WO2012101457A1

WO2012101457A1 - Exfoliation of layered materials

Info

Publication number: WO2012101457A1
Application number: PCT/GB2012/050189
Authority: WO
Inventors: Jonathan Nesbitt COLEMAN; Valeria NICOLOSI
Original assignee: Oxford University Innovation Ltd; College of the Holy and Undivided Trinity of Queen Elizabeth near Dublin
Current assignee: Oxford University Innovation Ltd; College of the Holy and Undivided Trinity of Queen Elizabeth near Dublin
Priority date: 2011-01-28
Filing date: 2012-01-30
Publication date: 2012-08-02
Anticipated expiration: 2013-07-28
Also published as: GB201101482D0

Abstract

The present invention provides a process for producing a two-dimensional material, the process comprising contacting a three-dimensional inorganic layered material with a solvent under conditions such that the three-dimensional material is exfoliated to form a two-dimensional material, wherein the three-dimensional material is a metal-containing material or a clay mineral, and wherein the solvent has a dispersive Hansen solubility parameter (δD) of from about 10 to about 22 MPa^1/2. Two-dimensional materials obtainable by said process are also provided.

Description

EXFOLIATION OF LAYERED MATERIALS

Field of the Invention This invention relates to two-dimensional materials and to processes for their production. In particular, the invention relates to a process for producing two-dimensional inorganic materials. The two-dimensional materials may be useful in various devices, including electronic, semiconductor and insulating devices. Background to the Invention

A variety of two-dimensional materials are known. Examples of such materials include graphene and boron nitride. Graphene in particular has generated huge interest in recent years due to its unique physical properties. However, a wide range of other two- dimensional materials exist, many of which have huge potential both for basic research and applications. These include metal chalcogenides (e.g. transition metal dichalcogenides; TMDs), metal oxides (e.g. transition metal oxides; TMOs), clay minerals and other layered structures such as Sb₂Te₃, TiNCI, Bi₂Te₃ and Bi₂Se₃. The latter materials are of particular interest due to their status as topological insulators and their thermoelectric properties. While interest is shifting to inorganic layered materials, the main hurdle is the lack of a simple method to prepare mono- or few-layer flakes of such materials in large quantities.

The most common inorganic layered materials are the transition metal dichalcogenides. These are potentially important because they occur in more than 40 different types with a wide range of electronic properties, varying from metallic to semiconducting. Layered transition metal dichalcogenides consist of hexagonal sheets of metal atoms sandwiched between two sheets of chalcogen atoms. While the bonding within these tri-layer sheets is covalent, adjacent sheets within a crystal are weakly bound by van der Waals interactions. Depending on the co-ordination and oxidation state of the metal atoms, transition metal dichalcogenides can be metallic, semi-metallic or semiconducting. For example, WS₂ is a semiconductor while PtTe₂, NbSe₂, TaS₂ and NiTe₂ are metals. In addition, superconductivity and charge density wave effects have been observed in some transition metal dichalcogenides. This versatility makes them potentially useful in many areas of electronics.

However, like graphene, inorganic layered materials must be exfoliated to fulfil their full potential. For example, films of exfoliated Bi₂Te₃ should display enhanced thermoelectric efficiency by suppression of thermal conductivity. Exfoliation of two-dimensional topological insulators such as Bi₂Te₃ and Bi₂Se₃ would reduce residual bulk conductance, highlighting surface effects. In addition, changes in electronic properties are expected as the number of layers is reduced, e.g. the indirect bandgap of bulk MoS₂ becomes direct in few layer flakes. Although exfoliation may be achievable mechanically on a small scale, liquid phase exfoliation methods are required for any realistic applications. Transition metal dichalcogenides can be exfoliated by ion intercalation. However, this method is time-consuming and extremely sensitive to the environment, and so is unsuitable for most applications. Furthermore, removal of the ions has been found to result in reaggregation of layers.

It has recently been shown that graphite can be exfoliated in various solvents to give monolayer sheets of graphene (see Hernandez et al, Nature Nanotechnology, 2008, 3, 563). In this method, the surface energy of the solvent used to exfoliate the graphite is matched to the surface energy of graphene. The surface energy of graphene is about 70 mJ/m², which is in the upper range of surface energies for most solvents. Accordingly, this method was considered unsuitable for exfoliating transition metal dichalcogenides and other inorganic, metal-containing layered materials, as the surface energies of such materials have been determined to be considerably higher than that of graphene. For example, Weiss et al (Physical review B, 1976, 14, 5392) teach that the surface energies of transition metal dichalcogenides such as MoS₂ and WS₂ are greater than 200 mJ/m². Since no solvent has a surface energy this high, solvent-based exfoliation techniques were considered to be unsuitable for exfoliating transition metal dichalcogenides and various other inorganic layered materials.

Summary of the Invention

It has surprisingly been found that certain three-dimensional inorganic layered materials, more particularly metal-containing materials and clay minerals, can be exfoliated in a variety of solvents to produce two-dimensional materials. In particular, it has been found that such materials can be exfoliated using solvents having a dispersive Hansen solubility parameter (5_D) of from about 10 to about 22 MPa^1/2.

Accordingly, the present invention provides a process for producing a two-dimensional material, the process comprising contacting a three-dimensional inorganic layered material with a solvent under conditions such that the three-dimensional material is exfoliated to form a two-dimensional material, wherein the three-dimensional material is a metal-containing material or a clay mineral, and wherein the solvent has a dispersive Hansen solubility parameter (5_D) of from about 10 to about 22 MPa^1/2. Two-dimensional materials obtainable by said process are also provided.

A process of the present invention may be used to prepare a wide range of two- dimensional materials. The materials may be useful in various products and devices, including electronic, semiconductor and insulating devices. Advantageously, the disclosed process may be non-destructive, insensitive to air and water, and may yield materials having minimal defects. Moreover, the process may allow two-dimensional materials to be produced in large quantities and the materials obtained may be of higher quality compared with materials obtained by conventional methods.

Brief Description of the Drawings

Figure 1 shows transmission electron microscopy (TEM) images of flakes of MoS₂ and WS₂. In all cases the scale bar is 100 nm.

Figure 2 shows a statistical analysis of flake size for MoS₂ and WS₂ flakes dispersed in N-methylpyrrolidinone (NMP). Here, L, w and N denote the flake length, width and number of monolayers per flake.

Figure 3 shows bright field scanning transmission electron microscopy (STEM) images (left) and high angle annular dark field STEM images (right) of flakes deposited from various dispersions produced in accordance with the present invention. Deschption of Various Embodiments

The present invention provides a process for producing a two-dimensional material by exfoliation of a three-dimensional inorganic layered material. Two-dimensional materials generally comprise one or more atomically thin crystalline repeating units referred to as monolayers (or nanosheets). The thickness of a monolayer will depend on the composition of the material. For example, a monolayer of graphene is generally one atom thick, whereas a monolayer of a transition metal dichalcogenide may comprise a hexagonal sheet of transition metal atoms sandwiched between two sheets of chalcogen atoms. A two-dimensional material of the present invention may comprise a single monolayer or a plurality of monolayers stacked together. Where more than one monolayer is present, the monolayers may be weakly bonded by van der Waals interactions. The two-dimensional material may comprise from 1 to about 20 monolayers, e.g. from 1 to about 10 monolayers, e.g. from 1 to about 5 monolayers. The two-dimensional material may have a thickness ranging from about 0.3 nm to about 10 nm, e.g. from about 0.3 nm to about 5 nm, e.g. from about 0.3 nm to about 2.5 nm. The two-dimensional material may have a lateral size ranging from about 50 nm to about 50 pirn, e.g. from about 50 nm to about 5 μιη. The dimensions of the two-dimensional material and the number of monolayers present may be determined using techniques known in the art, such as those described elsewhere herein.

The three-dimensional inorganic layered material, and hence the two-dimensional material produced therefrom, is a metal-containing material or a clay mineral. In an embodiment, the three-dimensional material is a metal-containing material. For instance, the three-dimensional material may be selected from the group consisting of metal chalcogenides, metal oxides, Sb₂Te₃, TiNCI, Bi₂Te₃ and Bi₂Se₃. The three- dimensional material may contain one or more transition metals. In an embodiment, the three-dimensional material comprises a transition metal dichalcogenide. In an embodiment, the three-dimensional material comprises a transition metal dichalcogenide of the formula MX₂, wherein M is Ti, Zr, Hf, V, Nb, Ta, Cr, Mo, W, Tc, Re, Ni, Pd or Pt; and each X is S, Se or Te. In an embodiment, the three-dimensional material comprises a transition metal oxide. In an embodiment, the three-dimensional material comprises a transition metal oxide of the formula MX₂, wherein M is Ti, Zr, Hf, V, Nb, Ta, Cr, Mo, W, Tc, Re, Ni, Pd or Pt; and each X is O. In an embodiment, the three-dimensional material comprises MoS₂, MoSe₂, MoTe₂, TaSe₂, NbSe₂, WS₂, Bi₂Te₃ or NiTe₂. In a particular embodiment, the three-dimensional material comprises MoS₂ or WS₂.

In an embodiment, the three-dimensional material is a clay mineral. The clay mineral may be selected from kaolin-serpentines (e.g. kaolinite, dickite, nacrite, halloysite, chrysotile, antigorite or lizardite), pyrophyllite-talcs (e.g. pyrophyllite or talc), smectites (e.g. beidellite, montmorillonite, nontronite, saponite or hectorite), vermiculites (e.g. vermiculite), illites (e.g. illite or glauconite), micas (e.g. muscovite, celadonite, phlogopite or taenolite), brittle micas (e.g. margarite), palygorskite-sepiolites (e.g. palygorskite or sepiolite) and chlorites (e.g. clinochlore).

According to a process of the present invention, the three-dimensional material is contacted with a solvent having a dispersive Hansen solubility parameter (5_D) of from about 10 to about 22 MPa^1/2. Hansen solubility parameters are used to describe solute- solvent interactions and relate to the dispersion, polar and hydrogen-bonding contributions to the cohesive energy density. The dispersive, polar and hydrogen- bonding Hansen solubility parameters are denoted by the symbols δ₀, δ_Ρ and δ_Η respectively, and represent the square root of the contribution to the cohesive energy density. The Hansen solubility parameters of various solvents are given in Hansen Solubility Parameters - A User's Handbook (CRC Press 2007), and can be derived using HSPiP software. In an embodiment, the solvent has a dispersive Hansen solubility parameter (5_D) of from about 12 to about 21 MPa^1/2, e.g. from about 15 to about 20 MPa^1/2.

In an embodiment, the solvent has a polar Hansen solubility parameter (δ_Ρ) of from about 1 to about 20 MPa^1/2, e.g. from about 2 to about 15 MPa^1/2, e.g. from about 4 to about 15 M Pa^1/2, e.g. from about 5 to about 10 M Pa^1/2.

In an embodiment, the solvent has a hydrogen-bonding Hansen solubility parameter (δ_Η) of from about 1 to about 25 MPa^1/2, e.g. from about 2 to about 20 MPa^1/2, e.g. from about 4 to about 14 MPa^1/2, e.g. from about 5 to about 10 MPa^1/2. As will be appreciated by those skilled in the art, the hydrogen-bonding Hansen solubility parameter is not restricted solely to hydrogen-bonding interactions, but relates to all interactions other than dispersive and polar interactions.

Particularly desirable results may be obtained using solvents having a well defined set of Hansen solubility parameters. In an embodiment, the solvent has at least one of, and preferably all of, a dispersive Hansen solubility parameter (5_D) of from about 12 to about 21 MPa^1/2, a polar Hansen solubility parameter (δ_Ρ) of from about 1 to about 20 MPa^1/2, and a hydrogen bonding Hansen solubility parameter (δ_Η) of from about 1 to about 25 MPa^1/2. In another embodiment, the solvent has at least one of, and preferably all of, a dispersive Hansen solubility parameter (5_D) of from about 15 to about 20 MPa^1/2, a polar Hansen solubility parameter (δ_Ρ) of from about 5 to about 15 MPa^1/2, and a hydrogen bonding Hansen solubility parameter (δ_Η) of from about 4 to about 20 MPa^1/2.

The solvent may also be defined by reference to the Hildebrand parameter (δ_τ). The square of the Hildebrand solubility parameter represents the sum of the squares of the dispersive, polar and hydrogen-bonding Hansen solubility parameters, i.e. δ_τ ² = (¾ +δ_ρ + δ_Η ² . In an embodiment, the solvent has a Hildebrand solubility parameter of from about 10 to about 40 MPa^1/2, e.g. from about 15 to about 35 MPa^1/2, e.g. from about 20 to about 30 MPa^1/2.

In an embodiment, the solvent has a surface energy of from about 60 to about 80 mJ/m², e.g. from about 65 to about 75 mJ/m², e.g. about 70 mJ/m².

In an embodiment, the solvent has a surface tension of from about 30 to about 50 mJ/m², e.g. from about 35 to about 45 mJ/m², e.g. about 40 mJ/m².

The solvent may be in the form of a single solvent or a mixture of two or more solvents. In an embodiment, the solvent comprises an organic solvent. In a particular embodiment, the solvent comprises a solvent selected from acetone, benzaldehyde, benzyl benzoate, benzyl ether, benzonitrile, bromobenzene, chlorobenzene, cyclohexyl- pyrrolidinone, chloroform, cyclohexane, cyclohexanone, dimethylacetamide, dimethylformamide, dimethylimidazolidinone, dimethylsulphoxide, N-dodecylpyrrolidone, formamide, isopropanol, methanol, N-methylformamide, N-methylpyrrolidinone, N- octylpyrrolidone, quinoline and N-vinylpyrrolidinone. Preferred solvents include cyclohexylpyrrolidinone, N-methylpyrrolidone, N-vinylpyrrolidone, isopropanol and dimethylformamide.

The three-dimensional material is contacted with the solvent under conditions such that the three-dimensional material is exfoliated to form the two-dimensional material. Typically, the three-dimensional material will be in the form of a powder which is mixed with the solvent. In an embodiment, the three-dimensional material is in the form of a powder having an average particle size of from 1 μιη to about 100 μπτι, e.g. from 1 μιη to 50 μιη. Exfoliation of the three-dimensional material may be facilitated by mixing the three-dimensional material with the solvent, and then applying energy, e.g. ultrasound energy, and/or a force, e.g. a centrifugal force, to said mixture. Energy may be applied to the mixture using any suitable means known in the art, e.g. by applying ultrasound or other sonic energy to the mixture. Preferably, a sonication bath is used. A centrifugal force may be applied using a centrifuge.

The two-dimensional material will generally be obtained in the form of a dispersion in the solvent, the dispersion comprising flakes of the two-dimensional material. In an embodiment, the dispersion comprises flakes of the two-dimensional material having from 1 to about 20 monolayers, e.g. from 1 to about 10 monolayers, e.g. from 1 to about 5 monolayers. Flakes of the two-dimensional material may be produced having a thickness ranging from about 0.3 nm to about 10 nm, e.g. from about 0.3 nm to about 5 nm, e.g. from about 0.3 nm to about 2.5 nm. The flakes may have a lateral size ranging from about 50 nm to about 50 μπτι, e.g. from about 50 nm to about 5 μιη. The dimensions of the flakes and number of monolayers may be determined using techniques known in the art, such as those described elsewhere herein. A process of the invention may be used to prepare dispersions comprising a plurality of two-dimensional materials. For instance, a dispersion may be produced comprising the two-dimensional material and one or more nanomaterials, e.g. another two-dimensional material or a one-dimensional material (e.g. a nanotube). In an embodiment, the one or more nanomaterials are selected from graphene, boron nitride, carbon nanotubes, inorganic nanotubes (e.g. metal oxide nanotubes) and metallic nanowires (e.g. silver nanowires). Alternatively, a dispersion may be produced comprising a plurality of two- dimensional inorganic materials selected from metal-containing materials and clay minerals. The versatility of the present process means it is possible to create such "hybrid" dispersions simply by adding one or more nanomaterials to the dispersion, or by mixing a dispersion containing the two-dimensional material with a dispersion containing one or more nanomaterials. Alternatively, the three-dimensional layered material may be contacted with the solvent in the presence of another three-dimensional layered material which is exfoliated by the solvent to yield a two-dimensional material. Preferred solvents for the production of such dispersions include cyclohexylpyrrolidinone, N- dodecylpyrrolidone, N-methylpyrrolidone, N-octylpyrrolidone and N-vinylpyrrolidinone.

The two-dimensional materials produced by a process of the present invention may be of improved quality and/or exhibit superior properties compared with two-dimensional materials obtained by conventional methods. The materials may display desirable properties such as e.g. high dielectric constants, high thermoelectric efficiencies and desirable mechanical properties. Moreover, reaggregation of the two-dimensional material may be minimised. It has been found that transition metal dichalcogenides produced by a process of the invention tend not to deviate significantly from their hexagonal structures, in contrast to transition metal dichalcogenides produced by e.g. lithium intercalation.

The process may further comprise forming a film comprising the two-dimensional material. In an embodiment, the process comprises forming a film comprising the two- dimensional material, wherein the film has a thickness of from about 5 nm to about 1 mm. Particularly where the film is for use in nanoelectronics, the film preferably has a thickness of from 1 to about 50 monolayers, e.g. from 1 to about 20 monolayers, e.g. from 1 to about 10 monolayers, e.g. from 1 to about 5 monolayers. Preparation of thin films, such as films of transition metal dichalcogenides, has previously been time consuming and complex and thus has limited the use of such materials in areas such as photovoltaics and solid-state batteries. However, the exfoliation process of the invention may allow films comprising such materials to be prepared. Depending on the three- dimensional layered material employed, the films may display metallic, semiconducting or insulating properties.

The film may be produced by any suitable method. The film may be formed on a substrate, which may be subsequently removed if desired. For example, the film may be produced by gravure coating, inkjet printing, Langmuir-Blodgett coating, Mayer rod coating, screen printing, spray casting, spin coating, or vacuum filtration. Particularly useful methods include vacuum filtration and spraying. For particularly thin films, such as those for use in nanoelectronics, the use of methods such as spray casting and Langmuir Blodgett deposition may be preferred.

A film may be formed comprising the two-dimensional material and one or more nanomaterials, e.g. another two-dimensional material or a one-dimensional material (e.g. a nanotube). In one embodiment, the one or more nanomaterials are selected from graphene, boron nitride, carbon nanotubes, inorganic nanotubes (e.g. metal oxide nanotubes) and metallic nanowires (e.g. silver nanowires). Such films may be produced from hybrid dispersions of the type described elsewhere herein. By using a mixture of the two-dimensional material and one or more nanomaterials, films having particularly desirable properties may be obtained. For example, one of the most promising low temperature thermoelectric materials is Bi₂Te₃. Preparation of Bi₂Te₃S and single-walled nanotube hybrid films may result in films with dramatically increased electrical conductivity and thus extremely large zT values (i.e. high thermodynamic efficiency). As another example, it has been found that the addition of graphene may result in films having particularly desirable mechanical properties and/or electrical conductivity.

The materials may also be used as reinforcing fillers in composite materials. Preparation of composite materials with exfoliated transition metal dichalcogenides has previously been considered impossible without ion intercalation. However, the present invention may facilitate formation of such composite material. By way of illustration, the two- dimensional material may be used to form a composite material comprising a polymer. The polymer may be a thermoplastic polymer such as e.g. a polystyrene, a polycarbonate or a polymethlymethacrylate. In an embodiment, the polymer is a polyurethane. Composites of the invention may exhibit desirable mechanical properties, such as high tensile strength.

Two-dimensional materials obtainable by a process of the present invention may be used in a wide range of devices or components including batteries, capacitors (e.g. nano- capacitors or super capacitors), dielectrics, electrodes (e.g. transparent electrodes), light emitting diodes (e.g. nano-light emitting diodes), solar cells (e.g. nano-solar cells), thermoelectric devices and transistors (e.g. nano-transistors). For instance, in nanoelectronics, semiconducting materials will be preferred in nano-transistors and nano-solar cells; conducting materials may be preferred in nano-electrodes; and insulating materials may be preferred in nano-capacitors or nano-transistors.

The following non-limiting Examples illustrate the present invention.

Materials and Methods

In the following Examples, commercially available MoS₂ and WS₂ powders were obtained from Aldrich. In addition, NbSe₂ and NiTe₂ were obtained from American Elements, MoTe₂ and MoSe₂ from Cerac Incorporated, TaSe₂ from Chemsavers and Bi₂Te₃ from Absco. All materials were used as supplied. All solvents were obtained from Aldrich and used as supplied.

Three different types of low power sonication baths were used: Branson 1510E-MT, Branson 2510E-MT and Bandelin Sonorex RK 1028H. In some cases, a horn probe sonic tip (VibraCell CVX; 750 W) was used. Centrifugation was carried out using a Hettich Mikro 22R centrifuge. UV-vis-IR absorption spectroscopy was performed with a Cary 6000i using 1 cm or 1 mm quartz or optical glass cuvettes. Transmission electron microscopy (TEM) was carried out with a Jeol 2100, operated at 200 kV. High resolution TEM images were taken with the Oxford-JEOL JEM2200MCO FEGTEM/STEM, fitted with two CEOS Cs aberration correctors, operated at 200 kV. Spatially-resolved electron energy-loss spectroscopy (EELS) was carried out in scanning TEM (STEM) configuration using a FEI Titan(TM) 80-300 S TEM operated at 80 kV. Selected area electron diffraction patterns were acquired using a JEOL 2010 operated at 200 kV. In some cases, electron microscopy was performed using an FEI Titan, operated at 300 kV in both bright field and high angle annular dark field modes. Scanning electron microscopy (SEM) was performed with a Zeiss Ultra Plus. Helium ion microscopy was performed with a Zeiss Orion Plus. The working distance was 5 mm and θ δ μηη aperture was used. The beam current was 0.5 pA with a tilt of 5 degrees. Atomic force microscopy (AFM) measurements were made with a Digital Instruments Nanoscope IIIA using silicon tips with a typical resonance frequency of 320 kHz. Tensile testing was carried out using a Zwick proline tensile tester with a strain rate of 0.5 mm/min (free standing films) or 50 mm/min (composites). Scanning Raman measurements were carried out using an NT- MDT Nova system.

Example 1 : preparation of dispersions

Samples were prepared by mixing MoS₂ and WS₂ powders in various solvents. The samples were then sonicated (10 ml cylindrical vial, starting concentration 1 mg/ml) in a low power sonic bath (Branson 2510E-MT) for 1 hour. The resulting dispersions were then centrifuged at 500 rpm for 90 min. After centrifugation, the dispersions remained opaque for some solvents. The supernatant (top two thirds of the centrifuged dispersion) was collected by pipette. The mass remaining in the supernatant was estimated by measuring the UV-vis-IR absorption spectrum.

Following this procedure, dispersions of MoS₂ and WS₂ were prepared using the following solvents: cyclohexylpyrrolidinone (CHP), N-dodecylpyrrolidone (N12P), benzyl benzoate, isopropanol, N-octylpyrrolidone (N8P), N-vinylpyrrolidinone (NVP), benzyl ether, dimethyl-imidazolidinone (DMEU), cyclohexanone, chlorobenzene, dimethylsulphoxide (DMSO), benzonitrile, chlorobenzene, chloroform, bromobenzene, N- methylpyrrolidinone (NMP), N-methylformamide (NMF), dimethylformamide (DMF), dimethylacetamide (DMA), benzaldehyde, quinoline, cyclohexane, methanol, acetone and formamide. Solvent parameter analysis

Effective solvents for exfoliation of the inorganic layered materials were found to have a dispersive Hansen solubility parameter (5_D) in the range of from about 10 to about 22 MPa^1/2. Particularly effective solvents were found to have a well defined set of Hansen solubility parameters, falling within the ranges given in Table 1 :

Table 1

Moreover, particularly effective solvents were found to have surface tensions of approximately 40 mJ/m², surface energies of approximately 70 mJ/m². Hildebrand solubility parameters of the solutes were found to be between 20 and 30 MPa^1/2.

Optimisation of dispersions

The concentration of material achieved after centrifugation could be improved by optimising the method of sonication, the sonication time and the centrifugation rate. To do this, small quantities (10 ml) of dispersion in 14 ml vials or round bottom flasks were sonicated. Many dispersions were sonicated using sonic bath or point probe (sonic tip) for different times. After sonication, the dispersions were centrifuged at different rates, 500 and 1500 rpm (45 or 90 min). The supernatant was decanted and the UV-vis-IR spectra measured. The dispersion quality was rated initially by the absorbance per unit length, A/I. Promising dispersions were studied by TEM to check for evidence of exfoliation. Optimum dispersion procedures for MoS₂ and WS₂ were found to be: sonication (10 ml in a 14 ml vial) with point probe (sonic tip), at nominal power output of 285 W (38%x750 W) for 60 min, followed by centrifugation at 1500 rpm for 45 min. The concentrations (after centrifugation) of MoS₂ and WS₂ dispersions in NMP were determined to be 0.3 mg/ml and 0.15 mg/ml for respectively.

Analysis of dispersions

Dispersions of MoS₂ and WS₂ in various solvents, but particularly NMP and I PA, were analysed by transmission electron microscopy (TEM). Typically, a drop of a given dispersion was placed on a holey carbon grid (400 mesh) and dried in ambient conditions. Low resolution TEM was performed in a Jeol 2100, operated at 200 kV while HRTEM images were taken with the Oxford-JEOL JEM2200MCO FEGTEM/STEM, fitted with two CEOS Cs aberration correctors, operated at 200 kV.

In general, large quantities of two-dimensional objects were observed. A selection of the monolayers and multilayers observed is shown in Figure 1. The length, L, and width, w, of a large number of flakes were estimated to generate flake size statistics. By careful analysis of the flake edge, it was also possible to estimate the number (N) of layers per flake for MoS₂ and WS₂. These results are shown in Figure 2. It can be seen that reasonable numbers of monolayers are observed in each case. Analysis of the intensity profiles show that the imaged MoS₂ and WS₂ flakes are likely to be single layers. Further evidence for exfoliation was provided by electron diffraction and electron energy-loss spectroscopy (EELS).

Example 2: deposition of individual flakes. Deposition of MoS₂ flakes

Samples for deposition were prepared by ice-cooled sonication of 7.5 mg/ml MoS₂ in NMP under a point probe for 1 hour at 38% amplitude. The dispersion was left to settle overnight and centrifuged at 1500 rpm for 45 minutes. The supernatant was decanted by pipette and retained for use. Silicon substrates with thermally grown 300 nm oxide were used for deposition and were cleaned by rinsing with isopropanol (I PA) and blow drying. The MoS₂/NMP dispersion (0.36 mg/ml final concentration) was diluted by a factor of 100 in I PA for spray deposition. The silicon substrate was maintained at 90 °C and 2 ml of the diluted dispersion was applied using an Evolution Airbrush at a pressure of 1.5 bar. Alignment marks were scored onto the substrate after deposition and debris gently blown away by compressed air. Analysis of flakes

The deposited flakes were characterised by scanning Raman, AFM and SEM. Scanning Raman measurements were carried out with a NT_MDT NTEGRA platform. An Argon ion laser at 488 nm was used with a 100X objective lens. The Raman spectra were recorded on a Renishaw Raman Spectroscope using a 1024 X 512 CCD camera. Subsequent tapping mode AFM analysis was performed on a Digital Instruments Nanoscope IIIA using silicon tips with a typical resonance frequency of 320 kHz. SEM images were taken with a Zeiss Ultra Plus SEM.

Both AFM and SEM analysis of the spray coated Si substrates showed large quantities of deposited objects. These ranged in lateral size from ~1 μΐη down to 50 nm. In order to confirm that the observed objects were MoS₂ flakes, Raman analysis was used. The Raman spectra obtained indicated not only that some flakes were very thin but that the MoS₂ structure was of the 2H polytype with no evidence of structural distortion. The spectra indicated that solvent exfoliation gives undistorted 2H-MoS₂ and that no additional structural transformations occur on exfoliation. As it was identified which flakes were MoS₂, more detailed AFM analysis could be performed. It was found that the bigger flakes, with lateral sizes of 500-1000 nm, tended to be thicker than the smaller flakes. This is believed to be due to aggregation during deposition/drying; SEM images of deposited flakes showed clear evidence of aggregation. The total flake thicknesses typically ranged from 5-7 nm, corresponding to approximately 5-7 monolayers.

STM analysis of individual MoS₂ flakes was also carried out. 30 nm of Au was sputter deposited at 1 Angstrom/s on a Si substrate with 300 nm of thermal oxide. MoS₂ flakes were then subsequently deposited using the spray technique described above. The sample was immediately introduced to vacuum and resistively heated at -100 °C for 10 minutes; there was no further surface treatment. STM measurements were made with an electrochemically etched W tip on an Omicron VT-STM with a base pressure of 5x10^"11 mbar. STM images clearly displayed strata which the height scan shows to be defined by steps of 1 nm in height. These steps were believed to represent individual monolayers. This height step was in agreement with the step heights observed by AFM. Electrical characterisation of single flakes

For electrical characterisation, MoS₂ flakes were sprayed onto highly doped silicon substrates with a 300 nm silicon dioxide as dielectric. Electrical contacts were prepared by chromium and gold e-beam evaporation (Temescal FC-2000) after definition in PMMA resist by e-beam lithography in a Zeiss Supra SEM equipped with Raith Elphy Quantum. After lift-off, several thin MoS₂ flakes contacted with two electrodes were obtained with separation as small as 250 nm. Source-drain current-voltage curves have non-linear characteristics, attributed to Schottky barrier contacts and possible trap states at the interfaces between MoS₂ flake and the substrate. The contacted MoS₂ flakes showed typically n-type semiconductor behaviour upon biasing the back gate. Individual flakes showed small on-off ratios (<10) with the mobilities between 0.01 and 0.02 cm²/Vs, as calculated from the linear regime with the gate voltage curves. This is considerably smaller than values of >0.5 cm²/Vs measured for micromechanically cleaved MoS₂.

Example 3: formation of films and composites Formation of films of MoS₂, WS₂ and BN Large volume (300 ml, 5 mg/ml) dispersions of MoS₂ and WS₂ (NVP) were sonicated for 48 hours in a low power sonic bath (Branson 2510E-MT). These were then centrifuged at 500 rpm for 15 minutes and decanted immediately. The concentration of these dispersions was measured as described above. The required volume was then filtered through a 0.45 μιη PVDF (polyvinylidene fluoride) filter membrane. The resulting films were then re-dissolved in 60 ml of a 50:50 DM F: NVP mixture by bath sonication of the coated membrane for 30 min. The resulting dispersions were then filtered onto alumina membranes. The typical pore size of these membranes was -0.02 μιη. Film thickness could be controlled by controlling the volume (and concentration) of dispersion filtered. Films could be prepared with a thickness ranging from a few nm to many tens of microns. Thicker films were robust enough to be free standing when removed from the filter paper. Films were removed from the filter paper by lubricating the filter membrane by first filtering a very small quantity of MoS₂ or WS₂ sediment. After removing any grit, the re-dispersed MoS₂ or WS₂ could be filtered and peeled off the membrane to give very good quality free standing films. All films were dried at 60 °C for 48 hours.

Formation of hybrid films Free standing hybrid films were prepared from 50:50 by weight MoS₂:graphene, WS₂:graphene and BN:graphene. To do this MoS₂, WS₂ were prepared as described above. BN films were prepared in a similar fashion. In each case the film mass was close to 50 mg. Graphene films were prepared by sonicating graphite in DMF for 4 days in a round bottomed flask in low power sonic bath. The dispersion was then centrifuged for 45 min at 500 rpm. The concentration of the graphene was measured as described above to be 0.995 mg/ml. From this dispersion, three portions were removed such that each contained 50 mg dispersed graphene. To prepare the hybrid dispersions the MoS₂, WS₂ and BN films were each placed in the graphene DMF dispersion described above and NVP added such that the DMF/NVP ratio was 50:50. These dispersions were bath sonicated for 30 min and then filtered through 0.45 μιη PVDF filter membranes. The films were dried at 60 °C for 48 hours, after which they could easily be peeled from the membrane.

Thin film hybrids (-200 nm) of MoS₂: graphene, MoS₂: single wall carbon nanotube (SWNT), WS₂: graphene and WS₂:SWNT were also prepared at 20, 40, 50, 60, 80 and 100 wt%. In more detail, MoS₂ and WS₂ dispersions were prepared at an initial concentration of 10 mg/ml in cyclohexylpyrrolidone (CHP). They were sonicated for 3 h in a horn probe sonic tip (VibraCell CVX; 750 W, 75% amplitude, 60 kHz), with ice cooling. The samples were allowed to settle overnight and centrifuged at 1500 rpm for 60 min. The SWNT (lljin nanotech) dispersions were prepared at an initial concentration of 1 mg/ml in CHP. These then received a milder sonication regime of 5 min of point probe sonic tip sonication (20% amplitude), 1 h bath sonication, followed by another 5 min of point probe sonic tip sonication. The SWNT dispersion was not centrifuged in order to improve yield. The graphene dispersions were prepared by adding graphite to NMP at an initial concentration of 3 mg/ml and sonicated for 4 days in a round bottomed flask in a low power sonic bath. This dispersion received a mild centrifugation of 500 rpm for 45 min. After centrifugation of the dispersions, the supernatant was decanted and retained for use. Their absorbance was measured and their concentration determined. Next, the dispersions were blended at the ratio required to give the desired mass fractions and were bath sonicated for 15 min to homogenise. The resulting dispersions were vacuum-filtered onto porous alumina filter membranes (0.02 μιτι pore size, 47 mm diameter). Their thickness was controlled by the volume of dispersion filtered and hence the deposited mass. The films were then dried under vacuum for 24 h at 60 °C.

Thick (-50 μίτι) free standing WS₂/SWNT hybrid films were also made. The SWNT and WS₂ dispersions used are the same as those used to make the thin film hybrids and dispersion preparation is detailed above. Once the dispersions concentrations were determined, they were blended in the ratio required to give the desired SWNT/WS₂ mass fraction. Mass fractions between 0 and 90% WS₂ were prepared. The mixture was then sonicated for 15 min in a sonic bath to homogenise. The resulting dispersions were vacuum-filtered onto porous alumina filter membranes (0.02 μιη pore size, 47 mm diameter). Their thickness was controlled by the volume of dispersion filtered and hence the deposited mass. Films were deposited in a layer by layer method. Ten separate aliquots were deposited sequentially to build up thick free standing films (between 50 and 60 pm). The films were dried under vacuum for 24 h at 60 °C. After drying, the films were peeled away from the filter membrane. Film thickness was measured using a micrometer, and the samples were cut for more detailed characterisation.

Formation of composites

Dispersions of MoS₂ and WS₂ were used to prepare films of known mass as described previously. These films were then placed in vials of thermoplastic polyurethane (TPU) dissolved in DMF at 50 mg/ml. The film mass and TPU/DMF volume were coordinated such that the nano-sheet:TPU mass ratios were 5 wt% or 20 wt%. These dispersions were sonicated for 30 min to disperse the nano-sheets in the TPU/DMF. The resultant composite dispersions were then poured into Teflon trays and the solvent slowly evaporated. The resultant films were dried at 60 °C under vacuum for 48 hours. Example 4: properties of films and composites Mechanical properties All free standing films and composites were cut into strips ~ 2.5 mm wide and -25 mm long. These were mechanically characterised by tensile testing. Three and four strips were measured per sample for free standing films and composites respectively. Tensile testing resulted in stress strain curves, from which the Young's modulus, Y, the tensile strength, UTS, and the strain at break, ε_Β were measured. In addition, for the composites, the stress at low strain (50%) was measured.

In all cases, it was found that the addition of graphene improved the mechanical properties. In the case of the composites, addition of 5% of all fillers improved modulus, stress at low strain and UTS. Increasing filler content to 20% generally increased the modulus and the stress at low strain. The strain at break was reduced at all filler contents. The increases in modulus are of the same order as the best nanotube reinforced elastomers. However, on addition of nanotubes to elastomers, the ductility generally falls catastrophically. This does not occur for these inorganic fillers. This suggests that such fillers may be of use to reinforce elastomers or even thermoplastics.

Electrical properties

The electrical properties of the free standing films were measured, and the results are shown in Table 2. As expected the BN film acted as an insulator, while MoS₂ and WS₂ films exhibited mid-range conductivity. Addition of graphene increased the conductivity significantly in all cases. Inorganic thick film

Inorganic Inorganic: graphene hybrid

(current flow perpendicular to

material (current flow in plane of film)

plane of film)

BN 10^"12 S/m 1270 S/m

MoS₂ 3.6x 10^"5 S/m 3380 S/m

WS₂ 1 .3x 10^"5 S/m 2940 S/m

Table 2: Electrical properties of MoS₂, WS₂ and BN free-standing films and hybrids of each material mixed with graphene. For all hybrids, the graphene content was 50 wt%.

Thermoelectric properties

Electrical conductivity, a_Dc, and Seebeck coefficient, S, also known as thermopower, were simultaneously measured (in plane) by suspending 30 x 7 mm samples of SWNT/WS₂ hybrid films between two thermoelectric devices with thermal paste. These devices were spaced 20 mm apart and used to generate temperature gradients. Silver paste was used to apply four lines on each sample to allow l-V measurements, which verified the Ohmic nature of the contacts. Temperature differences and thermoelectric voltages were measured using T-type thermocouples, containing copper leads with negligible thermopower (~1 .83 μ\//Κ), attached to the metal lines on opposite ends of the sample strip. Thermoelectric voltages were measured as the temperature gradient was increased to ±10 °C, which allowed the Seebeck coefficient to be taken from the slope. The DC conductivity and Seebeck measurements were made in the plane of the film. The Seebeck coefficient was typically between 60 and 80 μ\//Κ, significantly lower than the literature value of 1000 μ\//Κ for disordered WS₂ films. Importantly, the data indicated that the DC conductivity can be increased without significant degradation of S and so the power factor (a_DcS²). More importantly, adding nanotubes increases the power factor (by a factor of -750 much faster than it increases the thermal conductivity (by -30). This means the addition of nanotubes clearly improves the thermoelectric properties of layered compounds.

The dimensionless figure of merit was also calculated. It was found that the figure of merit increased with nanotube content, displaying typical values of -0.001. This is an important result as it shows that liquid phase exfoliation allows the addition of nanotubes (or other conducting nanoparticles) to layered compounds resulting in an increase in thermoelectric efficiency.

Example 5: exfoliation of other inorganic layered compounds

The techniques described above were used to exfoliate a number of other inorganic layered compounds, including MoSe₂, MoTe₂, TaSe₂, NbSe₂, Bi₂Te₃, and NiTe₂. Powdered material was added to a 10 ml volume of solvent in 12 ml vial at 10 mg/ml. This was point probe sonicated (30 min, 38% power) with ice cooling and immediately centrifuged at 1500 rpm for 60 min. Good dispersions were obtained, and the absorption spectra for NiTe₂ and NbSe₂ were found to be consistent with metallic behaviour as expected from the known electronic properties of these materials. The dispersions could easily be formed into thin films by filtration. Bright and dark field STEM was performed on holey carbon grids onto which had been cast a few drops of the dispersions. Figure 3 depicts bright field (left) and high angle annular dark field STEM images (right) of various flakes. In all cases, thin flakes were observed. In some cases, dark field STEM allowed layer counting. For example, the Bi₂Te₃ flake shown in Figure 3 contained 4-5 layers.

Claims

1. A process for producing a two-dimensional material, the process comprising contacting a three-dimensional inorganic layered material with a solvent under conditions such that the three-dimensional material is exfoliated to form a two-dimensional material, wherein the three-dimensional material is a metal-containing material or a clay mineral, and wherein the solvent has a dispersive Hansen solubility parameter (5_D) of from about 10 to about 22 MPa^1/2.

2. The process of claim 1 , wherein the three-dimensional material comprises a metal-containing material.

3. The process of claim 2, wherein the three-dimensional material comprises a metal-containing material selected from metal chalcogenides, metal oxides, Sb₂Te₃, TiNCI, Bi₂Te₃ and Bi₂Se₃.

4. The process of claim 3, wherein the metal-containing material comprises a transition metal dichalcogenide of the formula MX₂, wherein M is Ti, Zr, Hf, V, Nb, Ta, Cr, Mo, W, Tc, Re, Ni, Pd or Pt; and each X is S, Se or Te.

5. The process of claim 4, wherein the three-dimensional material comprises MoS₂ or WS₂.

6. The process of claim 1 , wherein the three-dimensional material comprises a clay mineral.

7. The process of any preceding claim, wherein the solvent has a dispersive Hansen solubility parameter (5_D) of from about 12 to about 21 MPa^1/2.

8. The process of claim 7, wherein the solvent has a dispersive Hansen solubility parameter (5_D) of from about 15 to about 20 MPa^1/2.

9. The process of any preceding claim, wherein the solvent has a polar Hansen solubility parameter (δ_Ρ) of from about 1 to about 20 MPa^1/2.

10. The process of any preceding claim, wherein the solvent has a hydrogen-bonding Hansen solubility parameter (δ_Η) of from about 1 to about 20 MPa^1/2.

1 1. The process of any preceding claim, wherein the solvent has a surface energy of from about 60 to about 80 mJ/m² and/or a surface tension of from about 30 to about 50 mJ/m².

12. The process of any preceding claim, wherein the solvent comprises an organic solvent.

13. The process of claim 12, wherein the solvent comprises a solvent selected from acetone, benzaldehyde, benzyl benzoate, benzyl ether, benzonitrile, bromobenzene, chlorobenzene, cyclohexylpyrrolidinone, chloroform, cyclohexane, cyclohexanone, dimethylacetamide, dimethylformamide, dimethylimidazolidinone, dimethylsulphoxide, N- dodecylpyrrolidone, formamide, isopropanol, methanol, N-methylformamide, N-methyl- pyrrolidinone, N-octylpyrrolidone, quinoline and N-vinylpyrrolidinone.

14. The process of claim 13, wherein the solvent comprises a solvent selected from N-methylpyrrolidone, N-vinylpyrrolidone, cyclohexylpyrrolidone, isopropanol and dimethyl formamide.

15. The process of any preceding claim, wherein the three-dimensional material is in the form of a powder.

16. The process of any preceding claim, wherein the process comprises mixing the three-dimensional material with the solvent to form a mixture and applying energy, e.g. ultrasound energy, and/or a force, e.g. a centrifugal force, to said mixture.

17. The process of any preceding claim, wherein a dispersion of the two-dimensional material is formed.

18. The process of claim 17, wherein the dispersion comprises flakes of the two- dimensional material having a thickness of 10 nm or less.

19. The process of claim 17 or claim 18, wherein the dispersion comprises one or more further materials selected from nanomaterials.

20. The process of claim 19, wherein the one or more further materials are selected from graphene, boron nitride, nanotubes and metallic nanowires.

21. The process of any preceding claim, wherein the process further comprises forming a film comprising the two-dimensional material on a substrate.

22. The process of claim 21 , wherein the film has a thickness of from about 5 nm to about 1 mm, e.g. from about 10 nm to about 500 nm.

23. The process of claim 21 or claim 22, wherein the film is formed by vacuum filtration, spraying, dip coating, Mayer rod coating, screen printing, inkjet printing or Langmuir Blodgett deposition.

24. The process of any preceding claim, wherein the process further comprises preparing a composite material comprising the two-dimensional material and a polymer.

25. The process of any preceding claim, further comprising forming a device comprising the two-dimensional material.

26. Use of a solvent for the production of a two-dimensional material by exfoliation of a three-dimensional inorganic layered material, wherein the three-dimensional material is a metal-containing material or a clay mineral, and wherein the solvent has a dispersive Hansen solubility parameter (5_D) of from about 10 to about 22 MPa^1/2.

27. A two-dimensional inorganic material obtainable by a process of any of claims 1 to 25, wherein the two-dimensional material comprises a metal-containing material or a clay mineral.

28. The material of claim 27, wherein the two-dimensional material comprises a metal-containing material selected from metal chalcogenides, metal oxides, Sb₂Te₃, TiNCI, Bi₂Te₃ and Bi₂Se₃.

29. The material of claim 28, wherein the material comprises a transition metal dichalcogenide of the formula MX₂, wherein M is Ti, Zr, Hf, V, Nb, Ta, Cr, Mo, W, Tc, Re,

Ni, Pd or Pt; and each X is independently S, Se or Te.

30. The material of claim 29, wherein the material is MoS₂ or WS₂.

31. The material of claim 27, wherein the two-dimensional material comprises a clay mineral.

32. The material of any of claims 27 to 31 , wherein the two-dimensional material is present in a film on a substrate.

33. The material of any of claims 27 to 32, wherein the material is dispersed in a solvent having a dispersive Hansen solubility parameter (5_D) of from about 10 to about 22 MPa^1/2.

34. A device comprising a two-dimensional material of any of claims 27 to 33.

35. The device of claim 34, wherein the device is selected from electrodes, capacitors, transistors, solar cells, light emitting diodes, thermoelectric devices, dielectrics and batteries.