Phyllosilicates and methods of preparation thereof

Abstract

A method of preparing a metal silicate hydrate material comprises contacting a calcium silicate material with a solution comprising a dissolved metal, wherein the metal comprises, consists essentially of or consists of cobalt and / or nickel, wherein the calcium silicate material comprises xonotlite, a carbonate material, and a cellulosic fibrous material.

Description

[0001] Phyllosilicates and methods of preparation thereof Field of the Invention The present invention relates to methods for preparing silicate hydrate materials. In particular, but not exclusively, the invention relates to methods for preparing cobalt silicate hydrate materials and / or nickel silicate hydrate materials. Background Certain metals, including in particular nickel and cobalt, represent a critical resource in certain industries such as rechargeable batteries, e.g. for electric vehicles, alloys, magnets, and catalysts, but have acute supply-chain risks and pose a threat to the environment. About 60 percent of global primary cobalt is produced in the Congo, much of which is then refined in other countries such as China. Small-scale artisanal mining has raised concerns from human rights organisations, and is believed to generate largeamounts of toxic waste, causing considerable health and environmental damage.As such, there is a desire globally to find solutions to remove cobalt and / or nickel from the environment in contaminated areas, and for end-of-life recycling of these metals. In parallel, certain phyllosilicate materials such as cobalt phyllosilicates and nickel phyllosilicates are known to act as catalysts in various reactions such as water- splitting reactions (Kim, J. S. et al. Amorphous cobalt phyllosilicate with layeredcrystalline motifs as water oxidation catalyst. Adv. Mater. 29, 1606893, DOI:10.1002 / adma.201606893 (2017)) and CO2 methanation in the case of nickelphyllosilicates. These may find use for example as metal catalysts in the production ofhydrogen. These materials are typically produced in small quantities, for example by co-precipitating aqueous solutions of cobalt chloride (CoCl2) with sodium metasilicate Na2SiO3. Whilst the immobilisation of heavy metals in calcium silicates has been generally investigated previously, no disclosure has been made of the preparation of a cobalt or nickel phyllosilicate. Examples of such generic work include: -Chen, Q., Tyrer, M., Hills, C. D., Yang, X. & Carey, P. Immobilisation of heavymetal in cement-based solidification / stabilisation: A review. Waste Manag. 29, 390–403, DOI: 10.1016 / j.wasman.2008.01.019 (2009) 55499624-1- Qi, G. et al. Preparation and evaluation of a mesoporous calcium-silicate material(MCSM) from coal fly ash for removal of Co(II) from wastewater. Chem. Eng. J. 279, 777–787, DOI: 10.1016 / j.cej.2015.05.077 (2015) -Baldermann, A. et al. Uptake of aqueous heavy metal ions (Co2+, Cu2+ and Zn2+)by calcium-aluminium-silicate-hydrate gels. Cem. Concr. Res.147, 106521, DOI: 10.1016 / j.cemconres.2021.106521 (2021) -Komarneni, S., Roy, D. & Roy, R. Al-substituted tobermorite: shows cationexchange. Cem. Concr. Res.12, 773–780, DOI: 10.1016 / 0008-8846(82)90041- 2 (1982) -Komarneni, S. & Roy, D. M. Tobermorites: a new family of cation exchangers.Science 221, 647–648, DOI: 10.1126 / science.221.4611.647 (1983) -Komarneni, S. & Roy, D. M. Method of storing radioactive wastes using modifiedtobermorite (1985) -US Patent 4,537,710 (Komarneni et al)- Komarneni, S., Roy, R. & Roy, D. M. Pseudomorphism in xonotlite andtobermorite with Co2+and Ni2+exchange for Ca2+at 25°C. Cem. Concr. Res.16, 47–58, DOI: 10.1016 / 0008-8846(86)90067-0 (1986) -Komarneni, S., Breval, E., Roy, D. M. & Roy, R. Reactions of some calciumsilicates with metal cations. Cem. Concr. Res.18, 204–220, DOI: 10.1016 / 0008- 8846(88)90005-1 (1988) -Shrivastava, O. & Glasser, F. Ion-exchange properties of 11-Å tobermorite.React. Solids 2, 261–268, DOI: 10.1016 / 0168-7336(86)80089-4 (1986) -Komarneni, S. Heavy metal removal from aqueous solutions by tobermorites andzeolites. Nucl. Chem. Waste Manag. 5, 247–250, DOI: 10.1016 / 0191- 815X(85)90001-4 (1985) It is an object of the invention to address and / or mitigate one or more problems associated with the prior art. It is an object of the invention to provide a means of removing cobalt and / or nickel from the environment in contaminated areas. It is an object of the invention to provide an improved method for preparing metal phyllosilicate catalyst materials such as cobalt phyllosilicates and / or nickel phyllosilicates. 55499624-1 Summary The present invention is based on the finding that it is possible to remove certain heavy metals such as cobalt and / or nickel from an aqueous solution whilst forming metal phyllosilicate materials such as cobalt phyllosilicates and / or nickel phyllosilicates. According to a first aspect, there is provided a method of preparing a metal silicate hydrate material, the method comprising: contacting a calcium silicate material with a solution comprising a dissolved metal, wherein the metal comprises, consists essentially of or consists of cobalt and / or nickel. The dissolved metal may comprise, may consist essentially of, or may consist of,Co(II) and / or Ni(II). Typically, the solution may be an aqueous solution. The calcium silicate material may comprise, may consist essentially of, or mayconsists of, xonotlite. The calcium silicate material may comprise, may consist essentially of, or mayconsist of, Ca6Si6O17(OH)2. The calcium silicate material may comprise at least 80 wt%, e.g. at least 85 wt%, e.g. at least 90 wt%, xonotlite. The calcium silicate material may further comprise a carbonate material, e.g. calcite. The calcium silicate material may comprise less than 15 wt%, e.g. less than 10 wt%, e.g. less than 7 wt%, of the carbonate material, e.g. calcite. The calcium silicate material may further comprise a fibrous material such as a cellulosic material. The calcium silicate material may comprise less than 10 wt%, e.g. less than 5 wt%, e.g. less than 3 wt%, of the fibrous material, e.g. cellulose fibres. The calcium silicate material may comprise or may consist of a material such as CALSITHERM®, SILCA®, MICROCAL®, Klimaplatte, or the like. Conveniently, thesematerial are typically used as building materials (termed “CS” board) and are low cost.Typically, the calcium silicate material may comprise: oxonotlite in an amount of at least 80% by weight of the calcium silicatematerial; oa carbonate material, e.g. calcite, in an amount of 15% or less by weightof the calcium silicate material, e.g. in an amount of about 1-10% by weight of the calcium silicate material; and55499624-1o a fibrous material, e.g. a cellulosic material, in an amount of about 10%or less by weight of the calcium silicate material, e.g. in an amount ofabout 1-10% by weight of the calcium silicate material. Advantageously, the use of CS board as the calcium silicate material isassociated with a number of non-obvious benefits in the context of the present invention,such as: -Improved reactivity of the xonotlite component towards Co and / orNi, compared to xonotlite alone; -The ready availability of CS as self-supporting sheets and blockswhich are capable of retaining their structural integrity during and after Co and / orNi sequestration. This is believed to be due to the presence of a fibrous material such as cellulosic fibres as a reinforcing agent in the CS board. The integrity ofthe material is advantageous, allowing it to be used as blocks and / or filter discsto treat a solution; -High porosity (around 0.9 or 90 percent by volume) which allowseasy penetration of a solution to be treated (such as waste water, tailings pondsin mining, and process liquors) into the material. This, in turn, allows the solutioninto direct and close contact with the large surface area of the xonotlitecomponent. -High rate of penetration of the solution into the material. In anembodiment, the sorptivity may be about 5-20 mm / min1 / 2, e.g. about 9.5mm / min1 / 2. This, combined with the high porosity means that the dry CS spongeabsorbs water both rapidly and in large quantities. For example, 1 m3 of CS maybe capable of absorbing about 0.8 m3of water. In some embodiments, the surface area of the calcium silicate material may beabout 10-30 m2 / g, e.g. may be about 20 m2 / g. In some embodiments, the bulk density of the calcium silicate material may beabout 100-500 kg / m3, e.g. about 200-300 kg / m3, e.g. about 267 kg / m3.The metal silicate hydrate material may be a metal phyllosilicate material. The metal silicate hydrate material may be amorphous or substantially amorphous. The metal silicate hydrate material may be an amorphous metal phyllosilicate material. 55499624-1The metal silicate hydrate material, e.g. the metal phyllosilicate material, mayhave the formula (I): M6Si6Ox(OH)y.zH2O (I)where M is Co and / or Ni;x = 4+z; and y = 28−2z Typically, z≈3, for Co and z≈4 for Ni. In an embodiment, the metal may comprise, may consist essentially of, or mayconsists of, cobalt, e.g. Co(II). The metal silicate hydrate material, e.g. the metal phyllosilicate material, may have the formula (Ia): Co6Si6O7(OH)22.3H2O (Ia)In an embodiment, the metal may comprise, may consist essentially of, or mayconsists of, nickel, e.g. Ni(II). The metal silicate hydrate material, e.g. the metal phyllosilicate material, may have the formula (Ib): Ni6Si6O8(OH)20.4H2O (Ib)Advantageously, the method may be carried out at ambient temperature or at room temperature, e.g. between about 5 and 40 °C, e.g. between about 10 and 35 °C, e.g. between about 15 and 30°C, e.g. between about 15 and 25°C, e.g. e.g. between about 20 and 25°C. Advantageously, the method does not include heating and / or cooling.Advantageously, the method does not include cooling. Advantageously, the method does not include heating. Advantageously, the method does not include heating and does not include cooling. The method may comprise contacting the calcium silicate material with the solution comprising the dissolved metal for at least 1 hour, e.g. at least 6 hours, e.g. at least 12 hours, e.g. at least 24 hours. Advantageously, the method permits large-scale sequestration of cobalt and / or nickel in low-cost materials under mild conditions, whilst simultaneously generating a phyllosilicate material capable of acting as a catalyst material in various reactions, e.g. water hydrolysis. 55499624-1 It was found that the use of a calcium silicate material such as xonotlite allows the sequestration of cobalt and / or nickel ions from solution with an uptake multiple times,e.g. at least five times, greater than that of other conventional materials, such as clays,zeolites and carbons, and several hundred times greater than its accumulation in plantsand biofilms. Advantageously, it was found that sequestration kinetics are zero-order in Co concentration, such that the rate of sequestration is maintained and / or is substantially constant even at low cobalt concentrations. Thus, provided a sufficient amount of the calcium silicate material is present, using this material permits sequestration ofsubstantially all, e.g. more than 90%, e.g. more than 95%, e.g. more than 99%, of themetal (e.g. cobalt and / or nickel). The method may comprise removing at least a portion of the metal (e.g. cobalt and / or nickel), from the solution. The method may comprise sequestering at least 90%, e.g. at least 95%, e.g. at least 99%, of the metal (e.g. cobalt and / or nickel), from the solution. The method may comprise releasing calcium into the solution. Conveniently, the method may comprise providing the calcium silicate material as a block of material. A “block” will be herein understood as a 3-dimensional volume ofthe material in solid form, and / or not provided in comminuted form such as a powder ora granule. The block may have any suitable shape such as a cube, a cuboid, a sphere, a cylinder, any other regular shape, or any irregular shape. Typically, the block may have a dimension, e.g. a height, length, width, and / or diameter, of at least 1 mm, e.g. at least 1 cm, e.g. at least 2 cm, e.g. at least 5 cm, e.g. at least 10 cm. Advantageously, using the calcium silicate material as a block mayimprove the ease of use of the material for decontamination, for example in freshwateror seawater environments, such as lakes, rivers, or coastal waters. In an embodiment, a block may have a length, width, and / or height of at least 100mm. A block may have a block may have at least two of a length, width, and height, of at least 100mm. A block may have a length, width, and height of at least 100 mm. The block may have a volume of at least 1 cm3, e.g. at least 10 cm3, e.g. at least 100 cm3, e.g. at least 1000 cm3. For example, if the block has a cubic shape, the block may have minimumdimensions of about 100 x 100 x 100 mm.55499624-1 For example, if the block has a cylindrical shape, the block may have minimumdimensions of 100 mm diameter base and a length of at least 50 mm.Alternatively, the method may comprise providing the calcium silicate material in comminuted form, e.g. as a granule, pellet, powder or the like. According to a second aspect, there is provided a metal silicate hydrate materialobtained or obtainable by a method according to the first aspect.The metal silicate hydrate material may be a metal phyllosilicate material. The metal silicate hydrate material, e.g. the metal phyllosilicate material, may have the formula (I): M6Si6Ox(OH)y.zH2O (I)Where M is Co and / or Ni;x = 4+z; and y = 28−2z. Typically, z≈3, for Co and z≈4 for Ni. In an embodiment, the metal may comprise, may consist essentially of, or mayconsists of, cobalt, e.g. Co(II). The metal silicate hydrate material, e.g. the metal phyllosilicate material, may have the formula (Ia): Co6Si6O7(OH)22.3H2O (Ia)In an embodiment, the metal may comprise, may consist essentially of, or mayconsists of, nickel, e.g. Ni(II). The metal silicate hydrate material, e.g. the metal phyllosilicate material, may have the formula (Ib): Ni6Si6O8(OH)20.4H2O (Ib)The present method may find applications in remediating and cleaning water and industrial wastewater from such activities as: mining, electroplating, battery recycling and more. The reaction product of the present method may be a poorly crystalline cobalt phyllosilicate or nickel phyllosilicate. Particular applications for such phyllosilicate compounds include catalyst precursor materials for several industrial reactions such as: water splitting, hydrogenation of furfural to furfuryl alcohol, and CO2 methanation. 55499624-1 The features described in relation to any aspect of the invention may equally apply to any other aspect and, merely for brevity, are not repeated. For example, features described in relation to compositions can apply in relation to methods, and vice versa. Brief Description of Figures The present invention will now be further described in detail and with referenceto the figures, which show:Figure 1a: powder X-ray diffraction [XRD] pattern of calcium silicate; Figure 1b: mid-IR spectrum of CS;Figure 2: Solution composition during the Co–CS sequestration reaction,showing: a–e: Concentrations of Ca (blue), Co (orange), Si (green); which are the means of three measurements with error bars showing the range; horizontal dotted lines showthe initial Co concentration b(Co)0; f: Amount of Ca released nCo vs amount of Coremoved nCa: composite plot of data from all solutions; regression line nCa= (1.02±0.03)nCo; g: Photographs of samples (decreasing initial concentration from left to right) taken before the solutions were decanted for analysis, loss of colour highlighting the progressive removal of Co from solution; Figure 3: Changes in solids composition during sequestration: a: XRD patterns of solid material formed by reaction of CS with Co(NO3)2solution of initial concentration b(Co)0=0.35m after 2h (red), 2d (blue) and 5d (black); tick marks show positions of Co-phyllosilicate (hk) reflections (also ICDD 21-871, 21-872); b: Mid-IR spectrum of thesame solid material shown in a; c,d: X-ray absorption spectra of the final product after 56d reaction and the standards used (Co(OH)2, Co2+(aq), Co3O4), where χµ(E) is the normalised XANES (c), |χ(R)| is the Fourier transform of the k3-weighted EXAFS χ(k) (d), and k is the wavenumber. Solid black lines (—) represent the final fits to the Fourier transform of the EXAFS; Figure 4: Kinetic phenomena in Co sequestration by CS: observations of stirredpowder reactions (25 mL solution mixed with 1.00 g CS powder) at 25 ◦C. a: Initial Coconcentration (b(Co)0 =0.35m), showing release of Ca2+into solution (Ca, blue points, ICP measurement), decrease of crystalline xonotlite (Xon, green points, XRD measurement), and growth of amorphous Co-S-H phase (Co-S-H, orange points, XRDmeasurement); y denotes fractional change; b: Decrease of Co2+ in solution at threeinitial molal concentrations: b10.35m (red points), b20.073m (grey points), b30.037m 55499624-1 (black points); dashed lines are linear fits to data, with best-fit rate constant k = 0.078±0.006, 0.072±0.005 and 0.071±0.001 mol / (kg CS h) respectively for the three concentrations; mean 0.074±0.004 mol / (kg CS h); Figure 5: images illustrating Sharp Front experiment and model: a: Photograph(left hand side) of a reacted sample after penetration of Co2+into a CS bed at 25◦C, showing the well-defined reaction zone (pink) and reaction front at x = xf. The initial Co(NO3)2concentration = 0.33 m and CS bed volume fraction porosity fb= 0.91. The sample was water saturated to avoid capillary transport of Co2+. The image was taken at t = 894hours and the reaction front position xfis 16.4mm below bed surface; b: Schematic of Sharp Front model (see text): xf(t) is the position of the reaction front at time = t, Lsthe length of the solution-filled part of the tube, cb(t)′the Co concentration at the bed surfaceand jm the Co mass flux across the bed surface; c: Photograph (bottom) showing howthe location of the reaction front is determined from analysis of the green channel image intensity (top), here 12.51mm at the midpoint of the erfc(x) fit (—), measured from the bed surface; Figure 6: Position of the Co sequestration reaction front xf(t): experimental data (circles) fitted to the Sharp Front model (solid line), Eqn 7: t = α0[−xf−α1ln(1−xf / α1)], with best-fit parameters α0= 44.53±1.17mm−1h, α1= 18.13±0.04mm. In this test, b(Co)0= 0.331m, Ls= 73.5mm, ρb= 203.5kg / m3and A = 90.1mm2. Inset shows that xfvaries as t1 / 2at early time, with the best-fit slope = 0.658±0.003mm h−1 / 2; Figure 7: Advance of the Co sequestration front in CS: four experimental datasets scaled to the dimensionless form of the SF model, Eqn 8 (−). Data as in Fig 6, blue filled circles; orange filled circles, b(Co)0 = 0.331m, Ls = 47.8mm, ρb = 139.9kg / m3; black open circles, b(Co)0 = 0.331m, Ls = 36.9mm, ρb = 221.4kg / m3; green open circles, b(Co)0 = 0.161m, Ls = 38.9mm, ρb = 273.6kg / m3. Inset shows the unscaled experimental data (xf,t) fitted to Eqn 7. For fit parameters, blue filled circles, see Fig 6; orange filled circles, α0 = 22.52±0.96mm−1h, α1 = 16.82±0.08mm; black open circles, α0 = 26.01±0.80mm−1h, α1 = 8.04±0.05mm; green open circles, α0 = 39.50±2.87 mm−1h, α1 = 3.60±0.04mm; Figure 8: Sharp Front analysis of four experimental tests of Co sequestration by replacement of Ca in CS beds (symbols key as Fig 7; uncertainties shown graphically);a: Replacement of Ca by Co at equilibrium, dashed line nCo = (1.00±0.01)nCa; b: Transportparameter K, showing increase of K with bed porosity fb; Figure 9: Solution composition during the Ni–CS sequestration reaction,showing: a–d: Concentrations of Ca (blue), Ni (orange), Si (green) which are the meansof three measurements with error bars showing the range; horizontal dotted lines show 55499624-1 the initial Ni concentration b(Ni)0; e: Amount of Ca released nNi vs amount of Ni removed nCa: composite plot of data from all solutions; regression line nCa = (1.06±0.01)nNi; f:Variation of pH in Ni-CS slurries during sequestration reaction: the initial Ni(NO3)2concentration b(Ni)0(m) 0.33 (pink), 0.071 (green), 0.034 (brown), 0.0035 (blue), 0control (orange), decreases upwards from the bottom;Figure 10: Changes in solids composition during sequestration: a: XRD patterns of solid material formed by reaction of CS with Ni(NO3)2solution of initial concentration b(Ni)0=0.33m after 2h (red), 15h (blue) and 2d (green); tick marks show positions of Ni- phyllosilicate (hk) and (00l) reflections); b: Mid-IR spectrum of the same solid material shown in a; Figure 11: Kinetic phenomena in Ni sequestration by CS: observations of stirredpowder reactions (25 mL solution mixed with 1.00 g CS powder) at 25 ◦C. a: Initial Niconcentration (b(Ni)0=0.33m), showing release of Ca2+into solution (Ca, blue points, ICP measurement), decrease of crystalline xonotlite (Xon, green points, XRD measurement), and growth of amorphous Ni-S-H phase (Ni-S-H, orange points, XRD measurement); ydenotes fractional change; b: Decrease of Ni2+ in solution at three initial molalconcentrations: b10.33m (red points), b20.071m (grey points), b30.034m (black points); dashed lines are linear fits to data, with best-fit rate constant k = 0.287±0.022,0.268±0.008 and 0.242±0.052 mol / (kg CS h) respectively for the three concentrations;mean 0.266±0.019 mol / (kg CS h);Figure 12: Sharp Front experiment: Time-series photographs of penetration ofNi2+ into a CS bed at 25°C, showing the well-defined reaction zone (green) and reactionfront at x = xf. The CS bed is contained in a flat-bottomed glass tube, i.d. 10.44 mm. Initial Ni(NO3)2 concentration = 0.33 m and CS bed volume fraction porosity fb = 0.93.The CS bed was pre-saturated with water to suppress capillary transport. The image onthe right taken at t = 1505 h shows the reaction front position xfinf when the sequestrationreaction is complete. Figure 13: Advance of the Ni sequestration front in CS: three experimentaldatasets scaled to the dimensionless form of the SF model, Eqn 6 (−). Orange filledcircles, b(Ni)0 = 0.332m, Ls = 67.7mm, ρb = 200kg / m3; green open circles, b(Ni)0 = 0.332m,Ls = 35.0mm, ρb = 185 / m3; blue filled circles, b(Ni)0 = 0.158m, Ls = 38.5mm, ρb = 240kg / m3;Inset shows the unscaled experimental data (xf,t) fitted to Eqn 5. For fit parameters, orange filled circles, α0 = 47.23±3.42mm−1h, α1 = 16.27±0.12mm; green open circles, α0= 23.83±1.40mm−1 h, α1 = 9.47±0.05mm; blue circles, α0 = 46.35±10.01 mm−1 h, α1 =3.85±0.10mm; 55499624-1Figure 14: Sharp Front analysis of four experimental tests of Ni sequestration byreplacement of Ca in CS beds (symbols key as Fig 13; uncertainties shown graphically);a: Replacement of Ca by Ni at equilibrium, dashed line nNi = (1.03±0.01)nCa; b: Transportparameter K, showing increase of K with bed porosity fb. Detailed Description In the present disclosure, reference is made to a number of terms, which havethe meanings provided below, unless a context indicates to the contrary. The nomenclature used herein for defining compounds, in particular the compounds according to the invention, is in general based on the rules of the IUPAC organisation for chemical compounds, specifically the “IUPAC Compendium of Chemical Terminology (Gold Book)”. For the avoidance of doubt, if a rule of the IUPAC organisation is in conflict with a definition provided herein, the definition herein is to prevail. Furthermore, if a compound structure is in conflict with the name provided for the structure, the structure is to prevail. The term “comprising” or variants thereof is to be understood herein to imply the inclusion of a stated element, integer or step, or group of elements, integers or steps, but not the exclusion of any other element, integer or step, or group of elements, integers or steps. The term “consisting” or variants thereof is to be understood to imply the inclusion of a stated element, integer or step, or group of elements, integers or steps, and the exclusion of any other element, integer or step or group of elements, integers or steps. The term “about” herein, when qualifying a number or value, is used to refer tovalues that lie within ± 5% of the value specified. For example, if a temperature isspecified to be about 5 to about 13 °C, temperatures of 4.75 to 13.65 °C are included. Reference to physical states of matter (such as liquid or solid) refer to the matter’s state at 25 °C and atmospheric pressure unless the context dictates otherwise. As explained above, the present inventors have discovered that it is possible to remove certain heavy metals such as cobalt and / or nickel from an aqueous solution whilst forming metal phyllosilicate materials such as cobalt phyllosilicates and / or nickel phyllosilicates. 55499624-1 Examples Experimental Materials The sequestrant used is Calsitherm® insulation board (Calsitherm® Silikatbaustoffe GmbH, Paderborn), a commercial product manufactured from lime and silica by steam autoclaving at 150–200◦C for 15–25h. Analysis of solutions Solution compositions are reported as molality b (unit mol / kg water, denoted mol / kgw or m). In transport models we use the amount concentration c (mol / m3) and themass concentration c′ = cMA (kg A / m3), where MA is the molar mass of substance A.Samples were prepared by adding 25mL of Co(NO3)2solution (concentration range 0.35m to 0.0035m) to 1.00g of CS in 50mL centrifuge tubes, and agitated on a shaker tray for 10min up to 56d. Each supernatant was separated by centrifuging, the liquid phase acidified and diluted with 3 percent HNO3 and divided into triplicate samples forICP-OES analyses. The solid was washed in DI water, centrifuged and dried to constantweight at 40◦C. The slurry pH was measured before centrifuging.Characterisation of solids Total carbon [TC] and total organic carbon [TOC] analyses TC and TOC analyses were carried out by Elemental Microanalysis Ltd (Devon, UK). For TC, the dried samples (40◦C) were weighed into silver capsules, placed in a combustion tube (1000◦C) and burned in pure oxygen. Combustion gases were passed over catalysts to ensure complete oxidation and absorption of halogens, sulphur and other interferences. CO2was separated on a chromatographic column and quantified using a thermal conductivity detector. TOC analyses were carried out in the same way on samples pre-treated with 15 M HCl to remove carbonates. X-ray fluorescence [XRF] XRF analyses were carried out by AMG Analytical Services Ltd. Analysis was carried out on powdered samples fused into glass beads with Li2B4O7 at 1270◦C and on six certified reference materials using a PANalytical AXIOS wavelength-dispersive instrument. 55499624-1 X-ray diffraction [XRD] XRD patterns were collected using a PANalytical Empyrean system in reflection mode (Bragg-BrentanoHDmodule) with a Cu anode (45kV and 40mA). The detector PHD lower level was set to 50percent, optimised to suppress fluorescence from the Co- containing samples. Slit and Soller settings were chosen to avoid beam spill on the sample (dia 16mm). Data were collected on powdered samples in back-loading holders from 5–80◦2θ, step size 0.0263◦2θ, count time 2s per step, samples spinning at 2rev / s. Quantification was based on a Partial-Or-No-Known-Crystal-Structure [PONKCS] method (55,56). As the ICSD structure files do not match exactly the hydrothermal xonotlite in CS, an ‘hkl phase’ with PONKCS style cell mass was developed from calibration mixes with CaF2(Sigma-Aldrich, >99.9percent) as a standard. A similar approach was used for the amorphous Co-S-H. TOPAS v5 and EVA v5.2 (Bruker Ltd) were used for data analysis and phase identification. Fourier transform infrared spectroscopy Powdered samples were analysed using an Agilent 4500a ATR FTIRspectrometer. Each sample was run with 64 scans at a resolution of 4 cm−1. X-ray absorption spectroscopy XAS Spectra were obtained from solid samples collected from all b(Co)0concentrations after 56d and from Co2+ (aqueous), β-Co(OH)2 and Co3O4) standards.Dried solids were diluted with BN and pressed into a 100–400 µm thick pellets for analysis at beamline B18 at Diamond Light Source. The Co K-edge (7.709keV) was measured using a Si(111) monochromator. Samples b(Co)0 = 0.35m and 73 mm, Co(OH)2 and Co3O4 were analysed in transmission mode, and samples b(Co)0 = 0.037m and 0.0035m, and aqueous Co2+in fluorescence mode with a 36-element Ge detector. Spectra were processed with DEMETER software package. ATHENA was used for data reduction and background subtraction and ARTEMIS to fit the R-space Fourier transform of the EXAFS spectra. Reaction front analysis The positions of the reaction front xf were obtained from images of the sedimented CS bed. The position of the bed surface did not change during the reaction, within the precision of the measurement, ≈ ±0.02mm. The green-channel intensity had the greatest RGB contrast and was extracted from the image pixel by pixel along the bed 55499624-1 (≈ 4.6µm / pixel). The front location xf was taken as the midpoint of an error-function fit(Figure 5c), with typical standard uncertainty 0.013mm. For fitting xf(t) data, Eqn 7 (seebelow for more detail) may be written as t= α0[−xf −α1 ln(1−xf / α1)] (9)with regression parameters α0 = Ls / K and α1 = xf∞. Errors and uncertainties Errors in measured quantities are either k = 2 expanded uncertainties estimated from standard deviations of replicates or estimates based on known accuracy of laboratory measurements. Errors in parameters obtained by least-squares regression are k = 2 uncertainties calculated from standard errors of regression with 0.95 confidence interval. Errors in derived quantities are then calculated by standard methods of error propagation. Results Calcium Silicate characterisation Calcium Silicate (‘CS’) consists largely of the calcium silicate hydrate mineral xonotlite Ca6Si6O17(OH)2(Xon 91.3 wt%), with calcite (Cal 6.3 wt%) and cellulose fibres (2.4 wt%) as minor components. Only xonotlite is believed to be chemically active in contact with cobalt solutions. Figure 1a shows the powder X-ray diffraction [XRD] pattern of calcium silicate,which apart from the absence of the weak reflection at 18.4◦ 2θ is in excellent agreementwith reference pattern ICDD 23-125 of an autoclaved xonotlite synthesised from lime and silica at 200◦C for 24h. Xonotlite has a strongly fibrous habit. The fibre axis is parallel to silicate double chains which lie along the

[0010] crystallographic direction. Double chains are formed by bridging-oxygen crosslinks shared between every third silicate tetrahedron of individual chains. Parallel double chains then lie between sheets of Ca polyhedra in the (001) crystallographic plane. In the ideal formula, xonotlite contains no molecular water; OH groups are located in Ca octahedra of the sheets. The mid-IR spectrum of CS is shown in Figure 1b. The main bands include the sharp band at 1197cm−1assigned36,37to the Si–O–Si stretching mode of the crosslinks of the double chain, the intense band at 960cm−1and the weaker one at 1068cm−1assigned to Si–O–Si stretching vibrations of non-bridging tetrahedra. The only feature in the water region 3000–4000cm−1is the small band at 3611cm−1assigned to the OH groups of 55499624-1 xonotlite. There is no evidence of molecular water in CS. Weak bands at 1425, 875 and 710cm−1match reference spectra of calcite. Co and Ni sequestration Solutions analysis The sequestration of Co by CS was tested by reacting powdered CS with aqueous Co(NO3)2 solutions of initial concentration b(Co)0 0.35, 0.073, 0.037 and 0.0035m (where m denotes molal concentration: mol / kg water) for periods of 2, 4, 6, 8hours, 1:14, 28 and 56 days. In all cases, 1.00g of CS was mixed with 25.0mL aliquotsof Co solution or a control aliquot of deionised water. The highest solution concentration contained slightly more Co than needed to completely replace Ca in the available xonotlite. At each sampling time, the supernatant was analysed for the Ca and Si released from CS, and for the Co remaining in solution. The results are illustrated in Figures 2a-2f. Fig 2a, Fig 2b and Fig 2c show that the removal of Co from solution isaccompanied by a synchronous and equivalent release of Ca into solution. The time forcomplete removal decreases with decreasing Co concentration in solution and takes roughly 100h at the highest concentration used (0.35m). At all times the concentration of dissolved Si is extremely small, generally less than 1.0±0.1mm (30ppm). In Figure 2d the initial amount of Co in the solution (0.0035 m) is removed in lessthan 1 hour. The final concentration of Ca in solution is close to the initial concentration of Co for all concentrations used, showing the same Co removal behaviour at both low and high concentrations. After reaction, the residual Co is extremely low. In solutions that remain in contact with excess CS, the mean Co concentration from dailymeasurements over the period from 4 to 9 days after first contact with CS is extremelysmall (0.06±0.04ppm) and below the method quantification level. The control solution (zero Co), as shown in Figure 2e, provides a useful baseline. The concentration of dissolved Si is in the range 1-5—3.0mm, while the Ca concentration is 1.0–1.5mm. These low concentrations (40—80ppm) are approximately as calculated from the known solubilities of xonotlite and calcite. Figure 2f shows that the amount of Co sequestered is equal to the amount of Careleased at all time steps and in all solutions. The replacement is stoichiometric throughout the reaction, with n(Ca) / n(Co) = 1.02±0.03. These observations show that sequestration occurs through the formation of a Co silicate phase, since the Si concentration in solution does not change during the 55499624-1 reaction. This product phase is denoted as ‘Co-S-H’ hereinafter for simplicity, followingcement chemistry notation in which ‘S’ is used for SiO2 and ‘H’ for H2O, with dashesindicating that the stoichiometry may be variable or not known. The clean-up of the strongly coloured Co2+ion from solutions of different initialconcentration is shown visually in Figure 2g.The sequestration of Ni by CS was tested by reacting powdered CS with aqueousNi(NO3)2 solutions of initial concentration b(Ni)00.33, 0.071, 0.034 and 0.0035m (wherem denotes molal concentration: mol / kg water) for periods of 2, 4, 6, 8 hours, 1:14, 28and 56 days, following the same methodology as with Co. The results are illustrated in Figures 9a-9f. Similar observations for Ni sequestration can be made as those summarised above for Co sequestration. Characterisation of the reaction product The product of the CS reaction with solution b(Co)0=0.35m, in which all Ca initially present is replaced by Co, is amorphous as shown by XRD and FTIR analyses (Fig 3a, Fig 3b). The only remaining sharp reflection is from minor calcite which does not take part in the sequestration reaction. The xonotlite reflections from the untreated CS areabsent in the fully reacted solid. There are two new features not present in xonotlite: abroad asymmetric reflection with a maximum intensity at 34.4◦2θ (d-spacing: 0.260nm), and a broad symmetrical reflection centred at 59.4◦2θ (d-spacing: 0.155nm). These features resemble the 2D (hk) diffraction bands first described in disordered layer materials such as graphite40and the sheet silicate halloysite. These broad reflections are also found in Co phyllosilicates produced in small quantities by co-precipitating aqueous solutions of cobalt chloride CoCl2with sodium metasilicate Na2SiO3rather than sequestering Co in a preformed solid material as we do here. Before the time required for full replacement (5d for b(Co)0=0.35m), Co-S-H and xonotlite co-exist as shown in Figure 3a. When the solution does not contain enough Co to replace all Ca in the xonotlite contained in 1.00g of CS, as in b(Co)0=3.5–73mm, Co-S-H and xonotlite also co-exist in the final product. Quantification of the xonotlite remaining and reaction product formed over time using b(Co)0=0.35m is shown in Figure 4a. From the observed.535nmand b = 0.930d-nm (with spacings of the b ≈ √3(hka)), in close agreement with thoseobtained previously bands in the reaction product, we obtain lattice spacings for thesilicate sheet for the Co phyllosilicates Co6Si8O10(OH)4 and Co6Si4O10(OH)8. This55499624-1 indicates that the reaction product although highly disordered contains intact fragments of silicate sheet structures. Further confirmation of this comes from FTIR and XAS analyses. FTIR spectra of Co-S-H lack the band at 1197cm−1assigned to the crosslinks in the double-chain structure of xonotlite, showing that the double chain structure is dismantled. The band at 1068cm−1broadens to a shoulder on formation of Co-S-H. Further evidence of silicate chain disruption comes from changes to the strong band at 960cm−1in the unreacted xonotlite (Si–O–Si stretching vibrations of non-bridging tetrahedra). In Co-S-H this is replaced by a strong, broad band at 985cm−1, assigned to the Si–OH stretching mode in silanol groups. The product is more highly hydrated than the starting CS as shown by broad bands in the region 3300–3650cm−1present in the reaction product but not in CS itself. The feature at 1640cm−1, also present only in the reacted product, is assigned tothe H–O–H bending mode of molecular water (not from OH groups), and confirms thatsome molecular water is present in the product. Together, these analyses show that the double chain structure of xonotlite has broken down. The Co-S-H product is hydrated and largely amorphous but has phyllosilicate-like features. XANES spectra (Fig 3c, Fig 3d) of standards and samples show that Co does not oxidise during reaction with xonotlite in the presence of NO3–, and that Co(OH)2is not present in the reacted samples. The Co co-ordination environment is similar in Co- S-H produced at all b(Co)0concentrations. To understand the Co-S-H structure, EXAFS spectra are fitted with relevant and available structures. As structural information does not exist for hydrated cobalt silicates / phyllosilicates, we choose hydrated silicates of Ni, Mg and Cu as their cationradii (0.83, 0.86, 0.87 Å, respectively) are similar to that of Co (0.89 Å) and the metalions have 6-fold co-ordination. We choose phyllosilicates and single and double chain silicates to determine if the chain structure of xonotlite is preserved. The models based on chain structures includes the metal-metal and metal-Si scattering paths at interatomic distances (2.6-2.8 Å, and 3.4-3.5 Å) that could not be fitted to the Co EXAFS. The structure of the Co-S-H phase is therefore not similar to these inosilicates. On the other hand, all Co EXAFS spectra could be fitted with the models based on the phyllosilicates. These fits include three distinct scattering paths which all statistically improved therespective fits: a Co-O scattering path with a coordination number [CN] of 6 at 2.085 Å,a Co-Co scattering path with a CN of 4.3-4.9 at 3.13 Å, and a Co-Si scattering path with a CN of 3.9-5.6 at 3.31 Å. We conclude that structure of the Co-S-H phase resembles a phyllosilicate, confirming the XRD and FTIR results. Furthermore, the EXAFS results 55499624-1 highlight that this phyllosilicate-like Co-S-H phase is the dominant reaction product even at the lowest concentration: b(Co)0=0.0035m. The product of the CS reaction with solution b(Ni)0=0.33m, in which all Ca initially present is replaced by Ni, is amorphous as shown by XRD and FTIR analyses (Fig 10a,Fig 10b). Similar comments apply mutatis mutandis during sequestration of Ni.Mass change on sequestration Accurate measurements of sample mass show that considerable water is taken up in forming Co-S-H. In tests on small blocks of CS the mass increased by 46 percent (Table 1) on Co-treatment, of which only 16 percent is accounted for by replacing of Ca by Co. Table 1. Mass change of CS blocks on complete Co reaction Sample MassMol Mol Bulk Solid Porosity ratio ratio ratio density density wa / wb nH2O / nXon ∆nH2O / nXon ρb ρs fkg / m3kg / m3Note a Note b Note c Note dA 1.483 13.8 3.2 415 2760 0.850B 1.482 13.8 2.8 – – –C 1.438 11.9 2.8 400 2635 0.849D 1.452 12.5 3.0 395 2675 0.852Mean 1.464 13.0 2.9 400 2690 0.850CS270 2540 0.895control Notes: (a) Sample weights before reaction (wb), and after reaction (wa). All samples were conditioned over LiCl saturated solution at 25.0◦C (RH 11.3percent) prior to weighing. (b) Mol ratio where nH2O is the water incorporated in the reaction product calculated from weight gain (wa−wb) after allowing for the mass change associated with replacement of Ca by Co; and nXon is the xonotlite amount before reaction. (c) ∆nH2O is calculated from weight loss on conditioning the reacted samples over molecular sieve 4A desiccant at 25◦C (RH < 0.1percent). (d) Bulk density, solid density and porosity arecalculated using the standard Archimedes buoyancy methods with water as thesuspending and saturating liquid. 55499624-1 Solution data show that all Ca is replaced by Co and that no Si is lost by the solid, so that the remaining mass gain must occur by incorporation of water, either as hydroxyl groups OH or as molecular water H2O or both. The mass of water in Co-S-H (here conditioned at 11 percent RH at 25◦C) corresponds to ≈ 13mol water per mol of xonotlite in the starting material. We write the reaction stoichiometry as: 6Co2++ Ca6Si6O17(OH)2+ 13H2O → Co6Si6Ox(OH)y.zH2O + 6Ca2+(1) where x = 4+z and y = 28−2z by charge balance. Samples lose only 20 percent of the incorporated water when conditioned over a molecular sieve 4A desiccant. We regard this as a rough measure of the amount of loosely held molecular water at that RH (<0.1percent at 25◦C). Therefore most of the incorporated water is chemically combined as OH. Water vapour sorption isotherms show that Co-S-H is hygroscopic and takes up about ten times as much water as CS over the RH range 0–80 percent at 25◦C. From Table 1, we have z ≈ 3, so that Co-S-H has the empirical formula Co6Si6O7(OH)22 ·3H2O. Eqn 1 can be written succinctly as: 6Co2+ + Xon + 13H2O → Co-S-H + 6Ca2+. (2)The molecular water combines reversibly and is zeolitic and non-structural. There is evidence for the hydrous nature of Co-S-H from the IR bands in the region 3300– 3650cm−1present in Co-S-H but not in CS. The weak feature at 1630cm−1confirms thata small quantity of molecular water is present in Co-S-H at ambient relative humidity, aswe find gravimetrically. The reaction scheme of Eqn 1 implies that the double-chain silicate unit (Si6O17)–10is dismantled during the reaction. This is supported by the disappearance of the IR band at 1197cm−1, previously assigned to the crosslinks between the double chains of xonotlite. We have shown the structure is disordered (XRD, IR, XAS), hydrated (IR, mass change) and resembles phyllosilicate fragments (XRD, XAS), which we expect to be edge-terminated with OH groups for charge neutrality. We have direct knowledge of the elemental composition from the starting xonotlite, the Ca / Co exchange reaction and the fact that Si is retained, together with the mass changes on forming Co-S-H. This combined with charge balance constraints leadsto Eqn 1. To determine z we make use of the dehydration mass change (z ≈ 3), whichleads us to Eqn 2 and provides the complete empirical formula. 55499624-1 Sequestration kinetics To determine the kinetics of the sequestration reaction, we use the ICP and quantitative XRD data. As evidenced in Figure 4a, in contact with Co solution, the release of Ca into solution occurs at a constant rate. The amount of Ca released tracks closely the disappearance of crystalline xonotlite in the CS as determined by quantitative XRD (Fig 4a), and also tracks the appearance of the Co-S-H phase. This constant rate is maintained down to low Co concentrations and until the reaction ceases abruptly, either because all available Co has been sequestered or until all xonotlite has been consumed. In Fig 4a, where the initial Co concentration is 0.35m, the timescale for complete release of Ca and complete decomposition of xonotlite is about 105h. The total amount of Ca released into solution is 7.60±0.30mmol, and the amount of xonotlite initially present in the solid CS is 1.30mmol. This mol ratio (7.60 / 1.30) = 5.8±0.2 is very close to the expected stoichiometric ratio of 6. How the amount of Co in solution decreases with time is a direct measure of therate of Co sequestration. Figure 4b shows this for three different initial Coconcentrations. Remarkably, the rate is the same for all: which means that the rate of sequestration is independent of the initial Co concentration over a tenfold variation in that concentration, and remains independent of Co concentration to the lowest measured levels. The reaction therefore has zero-order kinetics with respect to Co. The Co sequestration rate is controlled by the rate of release of Ca resulting from the dissolution or decomposition of xonotlite. Several such zero-order dissolution-controlled reactions are known in rock-fluid systems. For Co sequestration we write: dnCo / dt = −mCSk, (3)or nCo(t) = nCo(0)−mCSkt, (4)where nCo is amount of Co (mol) in solution, mCS the mass of CS sequestrant used, and k the rate constant. In these tests, the best-fit rate constant k0 = 0.074±0.004mol / (kg CS h) at 25◦C. In practical terms, this corresponds to a rate of removal in a well-mixed reactor of 105 kg Co per tonne of CS per day. It is likely that the quantity mCS is a proxy for the Si content of the solid phase, which is conserved (and therefore constant) throughout the sequestration process. The amount of Co-S-H formed is in direct proportion to the amount of Si present in the weight of CS sequestrant used. 55499624-1 The time for the complete removal of Co from solution is proportional to the initial amount of Co in solution, irrespective of concentration, as is diagnostic of a zero-order reaction. These values are 103±6, 23±1, 12±2 and 0.8±0.2hours for equal volumes of solution at the three initial concentrations b(Co)00.350, 0.073, 0.037 and 0.0035m, each in contact with 1.00g CS. Generally, the time for Co removal in these tests is 0.29(b(Co)0 / mm)h. We expect that the rate constant depends on the particle size of the CS, and of course on temperature. Further confirmation of the sequestration rate constant comes from filter-column experiments, in which a Co solution percolates through a CS bed at a controlled flow rate. Combined results show that CS sequesters Co2+from aqueous solution at a constant rate, independent of the Co concentration, up to a maximum capacity of about 7.5 mol Co per kg CS, or about 440g Co per kg CS. Corresponding observations in relation to Ni sequestration kinetics can be drawn from Figures 11a,11b. Selectivity Relatively few other elements are present in the natural environment as divalent metal cations at high solution concentrations. The results reported here show that Ca2+does not interfere with Co2+sequestration. Scoping experiments show that high concentrations of Mg2+do not interfere, and neither do the monovalent ions Na+and K+. It is also believed that Ni2+behaves in similar fashion to Co2+, and that both ions were found to be sequestered simultaneously from mixed Co2+ / Ni2+solutions. Sharp Front sequestration transport model When a Co solution is brought into contact with a sedimented powder bed or a block of CS the progress of the sequestration is visible as a moving reaction front. In the experimental arrangement of Figure 5a, a Co(NO3)2 solution containing a total initial mass ms0 of Co is in contact with the upper surface of a water-saturated CS bed of volume fraction porosity fb. The reaction front is steep and its position, xf, well defined: this prompts the use of a Sharp Front [SF] model. SF models have been used to describe water transport by capillary flow through porous materials, but a new SF model is developed here to describe a transport process that incorporates the chemical sequestration. 55499624-1 The SF test directly simulates Co removal from contaminated waters by blocks of CS, as loose powder would be more difficult to deploy and retrieve during environmental remediation. The SF model is highly successful in describing the movement of the reaction front therefore it should be widely applicable in modelling reaction zones in industrial and environmental mineral systems. Given that the SF system in Figure 5b is closed, the total mass of Co is constantand so ms+mr= ms0, with msthe total mass in solution, and mrthe total mass immobilized in the reaction zone. The solution mass concentration where Ls is the length occupied by the solution in the tube of cross-section area A. Then the Co mass flux across the upper surface of the bed at x = 0 is jm = A−1·dmr / dt. We write the simple linear transport relations: where xf =γmr, with γ the length of the reaction zone per unit mass of sequestered Co, and K is a transport coefficient, dimension L2T−1(length squared / time, same as diffusivity). We assume that the solution is well mixed, and that the Co concentration at the bed surface, c′b, is the same as the solution concentration, c′s. The massconcentration of Co, c′ = cMCo = bρsoln / (1+bMCo), where c is the amount concentration(mol Co / m3), b the molality (mol Co / kgw), MCo the molar mass of Co, and ρsoln the solution density. In terms of the experimental variable xf, Eqn 5 becomes: where xf∞ = γms0 is the final (equilibrium) position of the reaction front when all Co has accumulated in the reaction zone (ms = 0). The advance of the front is eventually halted by the removal of Co from solution. Integrating Eqn 6 with initial condition xf = 0 at t = 0 gives: 55499624-1 with X = (2T)1 / 2at early times. In experimental tests, the position xfof the reaction front was measured by image analysis as described in the ‘Methods’ section above. Figure 6 shows that the SF model describes the motion of the reaction front well.There are only two disposable fit parameters, α0and α1. In fact, α1= xf∞, the final location of the reaction front, and this is obtained by direct measurement with little error at the end of the experimental run. Since Lsis known, K can be determined immediately fromthe fit parameter α0 = Ls / K, and from the data of this test is found to be (4.58±0.10)×10−10m2s−1. Figure 7 shows data from four tests in which the volume and molality of the Cosolution were varied, as was the packing density of the CS bed. The scaled data (X,T) are described well by the dimensionless form of the SF model, Eqn 8, which therefore provides the master curve for the sequestration kinetics. An important quantity derived from the fit parameter α1 is the Co:Ca stoichiometry of the reaction. Since α1 = xf∞, then γ = α1 / ms0, the ratio of the final length of the reaction zone to the initial mass of Co in solution. The packing density of the bed ρb is known, and 1 / (Aγρb) is the mass of Co sequestered per unit mass of CS. We then compare the total amount (mol) of Co sequestered with the amount of xonotlite Ca originally present in the length α1 = xf∞, the equilibrium location of the reaction front. Figure 8a shows that this mol ratio is 1.00±0.01, demonstrating that all the Caavailable in the CS in the reaction zone is replaced by Co. The Ca / Co replacement is stoichiometric and complete, confirming the same result from solutions analysis of a stirred system using powdered CS. Figure 8b shows also that the transport parameter K is sensitive to the bedporosity f. The extrapolated value of K at f = 1, ≈ 9.2×10−10m2s−1, is close to reported values of the diffusivity of Co2+ion in water at similar concentrations. This shows that Co sequestration in the porous solid is controlled by the slow Co2+ion diffusion in the bed, not by the much faster kinetics of the sequestration reaction measured in the stirred- batch experiments using CS powder. In cases such as this, the SF experiment provides a simple way to determine accurately the diffusivity of the metal ion in a porous material. With reference to Figures 7 and 8 above, corresponding observations in relationto Ni sequestration can be drawn from Figures 13 and 14. 55499624-1 Conclusions The inventors have demonstrated the possible total sequestration of Co2+from concentrated solutions by using a calcium silicate hydrate material with exceptionally high sequestration capacity. The sequestration was shown to occur spontaneously at ambient temperature, removing 0.44kg of Co per 1kg of CS in under 5 days and is a replacement reaction that does not reverse when the Co-S-H product is in contact with aqueous Ca after its formation. This represents much higher sequestration capability than currently known materials. These results demonstrate that CS is a fast, easy and economic sequestrant for removing cobalt from solution, particularly for environmental and nuclear cobalt clean-up targets. Importantly CS can be deployed in large blocks rather than powders, as the SF tests have demonstrated. The reaction product Co-S-H is X-ray amorphous and has a large surface area. The silicate double chains of CS are destroyed in the sequestration which occurs by coupled dissolution-precipitation. There is evidence that Co-S-H contains fragments of phyllosilicate structures. Amorphous Co-phyllosilicates are known to act as water-splitting catalysts, and the present reaction represents a low cost route to manufacture such compounds, with the potential to replace expensive precious-metalcatalysts (for example, for use in water hydrolysis and / or hydrogen generation).Sequestration kinetics are zero-order in Co concentration, so that the rate is maintained even at low Co concentrations. The reaction goes to completion leaving at most only ppm residual Co. A new SF experiment design and analysis give access to chemical and transport parameters where Co solution is in direct contact with the solid sequestrant. SF methodology has wide applicability in fundamental materials studies, in process and environmental engineering. The SF and filter-column experiments provide the data required for applications scale-up. It will be understood that the present embodiments are provided by way of example only, and that various modifications can be made to the present embodiments without departing from the scope of the invention. For example, whilst most of the sequestration data has been generated using cobalt, it is believed that the same principle would apply in the sequestration of nickel. 55499624-1

Claims

CLAIMS1. A method of preparing a metal silicate hydrate material, the method comprising:contacting a calcium silicate material with a solution comprising a dissolvedmetal, wherein the metal comprises, consists essentially of or consists of cobalt and / or nickel, and wherein the calcium silicate material comprises: (i) xonotlite; (ii) a carbonate material; and(iii) a fibrous material.

2. A method according to claim 1, wherein the calcium silicate material comprises:(i) the xonotlite in an amount of at least 80% by weight of the calcium silicate material; (ii) the carbonate material in an amount of about 1-10% by weight of thecalcium silicate material; and (iii) the fibrous material in an amount of about 1-10% by weight of the calciumsilicate material.

3. A method according to any preceding claim, wherein the carbonate materialcomprises or consists of calcite.

4. A method according to any preceding claim, wherein the fibrous materialcomprises or consists of a cellulosic material.

5. A method of preparing a metal silicate hydrate material, the method comprising:contacting a calcium silicate material with a solution comprising a dissolved metal, wherein the metal comprises, consists essentially of or consists of cobalt and / or nickel, and wherein the calcium silicate material is provided as a block of material having a volume of at least 1 cm3. 55499624-16. A method according to claim 5, wherein the calcium silicate material is providedas a block of material having a volume of at least 1000 cm3.

7. A method according to claim 5 or claim 6, wherein the calcium silicate materialcomprises xonotlite.

8. A method according to any of claims 5 to 7, wherein the calcium silicate materialcomprises Ca6Si6O17(OH)2.

9. A method according to any of claims 5 to 8, wherein the calcium silicate materialfurther comprises a carbonate material, optionally calcite, and wherein the calcium silicate material further comprises a fibrous material, optionally a cellulosic material.

10. A method according to any preceding claim, wherein the dissolved metalcomprises, consists essentially of, or consists of, Co(II) and / or Ni(II).11 A method according to any preceding claim, wherein the solution is an aqueoussolution.

12. A method according to any preceding claim, wherein the metal silicate hydratematerial is a metal phyllosilicate material.

13. A method according to any preceding claim, wherein the metal silicate hydratematerial has the general formula (I): M6Si6Ox(OH)y.zH2O (I)where M is Co and / or Ni;x = 4+z; and y = 28−2z.

14. A method according to claim 13, wherein the metal silicate hydrate material hasthe general formula (Ia): Co6Si6O7(OH)22.3H2O (Ia)15. A method according to claim 13, wherein the metal silicate hydrate material hasthe general formula (Ib): 55499624-1Ni6Si6O8(OH)20.4H2O (Ib)16. A method according to any preceding claim, wherein the method is carried out atambient temperature or at room temperature.

17. A method according to any preceding claim, wherein the method does not includeheating or cooling.

18. A method according to any preceding claim, wherein the method comprisescontacting the calcium silicate material with the solution comprising the dissolved metalfor at least 1 hour, optionally at least 6 hours, optionally at least 12 hours, optionally atleast 24 hours.

19. A method according to any preceding claim, wherein the method comprisessequestering at least 90%, optionally at least 95%, optionally at least 99%, of the metal from the solution.

20. A metal silicate hydrate material obtained or obtainable by a method accordingto any of claims 1 to 19. 55499624-1

Images