Laminate membranes comprising a two-dimensional layer comprising polyaromatic functionalities

Abstract

This invention relates to membranes of two dimensional material and their uses in filtration. The membranes may include polyaromatic molecules which provide a improvement in the rejection observed for small solutes. The two dimensional material may be a transition metal dichalcogenide (TMDC) or hexagonal boron nitride (hBN).

Claims

1. A laminate membrane comprising: a plurality of nanoplatelets of a two-dimensional material comprising a transition metal dichalcogenide (TMDC) and/or hexagonal boron nitride (hBN), wherein each individual nanoplatelet is impermeable to liquid and the plurality of nanoplatelets are stacked in such a way as to form capillary-like pathways between the faces and sides of the nanoplatelets; and a plurality of polyaromatic molecules covalently bonded to the two-dimensional material.

2. The membrane of claim 1, wherein the two-dimensional material is a TMDC.

3. The membrane of claim 1, wherein the plurality of nanoplatelets is a mixture of a plurality of nanoplatelets of a first two-dimensional material comprising a TMDC and a plurality of nanoplatelets of a second two-dimensional material selected from the group consisting of a TMDC, graphene, and hBN.

4. The membrane of claim 1, wherein the polyaromatic molecules are dye molecules.

5. The membrane of claim 4, wherein the dye molecules comprise sunset yellow and/or crystal violet.

6. The membrane of claim 1, wherein the plurality of nanoplatelets is obtained from the corresponding bulk layered inorganic material using a solvent exfoliation method.

7. The membrane of claim 1, wherein the laminate membrane is comprised in a composite with a porous material.

8. A method of reducing the amount of one or more solutes in a liquid to produce a product liquid depleted in said solute or solutes; the method comprising: (a) contacting a first face of a laminate membrane with the liquid comprising the one or more solutes, wherein the laminate membrane comprises a plurality of nanoplatelets of a two-dimensional material comprising a transition metal dichalcogenide (TMDC) and/or hexagonal boron nitride (hBN); and a plurality of polyaromatic molecules covalently bonded to the two-dimensional material; wherein each individual nanoplatelet is impermeable to liquid and the plurality of nanoplatelets are stacked in such a way as to form capillary-like pathways between the faces and sides of the nanoplatelets, causing the liquid to pass through the capillary-like pathways between the faces and sides of the nanoplatelets; and (b) recovering the product liquid depleted in said solute or solutes from or downstream from a second face of the membrane; (c) optionally, recovering the solute or solutes from the first face of the membrane.

9. The method of claim 8, wherein the one or more solutes comprise one or more ions and corresponding counterions in which both ions have a hydration radius that is no larger than 1 nm.

10. The method of claim 8, wherein the method is a filtration method and wherein the product liquid is recovered as a liquid from or downstream from the second face of the membrane without the liquid having undergone a phase change.

11. The method of claim 8, wherein the liquid is an aqueous liquid.

12. The method of claim 8, wherein the one or more solutes comprise one or more ions and corresponding counterions in which both ions have a hydration radius that is no larger than 0.45 nm.

13. The method of claim 12, wherein the one or more solutes includes NaCl.

14. A method of reducing the amount of one or more non-ionic solutes in a liquid to produce a product liquid depleted in said solute or solutes; the method comprising: (a) contacting a first face of a laminate membrane with the liquid comprising the one or more solutes, wherein the laminate membrane comprises a plurality of nanoplatelets of a two-dimensional material comprising a transition metal dichalcogenide (TMDC) and/or hexagonal boron nitride (hBN); and a plurality of polyaromatic molecules covalently bonded to the two-dimensional material; wherein each individual nanoplatelet is impermeable to liquid and wherein the plurality of nanoplatelets are stacked in such a way as to form capillary-like pathways between the faces and sides of the nanoplatelets, causing the liquid to pass through the capillary-like pathways between the faces and sides of the nanoplatelets; and (b) recovering the solute or solutes liquid from or downstream from a second face of the membrane; (c) optionally, recovering any remaining product from the first face of the membrane; wherein the one or more solutes are each non-ionic species having a hydration radius that is no larger than 10 nm.

15. The method of claim 14, wherein the one or more solutes comprise non-ionic species having a hydration radius that is no larger than 1 nm.

16. The method of claim 14, wherein the method is a filtration method and wherein the product liquid is recovered as a liquid from or downstream from the second face of the membrane without the liquid having undergone a phase change.

17. The method of claim 16, wherein the non-ionic solute or each non-ionic solute is an organic molecule.

18. The method of claim 14, wherein the liquid is an organic solvent or solvent mixture.

19. The method of claim 14, wherein the liquid is an aqueous liquid.

20. The method of claim 14, wherein the method is a pervaporation method and the method of recovering the product liquid comprises allowing the liquid to evaporate from the second face of the membrane to form a vapour and subsequently condensing the vapour to form the product liquid.

21. The method of claim 20, wherein the non-ionic solute is an alcohol and the liquid is either water or a second alcohol with a smaller hydration radius than the first alcohol.

22. The method of claim 14, wherein the concentration of the one or more solutes in the product liquid is reduced by 50% or more relative to the concentration in the starting liquid.

23. The method of claim 14, wherein the plurality of nanoplatelets is obtained from the corresponding bulk layered inorganic material using a solvent exfoliation method.

24. The method of claim 14, wherein the plurality of nanoplatelets is a mixture of a plurality of nanoplatelets of a first two-dimensional material selected from the group consisting of a TMDC and hBN and a plurality of nanoplatelets of a second two-dimensional material selected from the group consisting of a TMDC, graphene, and hBN.

25. The method of claim 14, wherein the two-dimensional material is a TMDC.

26. The method of claim 14, wherein the two-dimensional material is hBN.

27. The method of claim 14, wherein the laminate membrane is comprised in a composite with a porous material.

28. A method of producing a laminate membrane of claim 7; the method comprising: a) depositing a plurality of nanoplatelets of a two-dimensional material onto a porous material to form the laminate membrane supported on the porous material; and b) contacting a first side of the laminate membrane with a first solution comprising a first concentration of the polyaromatic molecules and contacting the second side of the laminate membrane with a second solution comprising a second concentration of the polyaromatic molecules, said second concentration being lower than said first concentration, to covalently bond the plurality of polyaromatic molecules to the two-dimensional material to provide a membrane of claim 7.

29. The method of claim 28, wherein the method further comprises: obtaining the plurality of nanoplatelets of the two-dimensional material from the corresponding bulk layered inorganic material using a solvent exfoliation method.

30. The membrane of claim 1, wherein the plurality of polyaromatic molecules are charged polyaromatic molecules.

31. The membrane of claim 30, wherein the polyaromatic molecules are negatively charged.

32. The membrane of claim 1, wherein the laminate membrane has a thickness between about 100 nm and about 10 m.

33. The membrane of claim 1, wherein the polyaromatic molecules are covalently bonded to the two-dimensional material via a nitrogen atom in each polyaromatic molecule.

34. The membrane of claim 7, wherein the laminate membrane is sandwiched between layers of the porous material.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) Embodiments of the invention are further described hereinafter with reference to the accompanying drawings, in which:

(2) FIG. 1 shows: a) a schematic showing the exfoliation procedure; b) optical absorbance spectroscopy of resultant MoS.sub.2 nanoflakes dispersed in NMP after centrifugation; and c) the Raman spectrum of an individual MoS.sub.2 nanoflake after deposition onto a Si/SiO.sub.2 wafer for identification.

(3) FIG. 2 shows: I) photographs showing freestanding TMDC (MoS.sub.2) and composite (graphene/MoS.sub.2) membranes after removal from the supporting membrane; and II) photographs showing MoS.sub.2 membranes supported by PVDF filters along with SEM images showing the morphology of the TMDC paper.

(4) FIG. 3 shows: a) a thickness calibration curve of exfoliated MoS.sub.2 based PVDF membrane, (b-c) cross section and (e-f) top down.

(5) FIG. 4 shows the XRD patterns for the bulk starting materials before exfoliation (solid line), as well as for the exfoliated materials after filtration and formation of the membrane (dashed line), and also the pattern when the membrane is soaked in water for an extended period of time (7 days) (dotted line).

(6) FIG. 5 shows (a) a photograph of MoS.sub.2 based PVDF membrane and sandwich membrane. (b) a photograph of homemade H-beaker for MoS.sub.2 sandwich membrane functionalised by CV (left image), SY (middle image), and NR (right image) representing to cationic, anionic, and neutral dyes, respectively, with their molecular structures.

(7) FIG. 6 shows the determination concentration of dyes functionalised on MoS.sub.2 by the measurement of the lost absorbance of dyes mixing with dispersion of MoS.sub.2 compared to absorbance of pure dyes dissolved in IPA-water (volume ratio 1:1). The dyes functionalized on MoS.sub.2 were: (a) Crystal violet, (b) Sunset yellow, and (c) Neutral red compared to the pure dyes dissolved into IPA-water. In each case the pure dye dissolved in IPA is the upper line and the dye functionalized on the MoS.sub.2 is the lower line.

(8) FIG. 7 shows (a) Raman spectra of MoS.sub.2, CV, and MoS.sub.2 functionalised by CV. (b) Raman spectra of MoS.sub.2, SY, and MoS.sub.2 functionalised by SY. (c) Raman spectra of MoS.sub.2, NR, and MoS.sub.2 functionalised by NR. In each case, the lower line is MoS.sub.2; the middle line is CV, SY or NR, respectively; and the upper line is the dye functionalized MoS.sub.2.

(9) FIG. 8 shows: (a) Full PXRD of MoS.sub.2 with PVDF (black solid line), 2=20.17, used as reference peak position. (b) PXRD of the (002) peak position of the materials.

(10) FIGS. 9A and 9B show (a) XPS scan analysis of exfoliated and bulk MoS.sub.2. High-resolution XPS spectra showing Mo 3d, 3p, and S 2p shells of (b, d, f) exfoliated MoS.sub.2 and bulk (c, e, g) bulk MoS.sub.2. Each of the fitting peaks is labelled with corresponding orbital, as well as the binding energy in the parentheses.

(11) FIGS. 10A and 10B show: (a) XPS scan analysis of exfoliated MoS.sub.2 with MoS.sub.2 functionalised by CV, SY, and NR; (b-d) High-resolution XPS spectra showing Mo 3d, 3p, and S 2p shells, respectively. Each of the fitting peaks is labelled with corresponding orbital, as well as the binding energy in the parentheses.

(12) FIG. 11 shows: a) a photograph of the MoS.sub.2 membrane supported by a polymer (PVDF) demonstrating their flexibility. b) a cross-sectional SEM image of the formed MoS2 membrane supported on the PVDF showing the stacked layered nature of the MoS2. Membranes were produced with thicknesses varying from 1-10 m. c) TEM image showing the stacked layers of MoS.sub.2 d) a schematic showing the diffusion, driven by the large concentration gradient, of the solute (NaCl ions in this case) through the MoS2 laminate membrane which has been pre-functionalised by organic dye. e) a plot comparing the change in relative resistivity of the permeate side of a 6 m MoS2 laminate membrane for each dye molecule, when the feed side contains 1M NaCl and the permeate side contains 1000 times lower concentration (1 mM). f) a comparison of the sodium ion (Na+) percentage rejection under osmotic pressure measured by ICP-OES after 180 min for each membrane (6 m thickness), as well as the water permeation rate measured with external pressure (1 bar).

(13) FIG. 12 shows a) Raman spectra from the pristine MoS.sub.2 membrane, crystal violet (CV) powder, and a CV functionalised MoS.sub.2 membrane (MoS2/CV). (b) XPS spectra of the Mo3d region for a pristine exfoliated MoS.sub.2 membrane and (c) a CV functionalised membrane.

(14) FIG. 13 shows a) a plot of the rejection properties (R %=1C.sub.MoS2/C.sub.PVDF) of the dye functionalised MoS.sub.2 membranes (6 m thick). b) a plot of the permeation rates of each dye functionalised MoS.sub.2 membrane (6 m thick) for the different cationic species studied with a 1000 concentration gradient.

(15) FIG. 14 shows a plot of the rejection properties for the CV and SY 6 m thick MoS.sub.2 membranes for synthetic sea water containing mixed ionic solutes.

(16) FIG. 15 shows the NaCl rejection of (a, c, and e) plots of potential versus time using Ag/AgCl as electrodes and (b, d, and f) plots of relative resistivity versus time for MoS.sub.2 functionalised by 0.1 mM CV, SY and NR, respectively.

(17) FIG. 16 shows a Plot of Na.sup.+ rejection properties (R %=1C.sub.MoS2/C.sub.PVDF) and water permeation rate of 3 and 6 m in the different dyes at the same concentration (0.1 mM); the solid lines and column are % Na.sup.+ rejection and water permeation rate, respectively.

(18) FIG. 17 shows a) Plot of the rejection properties (R %=1C.sub.MoS2/C.sub.PVDF) of the 1.0 mM CV functionalised membranes at different thicknesses. b) Plot of the permeation rate of each membranes functionalised by 1.0 mM CV for the different cationic species studied with a 1000 concentration gradient.

(19) FIG. 18 shows the % Na.sup.+ rejection and water permeation rate of pristine MoS.sub.2, MoS.sub.2 functionalised by 0.1 and 1.0 mM CV in the different thicknesses; the solid lines and column are % Na.sup.+ rejection and water permeation rate, respectively.

(20) FIGS. 19A and 19B show: a) the concentration of EtOH passed into the permeate side for each membrane over 1 hour; b) concentration of EtOH remaining in the feed side for each membrane over 1 hour; c) concentration of EtOH passed through into the permeate side for each membrane over 1-5 hours; d) concentration of EtOH remaining in the feed side for each membrane between 1-5 hours.

(21) FIG. 20 shows the EtOH/water separation results using MoS.sub.2 membranes functionalised with crystal violet dye: a) Shows the concentration of ethanol passed through the functionalised MoS.sub.2 membranes (permeate side), up to 5 hours; b) Shows the concentration of ethanol remaining in the feed side, for the functionalised MoS.sub.2 membranes, up to 5 hours.

(22) FIGS. 21A and B show the separation of IPA/water mixtures through 1 m, 3, 5 MoS.sub.2 membranes and blank PVDF membrane: a) concentration of IPA passed through into the permeate side for each membrane over 1 hour; b) concentration of IPA remaining in the feed side for each membrane over 1 hour; c) concentration of IPA passed through the permeate side for each membrane between 1-5 hours; d) concentration of IPA remaining in the feed side for each membrane over between 1-5 hours.

(23) FIG. 22 shows the IPA/water separation results using MoS.sub.2 membranes functionalised by crystal violet dye, with MoS.sub.2 thicknesses of 1 m, 3 m and 5 m compared against a blank PVDF membrane. a) Shows the IPA concentrations present in the permeate side after 5 hours of stirring, for each membrane; b) Shows the IPA concentrations present in the feed side after 5 hours of stirring, for each membrane.

(24) FIG. 23 shows the pyridoxine/water separation results obtained after 6 hours; a) Shows the pyridoxine concentration in the permeate side after 6 hours for all membrane experiments; b) Shows the pyridoxine concentration in the feed side after 6 hours for all membrane experiments

DETAILED DESCRIPTION

(25) Throughout this specification, the term two dimensional material refers to nanoplatelets of TMDCs, hBN and mixtures thereof of which the laminate membrane is comprised. Thus, the term two dimensional material encompasses mixtures of WS.sub.2 and MoS.sub.2.

(26) Nanoplatelets may be single or few layered particles of the respective inorganic layered material.

(27) The term two-dimensional may mean a compound in a form which is so thin that it exhibits different properties than the same compound when in bulk. Not all of the properties of the compound will differ between a few-layered particle and a bulk compound but one or more properties are likely to be different. Typically, two-dimensional inorganic compounds are in a form which is single- or few layers thick, i.e. up to 10 molecular layers thick. A two-dimensional crystal of a layered material (e.g. an inorganic compound or graphene) is a single or few layered particle of that material. The terms two-dimensional and single or few layered are used interchangeably throughout this specification. Two-dimensional materials are not truly two dimensional, but they exist in the form of particles which have a thickness that is significantly smaller than their other dimensions. The term two-dimensional has become customary in the art.

(28) The term few-layered particle may mean a particle which is so thin that it exhibits different properties than the same compound when in bulk. Not all of the properties of the compound will differ between a few-layered particle and a bulk compound but one or more properties are likely to be different. A more convenient definition would be that the term few layered refers to a crystal that is from 2 to 9 molecular layers thick (e.g. 2 to 5 layers thick). A molecular layer is the minimum thickness chemically possible for that compound. In the case of boron-nitride one molecular layer is a single atom thick. In the case of the transition metal dichalcogenides (e.g. MoS.sub.2 and WS.sub.2), a molecular layer is three atoms thick (one transition metal atom and two chalcogen atoms). Thus, few-layer particles crystals are generally less than 50 nm thick, depending on the compound and are preferably less than 20 nm thick, e.g. less than 10 or 5 nm thick.

(29) The term multi-layered particle refers to a particle which exhibits similar properties to the same compound when in bulk. A more convenient definition would be that the term multi-layered particle refers to a particle that is 10 or more molecular layers thick.

(30) The inorganic compounds referred to in this specification are inorganic layered compounds. Thus, the term inorganic compound refers to any compound made up of two or more elements which forms layered structures in which the bonding between atoms within the same layer is stronger than the bonding between atoms in different layers. Many examples of inorganic layered compounds have covalent bonds between the atoms within the layers but van der Waals bonding between the layers. For the absence of doubt, the term inorganic layered compound is not intended to encompass graphene or other carbon based two-dimensional materials (e.g. graphene oxide, reduced graphene oxide, partially oxidised graphene).

(31) Many inorganic compounds exist in a number of allotropic forms, some of which are layered and some of which are not. For example boron nitride can exist in a layered graphite-like structure (hexagonal boron nitride or hBN) or as a diamond-like structure in which the boron and nitrogen atoms are tetrahedrally orientated. For the absence of doubt, it is hexagonal boron nitride that is referred to throughout this specification.

(32) TMDCs are structured such that each layer of the compound consists of a three atomic planes: a layer of transition metal atoms (for example Mo, Ta, W . . . ) sandwiched between two layers of chalcogen atoms (for example S, Se or Te). Thus in one embodiment, the TMDC is a compound of one or more of Mo, Ta and W with one or more of S, Se and Te. There is strong covalent bonding between the atoms within each layer of the transition metal chalcogenide and predominantly weak Van der Waals bonding between adjacent layers. Exemplary TMDCs include NbSe.sub.2, WS.sub.2, MoS.sub.2, TaS.sub.2, PtTe.sub.2, VTe.sub.2.

(33) The solutes to be removed from solution in the methods of the present invention may be defined in terms of their hydrated radius. Below are the hydrated radii of some exemplary ions and molecules.

(34) TABLE-US-00001 TABLE 1 Hydrated Hydrated Ion/molecule radius () Ion/molecule radius () K.sup.+ 3.31 Li.sup.+ 3.82 Cl.sup. 3.32 Rb.sup.+ 3.29 Na.sup.+ 3.58 Cs.sup.+ 3.29 CH.sub.3COO.sup. 3.75 NH.sub.4.sup.+ 3.31 SO.sub.4.sup.2 3.79 Be.sup.2+ 4.59 AsO.sub.4.sup.3 3.85 Ca.sup.2+ 4.12 CO.sub.3.sup.2 3.94 Zn.sup.2+ 4.30 Cu.sup.2+ 4.19 Ag.sup.+ 3.41 Mg.sup.2+ 4.28 Cd.sup.2+ 4.26 propanol 4.48 Al.sup.3+ 4.80 glycerol 4.65 Pb.sup.2+ 4.01 [Fe(CN).sub.6].sup.3 4.75 NO.sub.3.sup. 3.40 sucrose 5.01 OH 3.00 (PTS).sup.4 5.04 H.sub.3O.sup.+ 2.80 [Ru(bipy).sub.3].sup.2+ 5.90 Br 3.30 Tl.sup.+ 3.30 I 3.31 [UO.sub.2].sup.2 4.7

(35) The hydrated radii of many species are available in the literature. However, for some species the hydrated radii may not be available. The radii of many species are described in terms of their Stokes radius and typically this information will be available where the hydrated radius is not. For example, of the above species, there exist no literature values for the hydrated radius of propanol, sucrose, glycerol and PTS.sup.4. The hydrated radii of those species which are provided in the table above have been estimated using their Stokes/crystal radii. To this end, the hydrated radii for a selection of species in which this value was known can be plotted as a function of the Stokes radii for those species and this yields a simple linear dependence. Hydrated radii for propanol, sucrose, glycerol and PTS.sup.4 were then estimated using the linear dependence and the known Stokes radii of those species.

(36) There are a number of methods described in the literature for the calculation of hydration radii. Examples are provided in Determination of the effective hydrodynamic radii of small molecules by viscometry; Schultz and Soloman; The Journal of General Physiology; 44; 1189-1199 (1963); and Phenomenological Theory of Ion Solvation; E. R. Nightingale. J. Phys. Chem. 63, 1381 (1959).

(37) There are a variety of methods available for the exfoliation of bulk samples of TMDC and hBN to form nanoplatelets (Wang, Q. H.; Kalantar-Zadeh, K.; Kis, A.; Coleman, J. N.; Strano, M. S., Electronics and Optoelectronics of Two-Dimensional Transition Metal Dichalcogenides. Nat. Nanotechnol. 2012, 7, 699-712; Huang, X.; Zeng, Z.; Zhang, H., Metal Dichalcogenide Nanosheets: Preparation, Properties and Applications. Chem. Soc. Rev. 2013, 42, 1934-1946; Nicolosi, V.; Chhowalla, M.; Kanatzidis, M. G.; Strano, M. S.; Coleman, J. N., Liquid Exfoliation of Layered Materials. Science 2013, 340; Koski, K. J.; Cui, Y., The New Skinny in Two-Dimensional Nanomaterials. ACS Nano 2013, 7, 3739-3743). The methods can be broadly categorised into three classes: mechanical exfoliation, solvent exfoliation and chemical exfoliation.

(38) Mechanical Exfoliation

(39) Atomically thin flakes of TMDCs can be peeled from their bulk crystals by micromechanical cleavage using adhesive tape and then applied to target substrates, using the same techniques that were developed for graphene. Mechanical exfoliation typically produces single-crystal flakes of high purity and cleanliness that are ideal for fundamental characterization and for fabrication of individual devices. However, this method is not considered to be suitable for large scale production does not provide control over flake thickness and size.

(40) Solution Exfoliation

(41) TMDCs can also be exfoliated by ultrasonication or grinding in appropriate liquids for extended periods of time (2-12 hours), including organic solvents, aqueous surfactant solutions, or solutions of polymers in solvents. Typically, ultrasonication results in the mechanical exfoliation of layered crystals to give flakes that are a few hundred nanometres in size and vary in thickness down to monolayer. The exfoliated flakes are stabilized against re-aggregation through interactions with the solvent molecules. Thus the solvent is typically carefully chosen to produce high quality dispersed flakes of sufficient stability. This method has the highest degree of scalability combined with low costs of manufacture.

(42) Solution exfoliation techniques are the preferred methods for producing nanoplatelets for use in forming the membranes of this invention and membranes for use in the methods of this invention.

(43) Chemical Intercalation

(44) It is also possible to exfoliate many TMDCs into solution by intercalation of ionic species. The typical procedure involves submerging bulk TMDC powder in a solution of a lithium-containing compound (n-butyllithium) for extended periods of time (>1 day) to allow lithium ions to intercalate. This is then followed by exposing the intercalated material to water, usually with added brief sonication. The water reacts vigorously with the lithium between the layers to evolve H.sub.2 gas, which rapidly separates the layers. It is also possible to achieve this by electrochemical Li intercalation at room temperature over a relatively short (6 hours) time scale. Such chemical exfoliation methods produce gram quantities of submicron-sized monolayers, but the resulting exfoliated material differs structurally and electronically from the parent material. In particular, for MoS.sub.2 the process changes the electronic structure of the exfoliated nanosheets from semiconducting to metallic, and the Mo atom coordination is changed from trigonal prismatic (2HMoS.sub.2) to octahedral (1T-MoS.sub.2).

(45) Thus, throughout this specification, the term solvent exfoliation is intended to mean any exfoliation technique in which the layers of the multi-layered particles are cleaved by being subjected to energy in a solvent without prior activation of the multi-layered particles. The term specifically excludes techniques in which chemical species are intercalated between the layers of the multi-layered particles prior to the exfoliation step. The term chemical exfoliation is intended to mean any exfoliation technique in which the layers of the multi-layered particles are cleaved using a method that involves the intercalation of chemical species between the layers prior to the exfoliation step.

(46) Throughout the description and claims of this specification, the words comprise and contain and variations of them mean including but not limited to, and they are not intended to (and do not) exclude other moieties, additives, components, integers or steps. Throughout the description and claims of this specification, the singular encompasses the plural unless the context otherwise requires. In particular, where the indefinite article is used, the specification is to be understood as contemplating plurality as well as singularity, unless the context requires otherwise.

(47) Features, integers, characteristics, compounds, chemical moieties or groups described in conjunction with a particular aspect, embodiment or example of the invention are to be understood to be applicable to any other aspect, embodiment or example described herein unless incompatible therewith. All of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and/or all of the steps of any method or process so disclosed, may be combined in any combination, except combinations where at least some of such features and/or steps are mutually exclusive. The invention is not restricted to the details of any foregoing embodiments. The invention extends to any novel one, or any novel combination, of the features disclosed in this specification (including any accompanying claims, abstract and drawings), or to any novel one, or any novel combination, of the steps of any method or process so disclosed.

(48) The reader's attention is directed to all papers and documents which are filed concurrently with or previous to this specification in connection with this application and which are open to public inspection with this specification, and the contents of all such papers and documents are incorporated herein by reference.

EXAMPLES

Example 1Formation and Characterisation of Membranes of Inorganic 2-D Crystals

(49) Solvent stabilised dispersions of inorganic 2D crystals were produced by ultrasonication of commercially available bulk TMDCs (MoS.sub.2, MoSe.sub.2, WS.sub.2, WSe.sub.2, TiS.sub.2), as well as hexagonal boron nitride (hBN), in various solvents. Typically, MoS.sub.2 is dispersed in either N-Methyl-2-pyrrolidone (NMP) or a mixture of isopropanol/water (IPA/H.sub.2O) (7:3) at a concentration of 10 mg/ml and then exfoliated by bath ultrasonication for 12 hours (37 kHz, 40% amplitude, 25 C.). For hBN pure IPA was used for the exfoliation. After this sonication, the dispersions were centrifuged at 6000 rpm (3139 g) for 30 min to remove any unexfoliated material, the supernatant was then decanted and fresh solvent added before repeating the centrifugation to ensure a narrow distribution of flake dimensions and thicknesses. Similar dispersions were created using solution exfoliated graphene as a comparison and for the creation of composites. Once a stable and homogenous dispersion was created, the concentration of dispersed nanoflakes was determined by optical spectroscopy. FIG. 1 shows schematically the procedure of the exfoliation method used, as well as the characterisation of the subsequently produced MoS.sub.2 nanoflakes. These results show that after exfoliation the resultant flakes are few layered (1-3 layers in thickness).

(50) TMDC based filtration Membranes were produced by either using the IPA/H.sub.2O dispersions as prepared or by first diluting the NMP dispersed flakes in IPA by a factor of 20 and then filtering the dispersion through various supporting filtration membranes, typically a polyvinylidene fluoride (PVDF) membrane with a pore size of 0.1 m, using a syringe pump. Typically, a volume of 20-60 ml of the diluted dispersions are used to create the membrane with a thickness ranging from 1 to 10 m. During filtration the flakes in the dispersion are tightly stacked together to form a coherent paper which is supported by the PVDF filter. Composite membranes consisting of various mixtures of exfoliated 2D crystals (e.g MoS.sub.2/graphene, MoS.sub.2/WS.sub.2, graphene/hBN etc.) were created by simply mixing the dispersions of each individual material in differing ratios and briefly sonicating to ensure a homogenous distribution before filtration as for the pure membranes.

(51) Using the methodology described in the previous two paragraphs, the following membranes have been produced: MoS.sub.2, WS.sub.2, MoSe.sub.2, WSe.sub.2, TiS.sub.2, Bi.sub.2Te.sub.3, Graphene, hBN, MoS.sub.2/Graphene (1:1), MoS.sub.2/Graphene (1:3), MoS.sub.2/Graphene (3:1), MoS.sub.2/WS.sub.2 (1:1) and Graphene/hBN (1:1)

(52) FIG. 2 shows examples of the freestanding membranes created by peeling the TMDC paper off the supporting filter as well as the membranes created by leaving on the supporting PVDF filter. The SEM analysis shows that the individual nanoflakes stack closely together horizontally to form a continuous film. Individual nanoflakes are several hundreds of nanometers in lateral dimensions up to 1 m. The film in FIG. 2 was intentionally damaged to allow for imaging of the internal structure.

(53) FIG. 3(a) shows thickness calibration curve of MoS.sub.2 at three different mass (Table 1). The thicknesses were measured by SEM as shown in FIG. 3(a). The membrane was cut by the surgical blade in the middle membrane and attached on SEM cross-section stub. This calibration equation was used to make different thickness membranes that were 1, 3, 5, and 12 m. The MoS.sub.2 cross-section were shown in the straight line in FIG. 3(b-c). The lateral MoS.sub.2 were determined by top-down SEM images ca. 200-300 nm as shown in FIG. 3(e-f).

(54) TABLE-US-00002 TABLE 1 The thickness of exfoliated MoS.sub.2 based PVDF membrane. Mass exfoliated MoS.sub.2 based on PVDF/mg Thickness/m 0 0 0.43 2.50 0.04 0.92 4.95 0.09 1.42 7.90 0.20

(55) X-Ray diffraction (XRD) can also be used to determine the interlayer spacing and subsequently the size of ions which are predicted to pass through. This allows us to accurately predict the size exclusion properties of these membranes and of the composites. FIG. 4 shows the XRD patterns for the bulk starting materials before exfoliation (solid), as well as for the exfoliated materials after filtration and formation of the membrane (dashed line), and also the pattern when the membrane is soaked in water for an extended period of time (7 days) (dotted line). Typically used GO membranes suffer from significant swelling due to their hydrophilicity which makes the interlayer spacing increase significantly when wet, while this gives them a high water flux this makes them only useful for removal of large species (>0.9 nm). As seen in FIG. 4 the TMDC membranes do not suffer from any swelling when exposed to water for long periods of time, meaning that their dry interlayer spacing is approximately equivalent to that when in use for water filtration. This is one of the reasons why they are able to much more effectively sieve out smaller ionic species.

Example 2Formation and Characterisation of Membranes of Inorganic 2-D Crystals Functionalised by Dye Molecules

(56) A dispersion of MoS.sub.2 in isopropanol and water (1:1 ratio) was filtered onto PVDF membrane using a programmable syringe pump as shown in FIG. 5(a). Significantly, the MoS.sub.2 based PVDF membrane can be folded without a reduction in performance of the MoS.sub.2 membranes. This suggests that the flakes are not damaged by folding. This membrane was supported by a double side of polyethylene terephthalate (PET) with a controlled area of 26.910.sup.6 m.sup.2 attached by adhesive glue to form the MoS.sub.2 sandwich membrane.

(57) The MoS.sub.2 sandwich membrane was inserted into a H-beaker using two O-rings and supported with a spherical joint clip. Crystal violet, Sunset yellow, and Neutral red were used as cationic, anionic, and neutral dyes, respectively, to functionalise the MoS.sub.2 membrane. For each, 50.0 mL of 0.1 mM aqueous solution dyes were prepared as a feed side, and 50.0 mL of DI water as permeate side.

(58) The concentration of dyes in both feed and permeate sides were determined by UV-vis spectrophotometry every several days for 21 days. The molar extinction coefficients of dyes are 87,000 M.sup.1cm.sup.1 at 590 nm for CV, 22,200 M.sup.1cm.sup.1 at 482 nm for SY, and 13,900 M.sup.1cm.sup.1 at 482 nm for NR (the base form). All membranes in the H-beaker were cleaned for a week using ultra-pure water with hydrostatic pressure to remove residual dyes inside the membranes.

(59) The concentration of the dyes functionalised on MoS.sub.2 was determined by measuring lost absorbance of dyes mixing with dispersion of MoS.sub.2 compared to absorbance of pure dyes dissolved in IPA-water (volume ratio 1:1) as same as the dispersion of MoS.sub.2 as shown in FIG. 6(a-c).

(60) The mass of dyes functionalised on MoS.sub.2 were shown in Table 2. The saturated amount of dyes that attaches to the surface of the MoS.sub.2 dispersion was determined to be 30, 160, and 24 g per 1 mg of MoS.sub.2 respectively for CV, SY, and NR.

(61) TABLE-US-00003 TABLE 2 The saturated mass of Crystal violet, Sunset yellow, and Neutral red functionalised on MoS.sub.2 at different concentrations Mass of dyes (g) per 1 mg of MoS.sub.2 Prepared dyes Crystal Sunset Neutral concentration/M violet (pH 7) yellow (pH 7) red (pH 10) 1 11.0 1.7 3 2.8 24.1 5 24.1 8.0 8.6 10 29.4 22.2 24.7 20 29.4 47.4 24.0 50 152.8 100 159.2 Saturated dyes on ~30 ~160 ~25 MoS.sub.2
Powder X-Ray Diffraction Spectroscopy

(62) Powder X-ray diffraction (PXRD) patterns of the MoS.sub.2, bulk form and after exfoliation to form membranes as well as after dyes functionalisation, were obtained using a PANalytical X'Pert X-ray diffractometer. Using a Cu-K radiation source operating at 40 kV and 30 mA the patterns were records whilst spinning in the range 2=545, with a step size of 0.017 and a scan step time of 66 s. The position of the (002) peak was used to calculate d-spacing of the materials according to Bragg's law:

(63) $d=\frac{}{2\sin }$
where is the wavelength of the radiation source (0.15418 nm) and is half the position of the (002) peak. All X-ray patterns were corrected using PVDF at the 2of 20.17 as a reference peak.

(64) The X-ray patterns of MoS.sub.2 with PVDF membrane as supporter are shown in FIG. 8(a). FIG. 8(b) shows (002) peak position of bulk (black line) and after exfoliation (green line). It is clearly seen that the (002) peak after exfoliation is significantly broader with slightly bigger d-spacing compared to bulk form. This increased layer spacing can be explained the MoS.sub.2 flakes were restacking misorientations during filtration, as well as the untrasonication process which reduces the size from several microns to a few hundreds of nm. These would lead to increase of average layer spacing of MoS.sub.2. Moreover, PXRD pattern at (002) of MoS.sub.2 with dyes kept in water media to conserve the nano-channel in MoS.sub.2 after functionalisation were shown in FIG. 8b. MoS.sub.2 functionalised by dyes shows peak position as similar as pristine MoS.sub.2 indicating that MoS.sub.2 sandwich membrane did not undergo any significant swelling in water over 6 months

(65) Raman and X-Ray Photoelectron Spectroscopy of Exfoliated MoS.sub.2 Functionalised by Crystal Violet, Sunset Yellow, and Neutral Red.

(66) Raman spectra of MoS.sub.2 functionalised by Crystal violet, Sunset yellow, and Neutral red is shown in FIG. 7(a-c, upper line) compare to Raman spectra of pre-functionalised MoS.sub.2 and three type of dyes. The Raman shift of Sunset yellow and Neutral red molecules functionalised on MoS.sub.2 were similar to pure those dyes as shown in FIG. 7(b and e). Interestingly, in FIG. 7(a) in MoS.sub.2 functionalised by Crystal violet (upper line) shows a new splitting peak around 1900 cm.sup.1 compared to pure crystal violet (middle line). This shows that the vibration of Crystal violet had changed as somehow of the interaction between MoS.sub.2 and Crystal violet (e.g. the formation of a new covalent bond between nitrogen atom in crystal violet and Mo atoms that allow the new electron resonances in Crystal violet molecules).

(67) To understand this chemical functionalisation mechanism, Raman and X-ray photoelectron spectroscopies were utilised to analyse the pristine and dye treated MoS.sub.2 membranes. FIG. 8a compares the Raman spectra of the pristine MoS.sub.2 membrane, with the characteristic E.sub.2g and A.sub.1g MoS.sub.2 peaks labelled, the raw dye (CV) powder, and the functionalised MoS.sub.2 membrane (MoS.sub.2/CV). Clearly we can see that the characteristic peaks for the CV are present on the functionalised membrane. This Raman analysis was performed after the functionalised membrane has undergone multiple cleaning treatments with water and has been submerged in multiple aqueous solutions for prolonged periods of time (>6 months), indicating that the dye functionalisation is robust and not simply an adsorption reaction. This is supported by the XPS analysis shown in FIG. 8b-c which shows a dramatic difference in the Mo.sub.3d peak after functionalisation. The change in the spectrum corresponds to a significant increase in the amount of octahedral (Mo.sup.4+) co-ordinated molybdenum atoms, and indicates that the CV molecules have become covalently attached to the molybdenum within the membrane. This is also seen by shifts due to charge transfer in the sulphur peaks and changes in the Mo.sub.3p region, as well as presence of bonded nitrogen atoms (N.sub.1s) from the structure of CV itself. This shows a new peak binding energy between Mo 3p and N1s around 395.8 eV compared Mo 3p from pre-functionalised MoS.sub.2. Furthermore, MoS.sub.2/SY samples show peak shifts which indicate a change of vibration of SY on MoS.sub.2.

(68) XPS is commonly used to determine the composition and stoichiometry of MoS.sub.2. XPS spectra of MoS.sub.2, before and after exfoliation in IPA-water, are shown in FIG. 9(a). The binding energies are calibrated using adventitious carbon (C 1s) at 284.8 eV and the atomic percentage of the peaks were determined by optimised peak fit using a nonlinear Shirley-type background (70% Gaussian and 30% Lorentzian line shapes) with the Kratos library.

Example 3Discussion of Formation and Properties of Membranes of Inorganic 2-D Crystals Functionalised by Dye Molecules

(69) MoS.sub.2 membranes were first produced using syringe-pump assisted filtration, where different volumes of exfoliated MoS.sub.2 flakes dispersions as described above were filtered through polyvinylidene difluoride (PVDF) supporting membranes (100 nm pores) to produce membranes of the desired thickness. The exfoliated flake dimensions within the dispersion (0.01 mg/ml in isopropanol/water) were characterised and found to have thicknesses in the several (1-5) layer range and lateral dimensions of 200-300 nm. The PVDF supporting membranes alone were also characterised and found to exhibit no ionic selectivity or separation ability and thus do not affect the MoS.sub.2 membrane performance. FIG. 11a shows a photograph of one of the MoS.sub.2 membranes supported on a PVDF filter, along with a cross-section SEM image (FIG. 11b) which shows the laminar structure of the MoS.sub.2 sheets. Membranes of various thicknesses between 1-10 m were produced as specified. FIG. 11c shows a TEM image showing the stacked layers of MoS.sub.2. X-ray diffraction (XRD) was used to investigate and compare the (002) peak in the exfoliated MoS2 membranes, which corresponded to a d-spacing of 6.12 .

(70) Determination of Water Permeation Rate

(71) Water permeation rates were determined using two different techniques. Firstly, the osmotic pressure technique depending on concentration of solute was calculated using van't Hoff law:
=i[C.sub.sucrose]RT

(72) where is osmotic pressure, i is 1 for the van't Hoff factor, [C.sub.sucrose] is 1 M sucrose solution, R is gas constant, and T is the temperature in Kelvin, corresponding to osmotic pressure of 25 bar at room temperature.

(73) In this method, a sucrose solution (1 M) was placed on one side of the MoS.sub.2 membrane and distilled water on the other to allow for a direct comparison of relative performance. From the van't Hoff equation this corresponds to an osmotic pressure of 25 bar, and by measuring the change in volume of the water on the feed side the permeation rate can be approximated. This was found to be 4010.sup.3 L m.sup.2 h.sup.1 bar.sup.1 for a 3 m thick membrane and 11.610.sup.3 L m.sup.2 h.sup.1 bar.sup.1 for a 6 m thick membrane (both functionalised by dye), which compares very favourably to the GO membranes which exhibited a permeation of 810.sup.3 L m.sup.2 h.sup.1 bar.sup.1 for a 1 m thick membrane, demonstrating the increased water flux through the MoS.sub.2 membranes. This matches previous reports which have demonstrated water permeation rates 3-5 times higher for MoS.sub.2 membranes compared to GO. These values also compare favourably to those recently shown for MoS.sub.2 membranes produced through chemical exfoliation measured by osmotic pressure (1 M NaCl) of 2.210.sup.3 L.Math.m.sup.2.Math.h.sup.1.Math.bar.sup.1.

(74) Secondly, external pressure was used in a dead-end filtration setup to more accurately represent industrial applicability. The MoS.sub.2 sandwich membrane was cut and inserted in syringe holder with two O-ring at the top and bottom of membrane as well as adhesive glue to hold membrane and protect water leaking. The water permeation rate was calculated by weighing the water pushed through the controlled MoS.sub.2 membrane area (26.910.sup.6 m.sup.2) by the external pressure.

(75) The water permeation rate under external pressure was found to be 269.5 (23.7) L m.sup.2 h.sup.1 bar.sup.1 for a 6 m thick dye treated MoS.sub.2 membrane. This compares very favourably to similar external pressure results for pristine MoS.sub.2 laminates, which showed a permeation rate of 245 L m.sup.2 h.sup.1 bar.sup.1 for a much thinner 1.7 m membrane. This increased water permeation rate is likely related to the interaction of the water molecules with the molybdenum atoms that are present due to sulphur vacancies and defects produced through ultrasonication and at flake edges, which has been shown theoretically.

(76) Determination of Ionic Permeation Rate

(77) Ionic permeation was carried out in a homemade H-beaker as shown in FIG. 5b. A MoS.sub.2 sandwich membrane was inserted into the H-beaker using two O-rings as supporter. The starting concentration in feed and permeate side are 1 M and 1 mM, respectively, at the equal volume of 50.0 mL. Both solutions were stirred constantly during the experiment to minimise the concentration gradient. The ionic permeations were determined by in situ resistivity and potentiometric measurement as a function of time and ex situ increasing ions concentration by ICP-OES after three hours. Resistivity and potentiometric measurements were performed simultaneously by conductivity meter to measure the decreasing resistivity in permeate side and a potentiostat to measure the potential difference between two Ag/AgCl reference electrodes in both feed and permeate sides without applied current.

(78) Permeation of various solutes including; organic dyes and metal salts (KCl, NaCl, Na.sub.2SO.sub.4, CaCl.sub.2, and MgCl.sub.2) were measured using a homemade H-beaker setup (similar to that described above for the dye functionalization of the membranes). The membranes were first sandwiched between two polyethylene terephthalate (PET) sheets with pre-formed holes to ensure that the exposed membrane area was controlled and reproducible for each experiment. This sandwich was then sealed between the two containers with equal volumes of water, to minimise any hydrostatic pressure, and a feed side with a high concentration of the selected solute (1 M) and a permeate side with a thousand times lower concentration (1 mM) to ensure high diffusive pressure, and this is shown schematically in FIG. 11d. Both sides of the cell were constantly stirred with a magnetic bar to minimise concentration gradients. The rejection properties were measured using several techniques; optical absorbance spectroscopy was used for the dyes, in situ electrochemical techniques (both conductivity and potentiometric measurements) were used to determine the change in concentration of the ionic species, combined with ex situ inductively coupled plasma optical emission spectrometry (ICP-OES) to independently measure the concentration of the solutes. The different dye molecules investigated where a cationic dye; tris(4-(dimethylamino)phenyl)methylium chloride, commonly known as crystal violet (CV), an anionic dye; disodium 6-hydroxy-5-[(4-sulfophenyl)azo]-2-naphthalenesulfonate (sunset yellow, SY), and a neutral dye; 3-amino-7-dimethylamino-2-methylphenazine hydrochloride (neutral red, NR).

(79) The as-prepared MoS.sub.2 membranes exhibit excellent rejection properties for each of the large (1 nm) dye molecules, with very low permeation rates (CV: 5.3 mol m.sup.2 h.sup.1, SY: 0.9 mol m.sup.2 h.sup.1, NR: 12.5 mol m.sup.2 h.sup.1) for a 6 m thick membrane. However, the as-prepared membranes exhibited poor rejection of the group I and II metal cations listed above. Surprisingly, the filtration properties of the membranes were found to be transformed by their dye functionalization. On exposure to the cationic and anionic dye molecules for extended periods of time (21 days); they exhibited excellent rejection of simple ionic solutes. To visualise the dynamic permeation of the solute ions we can simply measure the change in resistivity of the permeate side of the membrane as a function of time, and as the ionic concentration increases the resistivity will decrease. Some discrepancy arises from the non-selectivity of the conductivity measurements, as exposure to atmospheric gases (e.g. CO.sub.2) also leads to a decrease in the resistivity over this time, thus by starting with 1 mM of chosen solute this affect is minimised.

(80) The plot in FIG. 11e compares the rejection performance of the dye functionalised MoS.sub.2 membranes as well as the bare supporting PVDF filter, both the CV and SY functionalised MoS.sub.2 membranes demonstrate that after 3 hours the relative resistivity has dropped by <10% indicating excellent rejection properties. The NR functionalised membrane however displays negligible rejection properties, similar to the bare PVDF membrane. This indicates that it is the interaction between the charged dye molecules and the slightly negative MoS.sub.2 flakes themselves that leads to the ionic rejection behaviour.

(81) To complement the resistivity measurements, in situ potentiometric measurements were also performed, measuring the potential difference between two silver/silver chloride (Ag/AgCl) reference electrodes, placed in each of the feed and permeate compartments, with no applied current (supporting information). By measuring the change in measured potential with time we can accurately detect any ionic concentration changes as any tendency to equalize the ionic concentrations will lead to a rapid decrease in the measured potential difference between the two reference electrodes, and these results were found to agree with the resistivity measurements shown. To validate these electrochemical results the increased concentration of sodium cations (Na.sup.+) present in the permeate side after three hours was determined by inductively coupled plasma-optical emission spectrometry (ICP-OES), these values are shown in FIG. 11f along with the water flux determined with external pressure.

(82) Ionic rejection can be defined as 1C.sub.MoS.sub.2/C.sub.PVDF where C.sub.MoS.sub.2 is the increased concentration of solute ions in the permeate side when the dye functionalised MoS.sub.2 membrane is in place, while C.sub.PVDF is the increased concentration when the blank membrane is present after three hours. The results in FIG. 11f give a Na.sup.+ rejection rate of 96.36 (0.27)% and 97.73 (0.63)% for the 6 m CV and SY thick MoS.sub.2 membranes, respectively despite the 1000 concentration gradient. The pristine MoS.sub.2, MoS.sub.2/NR, and bare PVDF however, showed much lower rejection properties. These results are in excellent agreement with the electrochemical results and indicate that the cationic and anionic dye functionalised membranes are able to efficiently reject NaCl (Na.sup.+). The water flux, calculated using external pressure, are also summarised for each corresponding membrane in FIG. 11f, where notably the CV functionalisation actually leads to an increase in the water flux compared to the pristine MoS.sub.2 membrane, indicating increased water transport properties whilst maintaining excellent ionic rejection.

(83) Due to the demand for desalination technology each of the commonly found ionic solutes in sea water were chosen to be analysed. The rejection rate, calculated after three hours with a 1000 concentration gradient as discussed previously, for each of KCl, NaCl, Na.sub.2SO.sub.4, CaCl.sub.2, and MgCl.sub.2 as a function of hydrated cationic radius are shown in FIG. 13a for a 6 m thick MoS.sub.2 membrane. The rejection rate for the SY functionalised membrane shows a consistent 98% rejection for each of the analysed solutes, increasing to 100% for the Mg.sup.2+, while the CV varies slightly more with rejection rates between 92-100%. It is also possible to calculate the permeation rate of these ionic solutes, by taking the concentration determined from ICP-OES as a function of time, and these ionic permeation rates are shown for the CV and SY functionalised membranes in FIG. 13b. These rates compare very favourably with the results for a similar thickness GO membrane (5 m) and feed concentration (1 M), namely a permeation rate of 2 mol.Math.m.sup.2.Math.h.sup.1 for these size ions. The ionic permeation achieved for the 6 m thick dye functionalised MoS.sub.2 membranes are approximately an order of magnitude lower at 0.15 mol.Math.m.sup.2.Math.h.sup.1 for the K.sup.+, Na.sup.+, and Ca.sup.2+ and a further order of magnitude lower for the Mg.sup.2+ at 0.025 mol.Math.m.sup.2.Math.h.sup.1. This combined with the significantly increased water permeation rate indicates that these membranes are promising for desalination and nanofiltration applications.

(84) Without wishing to be bound by theory, the predominant transport mechanism in this work is attributed to an electrostatic attraction between the cations in solution and the negatively charged MoS.sub.2 channels, with atomic radii only affecting the transport above the critical channel width. However, in our results we observe that although the XRD results indicate the channel width to be approximately 6.4 we observe high rejection of species with radii >4.2 , illustrating how the dye functionalisation has led to a tunability of these membranes.

(85) The change in permeation rates for the ions with a smaller hydration radius than 4.2 indicates that the nanocapillaries formed between the MoS.sub.2 sheets must be approximately 8.4 in width. This is in agreement the XRD results which show an interlayer spacing for the MoS.sub.2 membrane of 14 , leaving a free space after subtracting the thickness of a single MoS.sub.2 layer of 10 . This is slightly lower than that achieved for hydrated GO membranes of 13 , and is the source of the desired ionic selective filtration observed.

(86) To better replicate real world rejection properties of these membranes a mixed ionic solution of synthetic sea water (SSW) was tested, shown schematically in the FIG. 14. The concentration of the ionic solutes was certified as matching the standard composition of sea water, with the highest concentration components being NaCl (0.420 M), MgCl.sub.2.6H.sub.2O (0.0556 M), Na.sub.2SO.sub.4 (0.0288 M), CaCl.sub.2.2H.sub.2O (0.0105 M), and KCl (0.00926 M). FIG. 14a plots the rejection rate for each of the cationic components of the SSW after 3 hours as a function of the hydrated radii. Unlike the single ionic component solutions the larger doubly charged ions (Ca.sup.2+ & Mg.sup.2+) appear to have a lower rejection rate than the smaller ions, this can be attributed to their divalent nature which leads to a stronger electrostatic interaction with the abundant Cl.sup. anions present in the mixed solute. The interaction between multiple ionic components has not been analysed previously for similar laminar membranes and remains an area that must be better understood to realise industrial applications.

(87) The concentration of ions passed through membrane analysed by inductively couple plasma optical emission spectroscopy (ICP-OES) is shown in Table 3 and 4.

(88) TABLE-US-00004 TABLE 3 Ions rejection in percent calculated by ICP-OES. Membranes Dyes Ions rejection (%) feed Thickness K.sup.+ Na.sup.+ Na.sup.+ Ca.sup.2+ Mg.sup.2+ (mM) (m) (KCl) (NaCl) (Na.sub.2SO.sub.4) (CaCl.sub.22H.sub.2O) (MgCl.sub.26H.sub.2O) 0.1 mM CV 6 90.3 1.5 95.6 0.3 92.8 0.4 93.7 0.5 100 3 96.7 0.3 92.5 0.8 1 97.0 0.3 1 mM CV 5 98.4 0.1 98.0 0.1 87.1 0.2 99.2 0.2 100 3 94.9 0.2 97.0 0.2 84.6 0.1 95.3 0.6 100 1 97.6 0.2 97.13 0.3 88.2 0 97.8 0.2 100 0.1 mM SY 6 95.1 3.4 97.2 0.7 97.8 0.4 96.4 0.8 100 3 14.7 4.1 0.1 mM NR 6 36.0 0.7 3 39.3 0.7 Pre-dyes 5 27.0 3 18.8 1 11.3

(89) TABLE-US-00005 TABLE 4 Synthetic Sea Water (SSW) Rejection and each ion passed through membrane. Membranes Thickness Ions rejection (%) Dyes feed (m) K.sup.+ Na.sup.+ Ca.sup.2+ Mg.sup.2+ 0.1 mM 6 91.7 0.1 95.1 0.4 84.8 0 88.1 0.1 CV 3 88.7 0 90.5 0.6 79.8 0.1 84.8 0.3 1 mM 5 96.4 0 97.6 0.2 90.9 0 93.8 0 CV 3 93.8 0.1 94.6 0 90.9 0 90.2 0 0.1 mM 6 94.9 0.7 95.7 1.9 77.3 0.8 87.9 2.4 SY

Example 4Discussion of Formation and Properties of Membranes of Inorganic 2-D Crystals of the Invention in Separating Organic Molecules from Water

(90) The filtration and separation properties of transition metal dichalcogenides (TMDs) membranes, specifically molybdenum disulphide (MoS.sub.2) membranes, were explored to assess their capability for separation of organic-water mixtures. More specifically, the separation of alcohols (e.g. ethanol and isopropanol) from water was studied, as well as the separation of pyridoxine from water.

(91) The separation of ethanol-water mixtures and isopropanol-water mixtures were studied using both pristine and functionalised MoS.sub.2 membranes. These samples were analysed using gas chromatograph with a flame ionisation detector (GC-FID). Additionally, the separation of pyridoxine-water mixtures was also studied, and these samples were analysed using UV-Vis absorption spectroscopy.

(92) Experimental Methods

(93) Ultrasonic liquid phase exfoliation of molybdenum (IV) disulphide (MoS.sub.2) was the method used in this experiment to exfoliate bulk MoS.sub.2 to create submicrometer-thick MoS.sub.2 membranes. A known amount of the exfoliated MoS.sub.2 material was passed through a syringe pump at a rate of 10 ml h.sup.1 to create various thicknesses of MoS.sub.2 on PVDF membrane. The following membranes were used for all the experiments discussed in this report: blank PVDF membrane, 1 m-thick MoS.sub.2, 3 m-thick MoS.sub.2 and 5 m-thick MoS.sub.2 membranes.

(94) The membranes were each sandwiched between two plastic sheets, held together by Araldite epoxy resin. Once dried, the sandwiched membrane is placed between two U-shaped beakers, and this was the main experimental setup for the experiments. One of the beakers was filled with a known amount of an organic/water mixture, referred to as the feed side, and the remaining beaker was filled with the same amount of pure water, referred to as the permeate side.

(95) Ethanol-Water Separation

(96) Ethanol is soluble in water and rapid, effective separation of various EtOH/water mixtures is highly desired. The feed side contained a mixture of 1:1 EtOH/water, and the permeate side contained pure de-ionised water only. Samples were taken at various time intervals, which were then analysed using a gas chromatograph with a flame ionisation detector (GC-FID).

(97) The separation of EtOH/water was investigated using both pristine MoS.sub.2 membranes, and functionalised MoS.sub.2-crystal violet membranes with a thickness of 1 m, 3 m, 5 m on PVDF, compared against a blank PVDF membrane.

(98) The permeation data for the pristine MoS.sub.2 membranes are shown in FIG. 19. After 5 hours there is a noticeable increase in EtOH concentration in the permeate side of the blank membrane experiment, suggesting that a small amount of ethanol has passed through the membrane. To a lesser extent, there is also an increase in the ethanol concentration in the permeate side for the 1 m, 3 m and 5 m membranes. More ethanol appears to have passed through the 1 m-thick MoS.sub.2 membrane than the thicker MoS.sub.2 membranes, suggesting that the thicker MoS.sub.2 membranes separate ethanol and water more effectively.

(99) The ethanol concentration in the feed side, up to 5 hours, does not show a trend for any of the membranes. This implies there was very little change in ethanol concentration, which suggests the MoS.sub.2 membranes effectively separate water and ethanol up to 5 hours.

(100) In the permeate side, the blank PVDF membrane functionalised with crystal violet dye shows an increase in ethanol concentration (see FIG. 20a) after 5 hours. The 1 m, 3 m and 5 m MoS.sub.2 membranes also show an increase in ethanol concentration, however this is less than the blank PVDF membrane. The feed side shows the largest decrease in ethanol concentration for the blank membrane. For the 1 m, 3 m and 5 m MoS.sub.2 membranes, there is little change in concentration suggesting that the ethanol/water separation is efficient up to 5 hours.

(101) There is little difference when comparing the functionalised and pristine MoS.sub.2 membranes for EtOH/water separation. It is possible that the ethanol present in the feed side washed the crystal violet dye out of the membranes as crystal violet dissolves in ethanol.

(102) Isopropanol-Water Separation

(103) The experimental setup was the same as described for the ethanol-water separation experiments above, however isopropanol was substituted for ethanol. Hence the feed side contained a 1:1 ratio of IPA/water, and the permeate side initially contained pure de-ionised water only. Again, IPA/water separation experiments were conducted using both pristine MoS.sub.2 membranes and MoS.sub.2 membranes functionalised by crystal violet, at thicknesses of 1 m, 3 m and 5 m MoS.sub.3. These were compared against a blank PVDF membrane. All samples were analysed using the GC-FID.

(104) As shown in FIG. 21, the concentration of IPA in the permeate side increases for all the membranes with time. As expected, the concentration increase is highest for the blank membrane and lowest for the 5 m-thick MoS.sub.2 membrane, suggesting that more IPA has passed through the blank membrane than 5 m membrane. Hence the 5 m MoS.sub.2 membrane is more effective at preventing IPA permeation. In the feed side, there is a slight decrease in IPA concentration for the blank PVDF membrane experiment. There is very little change in concentration in the feed side for the MoS.sub.2 membrane experiments, suggesting that overall the membranes effectively separate IPA and water up to 5 hours.

(105) FIG. 22 shows IPA/water separation results using MoS.sub.2 membranes functionalised by crystal violet dye, with MoS.sub.2 thicknesses of 1 m, 3 m and 5 m compared against a blank PVDF membrane.

(106) Overall, the permeate side shows an increase in IPA concentration for the membrane experiments, suggesting that a small amount of IPA has passed through the MoS.sub.2 membranes. The blank PVDF experiment displayed the largest increase in IPA concentration as expected. In the feed side, there is little change in concentration for all the membrane experiments, implying that overall the membranes effectively separate IPA and water. As with the ethanol/water experiments, there appears to be little difference between the functionalised and pristine MoS.sub.2 membranes. Again, the IPA in the feed side mixture was suspected to dissolve, and hence remove, the crystal violet dye in the functionalised membranes.

(107) Pyridoxine-Water Separation

(108) These samples were analysed using UV-Vis spectroscopy as pyridoxine absorbs at 327 nm (in the UV region). The membranes used in this experiment were as follows: blank PVDF membrane and pristine MoS.sub.2 membranes of varying thicknesses (1 m, 3 m and 5 m). The permeate side consisted of pure de-ionised water only, and the feed side contained a 50 M pyridoxine solution dissolved in de-ionised water. FIG. 23 shows the pyridoxine/water separation results obtained after 6 hours.

(109) For all the membrane experiments, there was little or no change in the concentration of pyridoxine in both the permeate and feed side, which suggests that little or no pyridoxine is passing through the membranes. One possible reason for pyridoxine not passing through even the blank PVDF membrane is that it may functionalise the membranes by sticking to the PVDF and not passing through the capillaries. Further work is required to confirm this theory