first solvent), for example n-hexane, in a hydrophobic container. There is no limitation on the first solvent suitable for this purpose, provided the first solvent (i) is immiscible in the second solvent, (ii) it does not dissolve the fertiliser, (iii) it is less dense than the second solvent, (iv) two solvent method can also be done in reverse if MOF is hydrophobic, and (v) it is of easy removal (e.g. by having low boiling point).

The fertiliser is separately dissolved in a hydrophilic solvent (the second solvent), typically water, to form a concentrated solution of fertiliser. This solution is subsequently added to the suspension of MOF in the first solvent. The partially hydrophilic nature of the inner pores of the MOF framework favours diffusion of the fertiliser from the second solvent into the MOF framework. It was advantageously found that the hydrophobic nature of the first solvent and of the container further enhance the diffusion of fertiliser into the more hydrophilic MOF framework. Examples of first and second solvents suitable for this procedure include n-pentane, petroleum spirits, toluene, chloroform, dichloro methane or any solvent that demonstrates immiscibility. Independent from the infiltration procedure, the concentration of the fertiliser in the infiltration solution is not limited, provided the fertiliser infiltrates into the MOF framework. For example, the concentration of fertiliser in the infiltration solution may be between about 1 g/L to about 2000 g/L, between about 5 g/L to about 1500 g/L, between about 10 g/L to about 1000 g/L, between about 50 g/L to about 800 g/L, or between about 100 g/L to about 500 g/L.

The infiltration time is not particularly limited, provided there is no detectable change in the concentration of fertiliser in the impregnation solution following impregnation. For example, in some embodiments impregnation is effected for at least about 1 hour, at least about 5 hours, at least about 10 hours, at least about 18 hours, at least about 24 hours, or at least about 48 hours, and in any case for less than about 72 hours.

There is no limitation on the particular temperature at which the impregnation is performed, provided the MOF is chemically and mechanically stable at that temperature. In some embodiments impregnation is performed at a temperature that is lower than 100 °C, 90 °C, 75 °C, 50 °C, or 35 °C. Thus, the impregnation temperature may be between -50 °C and 75 °C, between -50 °C and 50 °C, between -50 °C and 30 °C.

In one embodiment, impregnation is performed at room temperature. As used herein, the expression 'room temperature' will be understood as encompassing a range of temperatures between about 20 °C and 25 °C, with an average of about 23 °C.

After infiltration is complete the fertiliser compositions are separated from the infiltration solutions and purified according to procedures that would be known to the skilled person.

In one embodiment, fertiliser and MOF precursors are combined together into a solution, and MOF formation is subsequently promoted. As a result, the forming MOF framework incorporates the fertiliser in situ within its framework according to a one pot synthesis.

It will be understood that this embodiment is based on traditional methods of MOF solution synthesis, for example traditional solvo-thermal methods, performed in the presence of dissolved fertiliser.

MOF precursors include those compounds known in the art that provide metal ions listed herein in the solution within a suitable solvent. Those compounds may be salts of the relevant metal ions, including metal-chlorides, -nitrates, -acetates -sulphates, -hydrogen sulphates, -bromides, -carbonates, -phosphates, and derivatives thereof, including mono- and poly- hydrate derivatives.

MOF precursors also include organic ligands of the kind described herein that coordinate the metal ion clusters in the MOF framework. The organic ligands include molecules that have at least two chemical moieties capable of coordinating a metal ion. In some embodiments, these groups comprise carboxylates, phosphonates, sulphonates, N- heterocyclic groups, and combinations thereof. There are no particular restrictions on the solvents that can be used to prepare the solution into which MOF precursors and fertiliser are combined together in the one pot synthesis, provided that (i) the MOF precursors are soluble in the solvent, and (ii) the fertiliser is compatible with the solvent. That is, the solvent will typically be one that does not adversely affect the activity of the fertiliser. Examples of solvent that may be used include methanol, ethanol, dimethyl sulfoxide (DMSO), acetone, water and mixtures thereof.

Provided the MOF forms, there is no particular limitation on the concentration of MOF precursors present in the solution of the one pot synthesis. Concentrations of MOF precursors in the solution can include a range between about 0.001 M and 1 M, between about 0.01 M and 0.5 M, between about 0.01 M and 0.2 M, between about 0.02 M and 0.2 M, between about 0.02 M and 0.15 M, between about 0.05 M and 0.15 M, between about 0.08 M and 0.16 M. The values refer to concentration of organic ligand as well as concentration of metal salt, relative to the total volume of the solution containing the MOF precursors and the fertiliser. The ratio between the concentration of organic ligands and the concentration of metal salts is not limited, provided the ratio is adequate for the in situ formation of MOF incorporating the fertiliser within its framework. In some embodiments, the organic ligand to metal salt ratio may range from 60: 1 to 1:60 (mol : mol), from 30: 1 to 1:30, from 10: 1 to 1: 10, from 5: 1 to 1:5, from 2.5: 1 to 1:2.5, from 2: 1 to 1:2, or from 1.5: 1 to 1: 1.5.

The concentration of fertiliser in solution with the MOF precursors in the one pot synthesis is not limited, provided the fertiliser is incorporated in situ within the MOF framework such that the resulting fertiliser composition has an amount of fertiliser as described herein. For example, the concentration of the fertiliser in the solution may be between about 1 g/L to about 2000 g/L, between about 5 g/L to about 1500 g/L, between about 10 g/L to about 1000 g/L, between about 50 g/L to about 800 g/L, or between about 100 g/L to about 500 g/L.

Once the fertiliser and MOF precursors are combined into a solution, MOF formation can be initiated, for example, by providing heat to the solution. Those skilled in the art would be aware of methods available to provide heat to the solution, which may include the provision of heat by a heat source such as an oven, for example a microwave oven, a hot plate, or a heating mantel. In some embodiments the formation of the incorporating framework is effected at a temperature that is lower than 200 °C, 150 °C, 100 °C, 90 °C, 75 °C, or 50 °C. Thus, the formation of the incorporating framework may be induced between 40 °C and 150 °C, between 75 °C and 150 °C, between 100 °C and 150 °C. The one pot synthesis is an efficient method to get low loadings or covalent attachment within the framework between the fertiliser and MOF. The present invention also provides a method of fertilising a plant growth medium, the method comprising adding a fertiliser composition according to the present invention to the plant growth medium.

Procedure for providing a fertiliser composition to a plant growing medium would be known to the skilled person.

Examples of such procedures include all known industrial-farming techniques and large scale machinery based fertilisation, such as fertilisation by means of toolbars, disc spreaders, hoppers, high capacity spreaders, side-press applicators or strip-till applicators.

Accordingly, the invention also provides a method of increasing plant growth rate in a growth medium, the method comprising adding a controlled release fertiliser composition according to the invention to the plant growth medium. In the context of this aspect of the invention, 'increasing plant growth' will be intended relative to the plant growth rate detected in the absence of the fertiliser composition in the growth medium.

Accordingly the invention further provides for use of a controlled release fertiliser composition according to the invention to increase plant growth rate in a growth medium, as well as use of a controlled release fertiliser composition according to the invention to modify release of fertiliser into a plant growth medium such that a normalised amount of nitrogen released into the medium (mg/mg) at a given dissolution time/dissolution volume ratio is lower relative to that of a fertiliser composition absent the MOF.

Specific embodiments of the invention will now be described with reference to the following non-limiting examples. EXAMPLES

Analytical techniques Determination of the amount of nutrient in soil and plant samples

Methods to determine the plant nutrient content of soil and plant samples include wet oxidation, dry combustion and extraction. The methods used in these experiments are based on methods suggested for fertiliser samples by the Association of Official Analytical Chemists (AO AC method).

A sample (e.g. soil or plant) is combusted in a stream of pure oxygen at 900-1450 °C. During the combustion step carbon is converted to C0 ₂ and nitrogen to NO _x . The gas mixture is taken by a helium carrier stream. One aliquot is taken for C0 ₂ measurement via Infrared (IR) spectroscopy. Another aliquot is cleaned from H ₂ 0 and C0 ₂ via chemical traps and NO _x is reduced to N ₂ in a copper catalyst bed. Impurities are removed from the resulting N ₂ stream and N ₂ is detected as the change of thermal conductivity compared to a nitrogen free helium gas stream.

Colorimetrical procedures were also carried out in this work. Automated colorimetrical procedures can be carried out via flow injection analysis (FIA) or segmented continuous flow analysis (CFA) as high throughput methods.

Analyte and reagents are subsequently or at once injected into a continuously flowing carrier stream. Mixing coils provide time for mixing and reaction. At the end of passing the analytical manifold the sample enters the detection unit. In case of NH ⁴⁺ -N and NO N detection this part is commonly represented by a colorimetrical cell that determines nitrogen concentration via photo spectrometry at 660 nm for NH ₄ and 540 nm for N0 ₃ [Standard Methods for the examination of water and waste water, 2013a-Apha method 4500-NO ₃ ; Standard Methods, ISO: Water Quality- Determination of ammonium nitrogen by flow analysis (cfa and fia) and spectrometric detection, 1997]]. CFA differs from FIA systems in introducing air bubbles for segmentation of the analyte stream. The segmentation can be used to determine the reproduciblity of the system [Gordon, Jennings, et al, Methods Manual WHPO 91-1, 1993].

Surface Area and Porosity Measurement ofMOFs

Gas adsorption measurements were performed to investigate the pore structure and surface area of synthesised and purchased porous materials, and were collected on an ASAP2420 from Micromeritics.

The pre-dried sample with fertiliser was loaded into the sample tube before sealing the sample tube with the filler rod and the transeal cap. The loaded sealed sample tube was then attached to the sample degassing vacuum manifold of the ASAP2420 and a heating jacket was attached to the bottom part of the sample tube (containing the sample). The sample was evacuated for at least 24 h to a residual pressure p < 0.133 Pa (1 μι ¾). Temperature when evacuating varied due to different thermal stabilities of different compounds.

The sample tube was weighed again and the difference in weight between the empty and the filled sample tube was noted as the dry sample weight. The adsorptive gas was set to nitrogen. The adsorption temperature was -196 °C achieved by plunging the sample tube into liquid nitrogen throughout the measurement, and data was taken in the pressure range of 0 < p/po < 0.99. Desorption isotherms were collected subsequent to adsorption isotherms and relative pressure was in the range of 0.9 < p/po < 0.05.

Desorption isotherms were only collected for new materials to receive general information about the pore structure. After data collection the relative pressure and corresponding adsorbed volume of nitrogen was used to determine the BET surface area and the total pore volume as explained previously.

Determination of total nitrogen content in dry matter

Total nitrogen of dry matter was determined by a LECO TruMac CN analyser. This method was used to analyse the nitrogen content of dried plant material and fertiliser materials. Plant and fertiliser material were ground before homogenising the sample. The analyser was calibrated using glycine (18.66 wt. % Nitrogen) and ethylenediaminetetraacetic acid (EDTA 9.58 % Nitrogen). Samples and calibration standards were burnt in an atmosphere of pure oxygen at 1350 °C.

Determination of total nitrogen content in liquid samples

Total nitrogen of a liquid sample was analysed with a Shimadzu TOC-VCSH/CSN + TNM-1 organic carbon and nitrogen analyser. This method was used to analyse liquid samples containing urea. The sample is injected into a catalyst packed combustion tube, at a furnace temperature of 720 °C. This causes the decomposition of the TN in the sample into nitrogen monoxide. The nitrogen monoxide is then taken by the carrier gas through a cooling and dehumidifying process and finally to a chemiluminescence gas analyser.

Determination of total nitrogen content in mineral nitrogen

Ammonia-nitrogen is determined as described by the international standard ISO 11732 [Standard Methods, ISO: Water Quality- Determination of ammonium nitrogen by flow analysis (cfa and fia) and spectrometric detection, 1997]. The sample is buffered to pH 12 and reacted with sodium salicylate and sodium nitroprusside before reaction with sodium dichloroisocyanurate solution (DCIC) in an automated flow injection analyser (Lachat QuikChem 8500 series 2). After reaction at 60 °C, the intensity of the colour is measured at 660 nm and the NH ₄ -N concentration is calculated from a set of calibration standards measured at the same time.

Nitrate and nitrite-nitrogen is determined as described in the APHA method 4500-NO ₃ - F [Standard Methods, For the examination of water and waste water, 2013a-Apha method 4500-NO ₃ ]. In this segmented flow analysis method nitrate is reduced to nitrite in an open tubular cadmium column in an atmosphere of helium and then reacted with sulphanilamide and N-(l-naphthyl)- ethylenediamine dihydrochloride (NEDD) in phosphoric acid. The pink colour formed is determined colorimetrically using an automated flow injection analyser (Lachat QuikChem 8500 series 2) at 540 nm. The N0 ₃ -N (also referred to as NO _x -N indicating the sum of nitrite and nitrate-nitrogen) concentration is calculated from a set of calibration standards measured at the same time. Conductivity of ammonium nitrate solutions

The ammonium nitrate concentration in 0.01 M CaCl ₂ solution was determined via conductivity measurements to supplement measurements via colorimetric methods. Since the correlation between conductivity and ammonium nitrate concentration becomes more non- linear with increasing salt concentrations, samples were diluted 1:5 by volume with deionised water before conductivity measurements. Incorporation of Fertiliser into MOF by post-synthesis infiltration

Two impregnation methods were applied:

Equilibrium adsorption: MOFs were soaked in excess impregnation liquid for 24 h, Vi _mp > > V _p . Excess solution was then filtered off and the impregnated material was dried.

Two solvents: Fertiliser molecules were also diffused into the MOF framework via a methodology called 'two solvents' for effective pore filling [Hill, Booth, et al.; Dalton Transactions, 2010, 39, 5306]. Porous materials were suspended in a hydrophobic solvent (n-hexane) in a hydrophobic container. N-hexane was chosen because (a) it is immiscible in water, (b) it does not dissolve fertiliser molecules (urea and ammonium nitrate), (c) its low density (p = 0.66 g/mL) allows for water and suspended matter to precipitate, (d) it is easy of removal due to low boiling point.

The suspension was stirred turbulently to achieve proper mixing and then subsequently an aqueous precursor of concentrated aqueous fertiliser solution was added to the suspension. Hydrophobic container and solvent were chosen to ease the hydrophilic fertiliser solution to adsorb at the more hydrophilic surface of porous materials. Therefore for extremely hydrophobic surfaces (e.g. Activated Carbon) this method may fail. Determination of Fertiliser Release Kinetics

Fertiliser release kinetics were determined similar to a method described previously by McLaughlin et al (Milani, McLaughlin, et al.; World Congress of Soil Science, Soil Solution for a Changing World, Brisbane, Australia, 1-6* August, 2010).

Fertiliser particles are packed into columns and a leaching solution was pumped through this column. The leachate is collected and analysed for nitrogen. The pelletised fertiliser materials were sieved and the fraction 2.25 mm < pellet-size < 3.35 mm was taken for determination of release kinetics. A plastic column with an inner diameter of 8 mm was subsequently built up as described in the following steps 1) rubber stopper with drain fitted on one end of the column; 2) one layer of filter paper fitted to the column size; 3) pelletised fertiliser materials aiming at 25 mg nitrogen; 4) one layer of filter paper fitted to the column size; 5) 10 mm tubing as spacer; 6) glass wool to keep previous 5 layers in place; 7) rubber stopper with drain fitted on the second end of the column.

The column was fixed with the fertiliser particles on the upper end vertically on a stand. The drain of the upper rubber stopper was attached to a tube leading to the fraction collector. 15 dry and clean 10 mL tubes were placed in a rack on the fraction collector. The drain of the lower rubber stopper was attached to a tube with an inner diameter of 1.143 mm leading through a peristaltic pump. The beginning of that tube was placed in a reservoir of 0.01 M CaCl ₂ solution at pH 6, which was chosen as leaching solution to imitate soil conditions.

The solution was sucked in by the peristaltic pump (Gilson minipuls 3) and pumped into the column until the solution reached the lower end of the fertiliser particle packing. The fraction collection and the pump were then started simultaneously. A flow rate of 1 mL/min was chosen. Fractions were collected. Fractions containing ammonium nitrate were analysed via colorimetric methods and via conductivity measurements. Fractions containing urea were analysed via total nitrogen detection. After each run the column was cleaned and all tubing was flushed with fresh 0.01 M CaCl ₂ solution. The results were then normalised and the release over time was measured and then plotted as the relative nitrogen dissolved (mg/mg) for a given dissolution time/dissolution volume ratio (see plots of Figures 3, 4 and 5). Materials

Fertilisers

Sources of nitrogen used in the Examples are urea (CO(NH ₂ )2), ammonium/ammonia (NH ₃ /NH ₄ ⁺ ) and nitrate (N0 ₃ ^~ ).

Soil

The soil used for plant experiments in this work was obtained from Mt Compass in South Australia. The sandy soil is nitrogen poor (0.03 wt. %) and has a low water maximum water holding capacity (MWHC = 18.4 wt. %). The low nitrogen content makes the soil suitable for nitrogen fertiliser investigations because nitrogen in the growing plant can be directly attributed to fertiliser nitrogen. Parameters of the soil are reported in Table 1.

Table 1: Mt Compass soil composition and properties.

Frame Wheat

Frame wheat was released at the Waite Institute in Adelaide, South Australia and is specially adopted to low rainfall areas. It is categorised as soft wheat with a minimum protein content of seeds > 10 wt. %.

The nitrogen content of the seeds was determined via total nitrogen determination with combustion as 2.21 wt. % at an average seed weight of 45.5 mg. EXAMPLE 1 - Synthesis of MOF

Zirconium-based MOFs A modified version of the procedure suggested by Schaate et al. (Schaate, Roy, et al., Chemistry- A European Journal, 2011, 17, 6643) was applied, using benzoic acid to control crystallisation and yield high crystalline nano particles.

UiO-66 synthesis

1.5 g zirconium(IV) chloride (ZrCl ₄ , 6.44 mmol) was dissolved in 125 mL DMF in a 500 mL Schott flask under stirring. 1.58 g benzene- 1,4-dicarboxylic acid (BDC) (9.52 mmol) were added to the solution, followed by another 125.5 mL DMF. After all reagents were dissolved 12.45 mL water (690 mmol) were added followed by 12.52 g benzoic acid (102.5 mmol).

The flask was capped and the reaction mixture was stirred until all reagents were homogenously dissolved. The solution was heated to 120 °C and kept at that temperature for 24 h resulting in precipitation of particles. The suspension was then allowed for cooling down to room temperature. The maximum yield after work up was approximately 1.26 g MOF/g organic ligand.

U1O-66-NH2 synthesis

The amino-functionalised U1O-66-NH ₂ was synthesised as described for UiO-66. Benzene- 1,4-dicarboxylic acid was replaced by 1.73 g 2-amino-l,4-benzenedicarboxylic acid (Amino-BDC) (9.55 mmol). The maximum yield after work up was approximately 0.425 g MOF/g organic ligand.

UiO-67 synthesis

UiO-67 was synthesised as described for UiO-66. Benzene- 1,4-dicarboxylic acid is replaced by 1.56 g biphenyl-4,4-dicarboxylic acid (BPDC) (6.44 mmol). Zr-BTB synthesis

Synthesis of the MOF Zr-BTB was not reported before. l,3,5-tris(4- carboxyphenyl)benzene (BTB) was used as ligand. Synthesis was carried out as described for UiO-66. Benzene-1,4- dicarboxylic acid is replaced by 3.57 g l,3,5-tris(4- carboxyphenyl)benzene (8.14 mmol). The maximum yield after work up was approximately 0.42 g MOF/g organic ligand.

Separation and purification

The suspension at room temperature was centrifuged until the supernatant became clear. The supernatant was discarded and replaced by approximately 250 mL of fresh DMF in order to dissolve unreacted reagents. After resuspending the particles centrifuging and replacing of DMF was repeated. After a third time of centrifuging and discarding of the supernatant the particle slurry was spread on a flat glass dish for drying. The slurry was dried at 100 °C under dynamic vacuum in a vacuum oven.

The resulting coarse particles were soaked twice in fresh DMF for 24 h. The coarse particles settle without centrifuging and used DMF can be decanted. The particles were then soaked in 5 mL fresh, dry methanol two times for 3 h and finally once for 24 h to replace DMF molecules in the pores with methanol molecules. After the methanol solvent exchange, particles were dried and activated at 60 °C under vacuum <6xl0-l Pa. The procedure described above for synthesis and work up is exemplary. The procedure was repeated several times with variation in scale (80-12,000 mg zirconium(IV)chloride). All quantities were scaled in the same ratio.

CAU-IO-NO2 aluminium-MOF

Synthesis was carried out as suggested by Reinsch et al.[Reinsch, Stock, et al., Chemistry of Materials, 2013, 25, 17]. 17.28 g of 5-nitro-l,3-benzenedicarboxylic acid (81.8 mmol) was dissolved in 48 mL DMF under gentle heating in a glass reactor. 152.2 mL water was added to the solution. 40.8 mL 2 M A1C1 ₃ -6H ₂ 0 in H ₂ 0 (81.6 mmol) were added to the solution. The glass reactor was sealed with a Teflon cap, then heated to 120 °C and kept at 120 °C for 15 h. Afterwards the reacted solution was allowed to cool down to room temperature. The maximum yield after work up was approximately 0.965 g MOF/g organic ligand.

CA U-10-OH aluminium- MO F

12 g of 5-hydroxy-l,3-benzenedicarboxylic acid (65.88 mmol) was dissolved in 48 mL DMF under gentle heating in a glass reactor. 158.4 mL water was added to the solution. 33.6 mL 2 M A1C1 ₃ -6H ₂ 0 in H ₂ 0 (67.2 mmol) were added to the solution. The glass reactor was sealed with a Teflon cap, then heated to 120 °C and kept at 120 °C for 15 h. Afterwards the reacted solution was allowed to cool down to room temperature. The maximum yield after work up was approximately 1.07 g MOF/g organic ligand. Separation and purification

For both MOFs described above a white powder precipitated during synthesis. The suspension was filtered via vacuum filtration. Then the powder was resuspended in 300 mL water under sonication for 30 minutes. The resuspended solution was filtered again. Both powders were then transferred to ceramic containers and activated at 200 °C for 2 h in a muffle oven.

The procedure described above for synthesis and work up is exemplary. The procedure was repeated several times with variation in scale (4-48 mL DMF). All quantities were scaled in the same ratio

ZIF-8

15 g Ζη(Ν0 ₃ ) ₂ · 6Η ₂ 0 (79 mmol) was dissolved in 1050 mL methanol. 30 g 2- methylimidazole (574 mmol) was dissolved in 1050 mL methanol. Both solutions were poured into a 2000 mL Schott flask and capped. The mixture was heated to 40 °C and kept at 40 °C for 48 h. The maximum yield after work up was approximately 0.173 g MOF/g ligand.

The white powder from above was collected by centrifuging until the supernatant became clear. After resuspending the powder in 240 mL fresh methanol, the suspension was centrifuged again. The process was repeated until methanol was exchanged three times to remove unreacted reagents. The so obtained wet powder was dried under nitrogen atmosphere at room temperature for 24 h.

The procedure described above for synthesis and work up is exemplary. The procedure was repeated several times with variation in scale (3.15 g Ζη(Νθ3)2·6Η ₂ 0). All quantities were scaled in the same ratio.

Biodegradable MOFs Bio-MOF-1 can be synthesised using a method reported by An et al.[An, Hupp, et al., Nature Communications, 2012, 3, 604], the content of which is incorporated herein in its entirety. The framework is constructed of zinc-adeninate building bricks and benzene- 1,4- dicarboxylic acid as linker. CD-MOF-1 can also be synthesised using a method reported by Smaldone et al. [Smaldone, Stoddart, , et al., Angewandte Chemie , 2010, 122, 1], the content of which is incorporated herein in its entirety.

The framework described by the aforementioned group is constructed of γ-cyclodextrin rings as organic linker and K ⁺ cations for the inorganic component, γ-cyclodextrin was replaced by β-cyclodextrin being approximately 25 times cheaper (Sigma).

EXAMPLE 2 - characterisation of MOFs

BET surface area and single point pore volume were calculated for each based on the N ₂ isotherm data. Micro pore volume (d < 20 A) was calculated from the pore size distribution. For MOFs the Tarazona (NLDFT) model for cylindrical pores was used. Surface areas, pore volume and micropore volume for the corresponding materials are shown in Table 2.

All materials show a BET surface area and micro pore volume similar to literature values. The micro pore volume of UiO-66-NH ₂ appears higher than that of the unfunctionalised UiO-66-MOF. It is remarkable that Zr-BTB, UiO-66 and UiO-66-NH ₂ show a significant higher total pore volume than the micro pore volume as perfect ordered crystals should only exhibit micro porosity. This behaviour is attributed to gas adsorption on and between nanoparticles. Table 2: Porosity characterisation. Surface area was determined via the BET method, total pore volume is a single point measurement and micropore volume was determined via a DFT model.

In relation to the pore size distributions Al-MOF CAU-10-NO ₂ shows a single sharp peak at approximately 6 A. The pore volume of CAU-10-OH is estimated to the pore volume of CAU-10-NO ₂ due to structural similarities. ZIF-8 shows sharp peaks at 9 and 15 A and the microporous activated carbon has a broad pore size distribution in the range of 6-9 A and 11-13 A. All Zr-MOFs show pores with a diameter of approximately 6 A.

The larger pore width is not as distinctive but can be found at approximately 7-8 A for UiO-66, U1O-66-NH ₂ and Zr-BTB. UiO-67 contains larger pores with diameters of approx 10-11 A. This is due to the extended linker in UiO-67 compared to UiO-66 (biphenyl-4,4- dicarboxylic acid to 1,4-benzenedicarboxylic acid).

A hysteresis is apparent during desorption in the gas isotherms of the zirconium-MOFs UiO-66, U1O-66-NH ₂ and Zr-BTB. This indicates mesoporous cavities as these cavities are emptied at lower pressures due to surface tension effects. The adsorption isotherm of these materials also shows a significant steepness in the range of 0 < p/ρθ < 0.6. These materials are supposed to consist of nanoparticles.

The slope can be attributed to inter-particle adsorption. Activated Carbon, UiO-67, ZIF-8 and CAU- IO-NO ₂ show micro porous behaviour with a low steepness of the isotherm for p/ρθ < 0.2 and no hysteresis. CAU-10-OH is showing non porous behaviour since no initial gas adsorption at low pressures can be observed. This occurs due to little attraction between nitrogen molecules and the pore surface. According to literature, gas adsorption experiments for CAU-10-OH with H ₂ O and H ₂ show porous behaviour [Reinsch, Stock, et al., Chemistry of Materials, 2013, 25, 17].

Stability ofMOFs in Water

100 mg of the porous materials UiO-66, UiO-66-NH ₂ , UiO-67, Zr-BTB, CAU-10-NO ₂ , CAU-10-OH, ZIF-8 as described were each filled into a separate 1.5 mL plastic tubes along with activated carbon as control. 1 mL water was added to each tube. The tubes were closed and vortexed to ensure wetting of all particles. The tubes were stored at room temperature for 48 hours. Excess water was then taken off the tubes with a pipette and pre- dried at 80 °C. Subsequently, samples were activated as described in their synthesis procedures. Nitrogen adsorption measurements were performed on all each samples except the CAU-10-OH samples to obtain the BET surface area, the pore volume and the pore size distribution after exposure to the aforementioned environment. Degassing temperature was 100 °C. CAU-10-OH was analysed for its PXRD pattern after the treatment as described in previously.

Stability ofMOFs in Fertiliser Solution

For infiltration of nitrogen fertiliser (urea and ammonium nitrate) into MOFs aqueous fertiliser solutions were used. Therefore stability in this environment is essential for the materials. Two times 100 mg of each porous material were filled each into 1.5 mL plastic tubes. To the tubes 1 mL of 800 g/L aqueous ammonium nitrate solution or 1 mL of 600 g/L aqueous urea solution were added.

Tubes were closed and vortexed to ensure wetting of all particles. The tubes were stored for 24 h at room temperature. After 24 h, excess fertiliser solution was taken off the tubes with a pipette. To remove fertiliser molecules from the porous materials, the following procedure was applied: 1 mL of water was added. Then the tubes were vortexed and the resulting suspension was allowed to settle. Excess solution was taken off and the procedure was repeated. Further 1 mL water was then added to the porous materials and soaked for 24 h. After 24 h, excess water was taken off again. Afterwards, samples were activated as described in their synthesis procedures.

The treated and activated samples were analysed for their BET surface area, pore volume and pore size distribution as described to see the impact of the aforementioned treatment. Degassing temperature was 100 °C. CAU-10-OH was analysed for its PXRD pattern after both treatments as described previously.

Stability in Moisture and Fertiliser Solution

The moisture stability was assessed via gas adsorption measurements. The materials' surface area is expected to change as a result of exposure to moisture or a moist environment. As discussed previously, some materials show mesoporous behaviour (UiO- 66, U1O-66-NH ₂ , Zr-BTB) resulting from small inter-nano article cavities. These cavities rather contribute to the total pore volume than to the micro pore volume and the surface area. Thus the surface area and the micro pore volume are taken into account when assessing the stability of the framework. Surface areas of the fresh and treated materials show UiO-66 and U1O-66-NH ₂ materials to be stable when treated against moisture and concentrated fertiliser solution. Hence materials are effectively applicable as fertiliser carrying materials.

The same statement can be made for CAU- IO-NO ₂ though the surface area measured after storage in water was measured approximately 25 % lower than the native material.

According to literature, the aluminium MOF is stable and the activation procedure is performed in water [Reinsch, Stock, et al., Chemistry of Materials, 2013, 25, 17].

The surface area of activate carbon does not change significantly after the treatment in water and ammonium nitrate solution but decreased approximately 50 % when treated with 600 g/L urea solution. Urea might be trapped in small irregular shaped and dead end pores of the activated carbon and were not totally removed while leaching. This might be a beneficial characteristic for a fertiliser application but could also lead to inefficiency when urea is never released. The activated carbon might even be unstable in 600 g/L urea solution but this is unlikely due to known exceptional stability of activated carbon. ZIF-8 was shown to be stable in water but dissolved in concentrated ammonium nitrate and urea solution and can thus not be loaded with solution based fertiliser molecules. This combination may therefore be a good candidate for a one pot synthesis method to get fertiliser into the ZIF-8 framework should these particular fertilisers be chosen.

UiO-67 and Zr-BTB both degraded in water over time, as seen by a decrease of surface area of more than 75 % for UiO-67 and more than 50 % for Zr-BTB after two days. The degradation of UiO-67 can be attributed to the enlarged organic linker compared to UiO- 66. The investigation of the cumulative pore volume and the pore width for all materials and all treatments tells a similar story. Micro pore volume (pore width < 20 A) of Zr-BTB and UiO-67 decreased after soaking in water and micro pore volume of activated Carbon decreased after treatment in urea solution.

Similarly cumulative pore volumes at larger pore widths (> 20 A) may change due to other effects than structural stability (see Zr-BTB, UiO-66 and UiO-66-NH ₂ ). For Zr-BTB, UiO- 66 and UiO-66-NH ₂ it is shown how pore volume is increasing at pore widths > 20 A which is attributed to inter-nanoparticle cavities. These inter-nanoparticle cavities may rearrange or compact and may thus lead to a change in total pore volume. Hence the assumption that surface area should be considered for stability statements is confirmed.

EXAMPLE 3 - Synthesis of fertiliser composition Nitrogen fertiliser molecules (urea and ammonium nitrate) were diffused into porous metal-organic frameworks and activated carbon (for comparison). The manufacturing of the fertiliser composition and the subsequent plant experiment are explained below.

To determine the fertiliser nitrogen content of the manufactured fertiliser composition (CN fertiliser SRX the total nitrogen content of the fertiliser composition (C SR) and the corresponding unloaded porous materials (CN Blank) was determined via the combustion method. The fertiliser nitrogen fraction was than calculated via equation 1

(CN fertiliser SR) = [(CN SR— CN Blank )/(C _N Fertiliser CN Blank)] * CN Fertiliser (Eql)

Method 1

All reagents required for the first manufacture of fertiliser compositions are listed below: UiO-66 and UiO-66-NH ₂ were synthesised, worked up and activated as previously described.

Activated carbon was used as purchased. The pore volume of these materials was determined following the procedures previously described.

Materials were loaded by equilibrium adsorption. 2 g of each material mentioned above was filled into a separate 50 mL plastic container containing a stir bar. 15 mL of aqueous 800 g/L ammonium nitrate solution were added to each container. The suspension was stirred for 17 h. After that time excess solution was filtered off via vacuum filtration and excessively washed with n-hexane. In order to minimise losses of fertiliser molecules through leaching n-Hexane was chosen as washing liquid as it does not dissolve ammonium nitrate. The resulting wet powder was dried at 100 °C for 5 h. After nitrogen analysis the fertiliser nitrogen content was calculated via equation 1 and the results are shown in Table 3.

Table 3 Nitrogen content in the fertiliser composition prepared according to method 1.

Method 2

UiO-66, UiO-66-NH ₂ , CAU-IO-NO ₂ and CAU-10-OH were synthesised, purified and activated as described previously. The pore volume of these materials was determined as previously.

Two different fertiliser solutions were prepared: (a) an aqueous 800 g/L ammonium nitrate solution, and (b) an aqueous 600 g/L urea solution.

All quantities needed for preparation are listed in Table 4. The volume of fertiliser solution (a) and (b) is calculated as the absolute pore volume for each porous material. Table 4 Preparation of urea and ammonium nitrate loaded porous materials.

For diffusion of the liquid fertiliser solution into the porous materials the methodology explained previously was applied. The amounts of porous material shown in Table 4 were filled into 500 mL polypropylene containers. 250 mL of n-hexane were added to the porous materials. The suspension was stirred with a stir bar to ensure a quasi-homogenous mixture. The volume of ammonium nitrate solution (a) and urea solution (b) was added dropwise to the corresponding containers with a 1 mL pipette. The suspension was stirred for 24 h.

Subsequently, particles were allowed to settle and a large part of the liquid phase was removed with a pipette. The residual slurry was dried under low dynamic vacuum at 60 °C for 48 h resulting in fine powder for each of the materials. After nitrogen analysis the fertiliser nitrogen content was calculated via equation 1 and the results are shown in Table 5. Table 5 Nitrogen content in the fertiliser composition prepared according to method 2.

Selective incorporation of Fertiliser into the MOF framework

Porous materials (UiO-66, UiO-66-NH ₂ , CAU-10-NO ₂ , CAU-10-OH, ZIF-8 and activated carbon) were tested for their selective incorporation of fertiliser molecules. To 200 mg of porous materials either 5 mL of 0.15 g/L aqueous ammonium nitrate solution or 5 mL 0.15 g/L urea solution was added. Tubes were closed firmly and shaken for 48 h. After 48 h the suspension was filtered with 0.45 μηι syringe filters and the filtrate was submitted for nitrogen analysis. Filtrates containing ammonium nitrate were analysed via colorimetric methods, filtrates containing urea were analysed for total nitrogen. Samples of 0.15 g/L ammonium nitrate and urea solution without MOFs were analysed as well.

The selective incorporation of fertiliser molecules was investigated by adding 200 mg of the porous materials to 1 mL of 0.15 g/L aqueous ammonium nitrate solution and urea solution respectively. Results for the change in NH ₄ ⁺ and N0 ₃ ^~ nitrogen concentration are shown in Figure 1. Compared to the stock solution the ammonium-nitrogen concentration did not change significantly for ZIF-8, CAU-10-NO ₂ and the zirconium MOFs (UiO-66 and UiO-66-NH ₂ ), whereas ammonium- nitrogen decreased when activated carbon or CAU-10-OH was added.

Activated carbon is known to remove ammonium nitrate from waste water and recently simulation results have shown that OH-functional groups are thermodynamically beneficial for ammonia capture in MOFs rather than NH ₂ -functional groups [Kim, Snurr, et al., Langmuir, 2013, 29, 1446]. Nitrate nitrogen was lower in all solutions after treatment with Zr-MOFs and Al-MOFs. This may be explained as N0 ₃ ^~ is attracted by positively charged metal sites within the framework. Nitrate concentration after treatment with ZIF-8 increased by more than 100% possibly due to unreacted zinc-nitrate from the synthesis and insufficient washing steps. A similar tendency of ammonium and nitrate adsorption in the porous materials is favoured inside the MOF framework as a result of charge neutrality.

Pelletisation

The powdery fertiliser compositions with MOFs and control carbon as well as powdery urea and powdery ammonium nitrate absent any porous materials were compacted with a hydraulic press without addition of water or binders. A stainless steel hollow cylinder with an inner diameter of 40 mm was placed on a bottom plate covered with Glad Wrap.

Approximately 3.5 g of a fertiliser material was equally distributed inside the cylinder on the bottom plate and covered with a thin plastic disc. A stamp was then fitted into the hollow cylinder and a pressure of approximately 39 MPa was applied with a hydraulic press. The powder was compacted to a flat disc with a thickness of approximately 2.5 mm (depending on the fertiliser composition) and a diameter of 40 mm. Discs for each fertiliser were manufactured this way. The disks of fertiliser composition were cut with a blade resulting in irregular shaped particles due to the brittleness of the discs. The obtained pellets were sieved and the size selected at 1.18 mm - 3.35 mm.

EXAMPLE 4 - characterisation of the fertiliser compositions

Porosity Measurement and Surface Analysis

Assuming all fertiliser diffused into the inner pore structure of the UiO-66-particles and no losses when drying the loaded material, the residual pore volume Vloaded can be calculated by Equation 2.

In Equation 2, m _M oF is the mass of MOF that is used; Vp is the unloaded specific pore volume (0.973 cm3 /g); C is the concentration of the fertiliser solution (0.8 g/mL ammonium nitrate, 0.6 g/mL urea) and p the fertiliser density. The right hand side of the equation are non-specific items for illustration.

For fresh UiO-66 the experimental residual pore volume is less than 50 % of the calculated minimal pore volume for the ammonium nitrate loading and approximately 60 % for the urea loading. Possibly small pores are blocked by the fertiliser molecules or residual water trapped in the pores and thus become inaccessible for nitrogen gas uptake. Whether pores are partly filled or fertiliser molecules are blocking pores from the particle outside is not certain but since UiO-66 exhibits an open pore structure with interconnected pores, pore blocking from the particle outside seems unlikely.

BET and XPS Surface Analysis

100 mg of UiO-66, UiO-66-NH ₂ , CAU-10-NO ₂ and CAU-10-OH were each filled into 50 mL plastic tubes. 20 mL of n-hexane was added to the tubes. 800 g/L aqueous ammonium nitrate solution according to 50 % v/v of the pore volume was added dropwise while stirring to each tube to allow diffusion of fertiliser solution into the pores. Tubes were closed and the suspension was stirred for 24 h at room temperature. After 24 h, n-hexane and water were evaporated at 60 °C under dynamic low vacuum in a vacuum oven for 48 h. The procedure was repeated with 600 g/L aqueous Urea solution instead of 800 g/L aqueous ammonium nitrate solution with a fresh set of porous materials. All samples but CAU-10-OH-samples were analysed for their BET surface area, pore volume and pore size distribution. The degassing temperature was 50 °C.

Energy Dispersive X-ray Spectroscopy (EDX)

UiO-66 was taken as an exemplary MOF into which urea and ammonium nitrate molecules could be diffused by the various methods. Scanning electron microscope pictures were taken of the resulting particles and energy dispersive X-ray spectra were taken of the particles' surfaces. UiO-66 was chosen because no nitrogen is supposed to be present in the framework and therefore a nitrogen peak in the EDX spectrum can be attributed to a fertiliser molecule.

The different loading and work up methods are listed below. Each method was carried out for 24 h at room temperature and 100 mg of fresh UiO-66-MOF were used. Each loading method was carried out with both 500 g/L ammonium nitrate solution as well as 500 g/L urea solution in a separate batch.

1) As synthesised UiO-66-MOF

2) Fertiliser solution according to 0.8 pore volumes was added dropwise. Particles were vortexed to ensure wetting of all particles

3) UiO-66-MOF was suspended in n-hexane and stirred. Fertiliser solution according to 0.8 pore volumes was added dropwise.

4) 50 mg of the sample prepared in 2) was washed with 10 mL water under vacuum filtration in order to wash off fertiliser molecules from the particles' surface.

All 7 samples (1 blank, 3 loaded with ammonium nitrate and 3 loaded with urea) were dried for 48 h at 50 °C under low dynamic vacuum and then processed and analysed via SEM and EDX.

Structure Analysis of Loaded UiO-66 Framework

The metal-organic framework UiO-66 was analysed for structural changes via powder X- ray diffraction after loading with ammonium nitrate molecules. 100 mg of each material was soaked in 1 mL 500 g/L aqueous ammonium nitrate solution for 48 h in a 1.5 mL plastic tube. Excess solution was then taken off with a pipette and 1 mL dichloromethane was added (DCM has a lower density than the ammonium nitrate solution and is immiscible with the aqueous solution). After vortexing for 10 s the suspension was centrifuged at 10,000 rpm for 2 min. An aqueous drop separated from the particles and was floating on top of the DCM. The aqueous drop and the excess DCM were taken off and the residual particles were dried under low vacuum at 80 °C for 24 h. The sample was prepared and submitted for PXRD analysis at the Australian synchrotron.

Nitrogen Release Kinetics

Fertiliser pellets manufactured as explained previously were leached under standardised conditions. The obtained release kinetics give information about the retention behaviour of ammonium nitrate and urea in MOFs in 0.01 M CaCl ₂ solution (pH 6). As this is imitating soil conditions, results may allow for predictions of fertiliser behaviour in plant experiments.

Fertiliser compositions in the form of pellets were favoured over those in the form of powder in terms of slow dissolution and hence improve the sampling accuracy. It was aimed at using approximately 25 mg of nitrogen for each dissolution experiment.

Figure 2 shows results for ammonium nitrate fertiliser composition and pure pelletised ammonium nitrate. Pure ammonium nitrate dissolved fastest (blue). 50 % of nitrogen was leached after less than 1.5 min. Regression curves for both Al-MOFs showed similar behaviour but according to the first data- points CAU-10-OH releases fertilisers slower than CAU-10-NO ₂ . The zirconium-MOFs showed similar release behaviour as well. 50 % of all nitrogen was dissolved after approximately 5 min or 5 mL of release volume. Notably no dependency on the functional group can be determined. All MOF-ammonium nitrate composites showed retention of ammonium nitrate compared to pure ammonium nitrate. The best performance in plant experiments was expected from Zr-MOFs as leaching was slowest in these MOFs. When leaching, or release profiling,was performed using the packed bed of pelletised Al-MOF fertiliser (CAU-10-OH/CAU-10-NO ₂ ) pellets appeared to have degraded completely. After 30 min the particles were completely converted into a slurry. The diffusion pathlength inside the pellets and hence the resistance to leaching can therefore be expected much lower than for the stable Zr-MOF fertiliser (UiO-66/UiO-66-NH ₂ ). The release kinetics of pure pelletised urea and urea loaded MOFs (UiO-66-NH ₂ /CAU-10- N0 ₂ ) are presented in Figure 4. Both MOFs showed similar retention behaviour for urea. 50 % of the soluble nitrogen was dissolved after approximately 7 min.

This was the slowest leaching or releasing examined in this work. CAU-10-NO ₂ also seemed to degrade completely. Taking this result into account the retention ability of CAU-10-NO ₂ is possibly even higher.

For plant growth experiments both MOFs showed promise as the urea release rate was decreased to less than 25 % of pure pelletised urea.

A comparison of CAU-10-NO ₂ and UiO-66-NH ₂ with regard to pore filling is presented in Figure 5. The plot illustrates that different MOFs incorporating the same fertiliser have different release profiles for the same nutrient. Urea dissolved and thus released only slightly slower than ammonium nitrate (release rate of 0.428 cm ^-1 for urea and 0.471 cm ^-1 for ammonium nitrate). Urea MOFs led to some release at a significantly lower rate than ammonium nitrate MOFs. CAU-10-NO ₂ urea release was approximately half as quick as CAU-10-NO ₂ ammonium nitrate. As mentioned before the nitro functional group of CAU-10-NO ₂ may be a reason for the retention of urea molecules.

Table 6 summarises the average release rates based on the regression curve for each fertiliser. Table 6 Initial average leaching rates.

The UiO-66 and U1O-66-NH ₂ MOFs showed the best retention behaviour for ammonium nitrate in 0.01 M CaCl ₂ regardless of their functional group. Urea was better retained in U1O-66-NH ₂ and CAU-IO-NO ₂ than ammonium nitrate and no difference in the urea release kinetics was determined for the two different MOFs. It also has to be taken into account that Zr-MOFs pellets remained stable when coming into contact with the leaching solution, whereas Al-MOFs degraded into a slurry. The diffusion or release rate of ammonium nitrate and urea could be decreased by incorporation in MOFs.

EXAMPLE 5 - Performance of the fertiliser compositions for plant growth

Plant Experiment with Fertiliser Composition in powder form

In preparation for planting, frame wheat seeds were pre-germinated on a wet tissue for 4 days. The soil was air dried and sieved through a sieve with 2 mm mesh size.

17 plant pots (370 mL) with a bottom diameter of 6 cm a top diameter of 7.3 cm and 4 holes for dewatering were lined with a tissue to cover the holes. 150 g of Mt Compass soil were filled to the lined pots ensuring no soil is running out of the pot's holes. With a spatula a small cavity was pressed into the soil. Powdery fertiliser compositions were filled into that cavity. Table 7 summarises the amount and type of fertiliser material applied to each pot. 50 mg Nitrogen were supposed to be in each pot (due to a later analysis of the fertiliser composition, the nitrogen content was estimated and therefore the applied nitrogen differs from 50 mg). A further 150 g of Mt Compass soil was added on top of the fertiliser composition. 178 mL of deionised water were poured into each pot to leach the soil.

The leachate was running out of the holes at the bottom of the pot and was collected overnight in saucers below the pots. 15 mL of the leachate were filtered with a 0.45 μιη syringe filter and submitted for nitrogen analysis. Leachate from pots containing ammonium nitrate as fertiliser was analysed for NH ₄ ⁺ -N and N0 ₃ ^~ -N, leachate from pots containing urea as fertiliser were analysed for total nitrogen. Soil was allowed to dry for 3 days. Three pre-germinated frame wheat seeds were then added to each pot, spaced evenly around the centre of the pot in a depth of approx. 4 cm. For each fertiliser composition four replicate pots were planted.

Table 7 Fertiliser applied to plants, first plant experiment.

Plant experiment with Pelletised Fertiliser Compositions

This was similar to the above experiment. 40 plant pots with 150 g each of soil received pelletised and sieved fertiliser material on top of the soil within a radius of 13.5 mm from the pot centre. Table 8 summarises the amount and kind of fertiliser composition applied to each pot.

Table 8 Fertiliser applied to plants, second plant experiment.

Plant Care

For both plant experiments the wheat plants were treated the same. Three nutrient stock solutions were prepared, containing all necessary nutrients for growth of the frame wheat plants. Nitrogen was only supplied by the fertiliser material and minor amounts present in the soil and the seeds. The compositions of the stock solutions are presented in Table 9.

Table 9 Stock solutions composition.

200 mL of stock solution A, 200 mL of stock solution B and 100 mL of stock solution C were added to 1.75 L deionised (DI) water resulting in 2.25 L of the so called Ruakura solution. Four days after planting the seeds were thinned out to two per pot. After that alkathene beads were added to each pot.

The soil was covered completely with one layer of alkathene beads to minimise moisture loss of the soil during the growth phase. 5 mL of Ruakura solution was initially added to each pot, thereafter 7.5 niL of Ruakura solution were added twice per week. The plants were watered daily to 50 % of their maximum water holding capacity (MWHC: 18.4 g water per 100 g soil). Knowing fertiliser, pot and dry soil weight, the desired weight after watering was calculated.

Harvesting and Analysis

35 days after planting visual signs of plant growth and nutrient deficiency were documented. Afterwards the wheat plants were harvested. Plants were cut with scissors right above the alkathene beads, stored in paper bags and dried at 40 °C for three days. Furthermore a soil subsample was taken with a boring rod. The analysis performed is listed below.

The dry, harvested plant material was weighed and then milled in a hammer mill in preparation for nitrogen analysis- The milled plant material was analysed for total nitrogen via the combustion method. The soil subsample was ground with mortar and pestle, dried at 40 °C and extracted with 2 M KC1 solution (1: 10 soikKCl solution ratio by weight) for 1 h at room temperature under shaking.

The extracts were first centrifuged and then filtered with 0.2 μιη syringe filters. Filtered extracts from pots containing ammonium nitrate as fertiliser were analysed for NH ₄ ⁺ - N and N0 ₃ ^~ -N, extracts from pots containing urea as fertiliser were analysed for total nitrogen

Planting experiments showed that when no fertiliser was applied, the soil contains similar amounts of ammonium and nitrate. With the application of fertiliser in its various forms, the results indicated that ammonium was released more than 25 times less than nitrate. The favoured releasing of nitrate is a common phenomenon. The charge imbalance of the leaching of NO ions can be compensated by leaching of positively charged metal ions for instance.

The nitrogen distribution for each fertiliser as well as for a planting pot without addition of fertiliser is presented in Figure 6. Nitrogen taken up by the plant is presented blue, nitrogen in the leachate red and nitrogen in the soil purple.

UiO-66 with fertiliser appears to provide small amounts of nitrogen to the plant, whilst the majority of nitrogen went straight through the soil with the leachate. Less than 16 % of all nitrogen was taken up by the plants and more than 75 % was leached. This performance was similar to the results for pure powdery ammonium nitrate (10.5 % uptake and 80.1 % leached). More than 67 % of all nitrogen was used for plant growth in case of UiO-66- NH ₂ . Approximately 20 % are lost through leaching. For the activated carbon fertiliser approximately 50 % of all nitrogen is taken up by the plants and approximately 25 % is leached. Activated carbon and UiO-66-NH ₂ are thus more suitable to deliver nitrogen slowly in soil. The use of nitrogen could be increased by more than 450 % compared to pure ammonium nitrate. The nitrogen reserve left in soil is increasing when plant uptake is increasing. This is indicating that nitrogen was not limited at the time of harvesting in case of UiO-66-NH ₂ and activated carbon.

Figure 7 shows the nitrogen retention abilities for the different fertiliser applied in more detail. Nitrate is leached more than ammonium in all cases. This may be due to the ammonium retention ability of the soil as mentioned above. As the powdery fertiliser was applied to the pot centre, the adjacent soil layer can retain some of the leached NH ₄ ⁺ and N0 ₃ ^~ ions.

Dry plant weight and nitrogen content of dry plant material as depicted in Table 10 further demonstrates the distribution of nitrogen. Nitrogen content of plants is less than 2 % for plants fed with UiO-66 fertiliser and ammonium nitrate. Nitrogen content is higher than 4 % for plants fed with UiO-66-NH ₂ and activated carbon fertiliser. Table 10 Average plant weight and nitrogen content with standard deviations after harvesting and drying.

Surprisingly, wheat plants grew without any addition of nitrogen sources. Nitrogen in plants (2.27 mg) can only be supplied by the seeds and minor amounts present in the soil. Thus more nitrogen is present in the soil than originally assumed (0.03 wt. %). Nitrogen deficiency was clearly visible for UiO-66 fertiliser, ammonium nitrate and no fertiliser as leaves were yellow. Plants supplied with pure ammonium nitrate and urea even have less leaves than plants supplied with the MOF-based fertiliser composition.

Zirconium MOFs (UiO-66 and UiO-66-NH ₂ ) and Activated Carbon were loaded with ammonium nitrate by soaking in aqueous 800 g/L ammonium nitrate solution. Then fertilisers containing approximately 50 mg of soluble nitrogen were applied to wheat plants in a powdery form. The results indicate that UiO-66-NH ₂ and activated carbon retain ammonium nitrate within their framework and prevent leaching. The fertiliser was still able be taken up by the plants. UiO-66 instead was unable to retain most of the ammonium nitrate. Activated carbon showed selective uptake of ammonium nitrate. UiO-66-NH ₂ differs from UiO-66 only regarding the functional group. The more hydrophilic pore surface of UiO-66-NH ₂ due to its amino-functional group might be the reason for a more effective pore filling than for UiO-66. Attractive forces are expected between ammonium and nitrate ions and the polar amino-functional group that can explain the improved retention behaviour for UiO-66-NH ₂ . For the second plant experiment an improved loading method was applied to fill porous materials with ammonium nitrate and urea respectively. The 'effective pore filling method' improved diffusion of fertiliser molecules into porous structures. Loaded materials were pelletised to slow down the nutrient release and imitate real life supply of fertiliser. The results are shown Figure 8 (ammonium nitrate is abbreviated A.N.)

For all pelletised fertiliser compositions the relative nitrogen uptake in the plants is higher than for pure ammonium nitrate and urea. The nitrogen uptake is in accordance to the leached nitrogen which decreases when plant uptake is increasing. Three MOF-fertilisers with functional groups (CAU-10-OH, UiO-66-NH ₂ and CAU-10-NO ₂ ) showed the highest nitrogen uptakes in plants concerning ammonium nitrate delivery. As for powdery fertiliser compositions the non-functionalised UiO-66 showed the lowest ammonium nitrate retention ability of all porous materials. Nevertheless the relative nitrogen uptake of plants was more than twice as determined for pure ammonium nitrate. MOFs with polar functional groups appears to be highly suitable as control release matrix possibly due to attractive forces between pore surface and ammonium nitrate ions.

Pure urea was leached less than pure ammonium nitrate. Hence nitrogen supply for the plants was increased in the case of urea. This might be due to the lower solubility of urea in comparison to ammonium nitrate. UiO-66-NH ₂ loaded with urea showed similar fertiliser retention behaviour as pure urea. The relative nitrogen uptake of the plants is only approximately 30 %. According to the results derived from pot experiments, UiO-66-NH ₂ seems to a promising MOF for the controlled release of ammonium nitrate rather than urea.

CAU-10-NO ₂ revealed the lowest nitrogen leaching of all fertiliser compositions manufactured in this work. Only 22 % of all nitrogen was lost at the initial leaching step. Nevertheless the nitrogen content determined in the harvested plant material is comparatively low (approximately 44 %) over the period of testing. The nitrogen retained in soil was more than 30 %. This indicates that some nitrogen is still retained in the pelletised MOF and may yet still be available for plant uptake. This could represent a highly stable, very slow release system. Good retention behaviour (and thus very slow release rates) of CAU-10-NO ₂ could already be predicted from its urea dissolution kinetics. A reason may be high attractive forces between the amine group of urea and the nitro group of the MOF. Figure 9 summarises the results so far.

Table 11 presents the average plant weight for each fertiliser applied and the corresponding nitrogen content. Again the pure ammonium nitrate and pure urea showed the worst performance regarding the nitrogen content of the dried plant samples. Nevertheless the dry biomass does not significantly differ from plants supplied with MOF-based fertiliser composition. Obviously nitrogen deficiency started early for those plants and growth was not inhibited for a long period of time.

Table 11 Average plant weight and nitrogen content after harvesting and drying.

Nitrogen

Plant weight Stdv.

content

[mg] [wt. %]

[wt. %]

CAU-10-OH + ammonium nitrate 688 2.86 0.24

UiO-66-NH ₂ + ammonium nitrate 619 3.66 0.75

AU-IO-NO ₂ + ammonium nitrate 617 3.44 0.08

UiO-66 + ammonium nitrate 623 2.75 0.54 ammonium nitrate 576 1.54

CAU-IO-NO2 + urea 542 4.02 0.42

UiO-66-NH ₂ + urea 548 3.13 0.58

urea 680 2.22

no Fertiliser 165 1.10 0.06

The nitrogen result is supplemented by the visual signs of the plants before harvesting. With MOF materials leaves occur fresh and dark green, whereas plants supplied with only ammonium nitrate and/or urea are paler and smaller which are signs for nitrogen deficiency. Figure 10 shows the relative nitrogen use efficiency of conventional fertilisers compared to the MOF materials. None of the fertilisers achieved 100% efficiency (data points relative to the blue line) but that is not unexpected given that in soils nitrogen can also be lost by gaseous evolution, and also that some leaching losses were expected (and measured). The MOF materials were much more effective than conventional fertiliser (red line) due to the controlled release of nitrogen from the products and the retention of nitrogen preventing leaching losses. The most effective materials per unit N applied were CAU-10-OH, activated carbon, and UiO-66-NH2. The data also indicate that introduction of polar groups into the MOF is beneficial for retention of ammonium nitrate.

Throughout this specification and the claims which follow, unless the context requires otherwise, the word 'comprise', and variations such as 'comprises' and 'comprising', 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.

The reference in this specification to any prior publication (or information derived from it), or to any matter which is known, is not, and should not be taken as an acknowledgment or admission or any form of suggestion that that prior publication (or information derived from it) or known matter forms part of the common general knowledge in the field of endeavour to which this specification relates.