Field

[0001 ] This specification relates to exfoliation of agglomerated materials so as to produce a dispersion of discrete molecular layers.

Priority

[0002] This application claims priority from Australian Provisional Patent Application No. 2015904218, the entire contents of which are incorporated herein by cross-reference.

Background

[0003] Various substances which are composed of extended benzenoid structures have recently attracted a great deal of attention due in part to their thermal and electrical properties. A well- known example of such materials is graphene, the exfoliated form of graphite. Another example is carbon nanotubes, which may be regarded as cylinders of graphene-like material. These may be single walled (SWCNT) or multi walled (MWCNT). Other examples include carbon whiskers, fullerenes and carbon fibres. Certain inorganic materials, such as molybdenum sulfide, are also known to have a graphite-like layered structure, and it is known that such materials can be separated into their deagglomerated layers. In this specification the term "molecular layer" will be used to encompass materials as set out above having extended benzenoid sheets (either flat, as in graphene, cylindrical as in carbon nanotubes, or in any other form) as well as inorganic materials in the form of extended sheets, and the term "agglomerated molecular layer" will be used to refer to the agglomerated form of such materials (i.e. graphite, agglomerated nanotubes etc.).

[0004] One problem with manipulation of these materials is their tendency to agglomerate, driven by the energetic benefit of extensive van der Waals forces, specifically π-π interactions. Thus graphene can agglomerate to form a graphite-like material, and indeed occurs naturally as graphite, which may be viewed as agglomerated graphene sheets. Similarly, carbon nanotubes can agglomerate to form ropes, bundles or aggregates. [0005] Charge transfer (CT) interactions are intermolecular interactions between π electron rich (donor) and π electron deficient (acceptor) molecules. CT interaction can be effectively used to cleave and interrupt π-π interactions between molecular layers resulting in deaggregation. The resulting CT complex shows a characteristic absorption band in the visible region providing evidence of association. For this to be realized, the design and strength of the electron acceptors is important. This is because the breaking of π-π interaction has energetic consequences. Weak electron acceptors or donors such as fullerene (C ₆ o) do not match the energetic cost associated with such cleavage and hence are poor CT additives.

[0006] Graphene is a two dimensional crystalline nanomaterial that consists of a single atomic layer of carbon bonded together in a hexagonal lattice. The term "graphene" is also used to refer to small numbers, e.g. less than about 10, of such layers which are laminated together. Graphene has attracted a great deal attention due to its extraordinary physical and chemical properties. Recently, research has been geared towards increasing the quality and yield of graphene exfoliation so as to ensure practical applications in various areas. Direct liquid phase exfoliation of graphite powder in a well-chosen organic solvent has the potential to produce materials for use in applications such as functional coatings, conducting inks, composite, batteries, supercapacitors and top down approaches to electronics. In this method, solvents that have matching surface energies with graphene, such as N-methylpyrrolidone (NMP) or dimethylformamide (DMF), can be directly used to exfoliate graphite through simple sonication.

[0007] However, this method has a significant disadvantage in that the yield is very low, typically 0.01 mg/mL of exfoliated graphene, and improving the yield requires very long sonication times which is often impractical for large scale applications. Furthermore, the process is limited to solvents with a well matched surface energy, which are often either too expensive or have high boiling points rendering further processing impractical or of limited utility.

[0008] Extending liquid phase exfoliation of graphene to lower boiling point, non-polar solvents such as chloroform, which have relatively poor matching surface energy, would thus be advantageous in increasing the range of solvents available for many graphene applications such as preparation of polymer-graphene composites, since many commodity polymers are soluble in these solvents. Also, reducing the sonication time would be advantageous, since extended sonication times can cause rupture of graphene sheets, leading to a poor quality product with small sheets. [0009] Moreover, since the best solvent that is currently known for liquid phase exfoliation of graphene in organic solvent is NMP, a solvent that is currently on the European candidate list of substances of high concern due to its toxicity, it would be beneficial to identify alternative, greener solvents for graphene exfoliation so as to meet environmental and safety standards. In addition to exfoliation in greener organic solvents, eliminating the need for solvents all together during composite formation would not only be highly attractive from an environmental conservation perspective but would also greatly reduce the processing costs associated with highly priced solvents.

[00010] There is also a need, in certain applications, to modify the electrical properties of graphene. The zero gap characteristic of pristine graphene, with attendant difficulty in development of a bulk non-covalent approach to both open and tune the band gap without destroying the sp ² graphene basal planes, has also limited its applications in electronic devices. For instance, graphene based transistors have been considered as a potential substitute for the conventional silicon semiconductor-based microelectronics. This is due to the fact that modern logic circuits are based on silicon complementary metal oxide semiconductor (CMOS) technology. Thus, band gap tuning and exploring the electronic properties of graphene is of fundamental importance for many applications, especially in field effect transistors (FETs) where both p-type and n-type conductions are desired to construct complex logic circuits. Solution processing of bulk non- covalently doped graphene without the need of special equipment would open up additional potential of low cost method for fabrication of semiconductor and electronic devices.

[0001 1 ] It is an aim of the present invention to at least partially overcome at least one of the above disadvantages of previous methods and products. It is a further aim to at least partially satisfy at least one of the above needs.

Summary of Invention

[00012] In a first aspect of the invention there is provided a method for producing a dispersion of molecular layers in an organic liquid, said method comprising: a) combining an agglomerate of said molecular layers with an electron-deficient aromatic compound; b) before or after step a), combining said agglomerate with the organic liquid; and c) after step a), providing mechanical energy to said agglomerate so as to separate said agglomerate into molecular layers. [00013] The agglomerate of molecular layers may be graphite or it may be aggregated carbon nanotubes or it may be an inorganic layer material such as the transition metal dichalcogenide molybdenum disulfide or it may be a mixture of these. In the latter case, the method may form a hybrid material.

[00014] The following aspects may be used in conjunction with the first aspect, either individually or in any suitable combination.

[00015] Step b) may be conducted after step a). In this case step c) may be conducted at least in part after step b).

[00016] The step of providing mechanical energy may comprise sonicating a mixture which comprises the agglomerate, the electron-deficient aromatic compound and the organic liquid.

[00017] The electron-deficient aromatic compound may have one or more pendant hydrophobic chains. The hydrophobic chains may be hydrocarbon chains. They may be branched chains. The electron-deficient aromatic compound may be a cyclic imide. It may be a bisimide. The imide may have a hydrophobic chain attached to the nitrogen atom thereof.

[00018] The ratio of the amounts of agglomerate to the electron-deficient may be between about 1 : 10 and about 1000: 1 on a weight basis.

[00019] The method may additionally comprise step d) separating residual agglomerate from the liquid following steps a), b) and c).

[00020] The method may additionally comprise filtering the dispersion so as to provide a film of the exfoliated molecular layers.

[00021 ] In one embodiment there is provided a method for producing a dispersion of molecular layers in an organic liquid, said method comprising: a) combining an agglomerate of said molecular layers with an electron-deficient aromatic compound; b) after step a), combining said agglomerate with the organic liquid; and c) after step b), sonicating the organic liquid containing the agglomerate so as to provide mechanical energy to said agglomerate so as to separate said agglomerate into molecular layers. [00022] In another embodiment there is provided a method for producing a dispersion of molecular layers in an organic liquid, said method comprising: a) combining an agglomerate of said molecular layers with an electron-deficient aromatic compound and grinding said agglomerate and said electron-deficient aromatic compound together; b) after step a), combining said agglomerate and electron-deficient aromatic compound with the organic liquid; and c) after step b), sonicating the organic liquid containing the agglomerate and electron-deficient aromatic compound so as to provide mechanical energy to said agglomerate so as to separate said agglomerate into molecular layers.

[00023] In a further embodiment there is provided a method for producing a dispersion of molecular layers in an organic liquid, said method comprising: a) combining an agglomerate of said molecular layers with an electron-deficient aromatic compound and grinding said agglomerate and said electron-deficient aromatic compound together; b) after step a), combining said agglomerate with the organic liquid; and d) separating residual aggregated agglomerate from the liquid.

[00024] In a second aspect of the invention there is provided a dispersion of molecular layers produced by the method of the first aspect.

[00025] In a third aspect of the invention there is provided a film comprising molecular layers, said layers having an electron-deficient aromatic compound on a surface thereof. The film may consist of, or consist essential of, the molecular layers and the electron-deficient aromatic compound on a surface thereof.

[00026] The electron-deficient aromatic compound and the molecular layers may form a charge transfer complex. The film may have a conductivity of at least about 5000 S/m. In this instance, the molecular layers may comprise an extended benzenoid structure, for example graphene.

[00027] The film of the third aspect may be made by preparing a dispersion by the method of the first aspect and then filtering said dispersion to provide the film. Thus the process of the first aspect may be used to make the film of the third aspect.

[00028] In a fourth aspect of the invention there is provided a dispersion of particles, said particles comprising a charge transfer complex between molecular layers and an electron deficient aromatic compound. [00029] The dispersion may be made by the process of the first aspect. The process of the first aspect may make the dispersion of the fourth aspect. The film of the third aspect may be made by filtering the dispersion of the fourth aspect. The filtering may be through a filter of pore size less than about 50nm.

Brief Description of Drawings

[00030] Figure 1 is a diagram illustrating the concept of exfoliation of graphite into pristine graphene caused by CT interaction. Fig. 1 also includes structures of electron acceptors Al and A2.

[00031 ] Figure 2 shows optical images of a) acceptor Al plus graphite a) before, and b) after, grinding.

[00032] Figure 3 shows XRD diffraction patterns: (a) raw graphite powders; (b) graphite ground with Al; (c) graphite ground with A2; (d) graphite exfoliated with Al in NMP; (e) graphite exfoliated with A2 in NMP.

[00033] Figure 4 shows optical images of graphene dispersions: a) in NMP at different concentrations of Al before sonication (Electron acceptor concentration shown in %w/w); b) in NMP at different concentrations of A l followed by sonication for 30 minutes and centrifugation, c) in DMF at different concentrations of Al followed by sonication for 30 minutes and centrifugation; and d) in CHC1 ₃ at different concentrations of Al followed by sonication for 30 minutes and centrifugation..

[00034] Figure 5 shows the effect of weight of a) Al and b) A2 on the concentration of exfoliated graphene in NMP, DMF and CHCI ₃ .

[00035] Figure 6 shows a summary of concentrations of graphene dispersions in organic solvent Cyrene, NMP, DMF, chloroform, DCM (dichloromethane), acetone, anisole, 1,2-DCB, CH- ONE (cyclohexanone), epoxy resin (bisphenol A diglycidyl ether) and MAA with or without addition of acceptor. [00036] Figure 7 shows UV/Vis spectra of A l (dashed line) and graphene plus A l (solid line) in NMP at room temperature and pressure showing evidence of CT complexation via appearance of CT band at about 380 nm.

[00037] Figure 8 shows SEM images of surface of films of a) without acceptor b) with Al and c) with A2 exfoliated graphite.

[00038] Figure 9 shows Raman spectra 532.1 nm for a) various graphite exfoliated films at different acceptor concentrations in NMP; b) the raw material graphite shown for comparison.

[00039] Figure 10 shows concentrations of MWCNTs dispersed in DMF, Cyrene, NMP and chloroform without addition of acceptor (G) and with acceptors Al and A2.

[00040] Figure 1 1 shows photographs of MWCNTs a) mixed with Al or A2, MWCNTs ground with Al or A2, b) MWCNTs dispersed in NMP with Al or A2 before sonication c) MWCNTs sonicated with Al or A2 for 30 minutes, the centrifuged at 3000 rpm and the top half of the resulting mixture separated and d) stable dispersions of MWCNTs in various solvents.

[00041 ] Figure 12 shows an NMR spectrum of Al .

[00042] Figure 13 shows an NMR spectrum of A2.

Description of Embodiments

[00043] The present invention relates to a method for dispersing molecular layers in an organic liquid. The term "molecular layer" covers substances composed of carbon sheets or other layer materials such as transition metal dichalcogenides (e.g. molybdenum sulfide, tungsten sulfide, molybdenum diselenide, tungsten diselenide and bismuth telluride), clays, talc and boron nitride (particularly hexagonal boron nitride, "h-BN"), either fully isolated or in only a few layers, as well as extended benzenoid sheet materials (i.e. made up of networks of sp carbon atoms, and designated herein as SP2C). The particles of such materials may comprise between 1 and about 10 layers, or 1 and about 5, e.g. about 1 , 2, 3, 4, 5, 6, 7, 8, 9 or 10 layers. The layers may be planar, e.g. graphene, molybdenum sulfide etc. Alternatively the SP2C may be in the form of carbon nanotubes, in which the layers are in the form of cylinders, or buckminsterfullerenes in which the layers are in the form of spheres or approximate spheres. These may be viewed as cylindrical or spherical forms of graphene. Alternatively equivalents of these structures (i.e. nanotubes, buckminsterfullerenes) may be constructed from other layer forming materials as described above, i.e. metal dichalcogenides, clays, talc and boron nitride, and these materials are also envisaged as being encompassed by the term "molecular layer". The carbon nanotubes may be single walled (SWCNT) or multiwalled (MWCNT). As described above, they may have from 1 to about 10 walls. Other examples of molecular layers include fullerenes, carbon whiskers and carbon fibres. The molecular layers may be regarded as electron rich molecular networks or as pi-electron donating networks.

[00044] In either of the above options (graphene, carbon nanotubes etc.), π-π interactions can cause aggregation of the particles. In many applications however it is desirable to have the layers separated rather than agglomerated. In the past this has been achieved by extended sonication, however this can lead to rupture of the particles, resulting in smaller or poorly formed particles. The present invention presents a method for deaggregation (delamination in the case of graphite, which is simply laminated or aggregated graphene, and other laminar materials) to form the molecular layers under relatively mild conditions so as to facilitate the deaggregation.

[00045] The starting material for the method of the invention is the aggregate of the molecular layers. Thus in the case of graphene, the starting material is graphite and in the case of carbon nanotubes the starting material will be an aggregate, commonly a rope or bundle of carbon nanotubes. This is combined with an electron-deficient aromatic compound. Such compounds may be regarded as electron acceptors. The inventors have found that the electron-deficient aromatic compound can interpose between the individual layers or tubes of the molecular layers so as to reduce the interactions between them. This greatly facilitates the deaggregation process. A certain amount of mechanical energy needs to be introduced into the system in order to achieve the desired deaggregation. The inventors have surprisingly found that even grinding the aggregates with an electron-deficient aromatic compound can lead to a certain degree of deaggregation. However in order to achieve bulk deaggregation it may be useful to apply other forms of energy input, e.g. vortex mixing, ball milling or mild sonication (either indi vidually or in combination) in addition to, or instead of, grinding.

[00046] In one form therefore, the method of the invention involves combining the aggregates with an electron-deficient aromatic compound and then providing mechanical energy by grinding the two together. Following this, a suitable carrier liquid, commonly an organic liquid, is added so as to suspend the aggregates and the deaggregated molecular layers. The remaining aggregates can then be removed if required. This may be achieved by centrifugation, settling, filtration through a suitable filter which has pore size between the size of the deaggregated molecular layers particles and the aggregates, or by some other suitable method. Alternatively, the suspension can be subjected to further mechanical energy, commonly by sonication, in order to deaggregate at least a part of the remaining aggregates. Following this, any residual aggregates may be removed as described above.

[00047] In another form, the aggregates may be simply suspended in a suitable carrier liquid with the electron-deficient aromatic compound and the resulting suspension sonicated.

[00048] As will be recognised from the two options set out above, the important aspect of the invention is that mechanical energy, either in the form of grinding or sonication or indeed in any other suitable form, or in a mixture of such forms, is applied to a mixture of the aggregates and an electron-deficient aromatic compound and at some stage a carrier liquid is added. The precise stage at which the liquid is added is not critical, and the mechanical energy may be supplied at a single stage or at multiple stages.

[00049] The nature of the carrier liquid is not critical to this process. The carrier liquid may be any liquid in which the end use requires the molecular layers to be dispersed. It is commonly an organic liquid. It may be a polar liquid or may be a non-polar liquid. It may be aromatic or may be non-aromatic. It may for example be a polar aprotic solvent, an alcohol, an ether, an ester, benzene, a benzene derivative, a heterocycle (e.g. pyridine, dioxane) or some other suitable liquid. In some instances it may be mixed with water, and in other instances it may be substantially anhydrous. The carrier liquid may be a liquid in which the electron-deficient aromatic compound has a significant solubility. It may have a solubility at the temperature at which the method is conducted of at least about 0.1 %, or at least about 0.5 or 1% on a w/w or w/v basis. In some embodiments the liquid is not an electron donor, or at least is a weaker electron donor than the molecular layers. This prevents or inhibits formation of a donor-acceptor pair with the electron-deficient aromatic compound which might impair the ability of the electron-deficient aromatic compound to interact with the molecular layers or aggregate thereof. [00050] In the present context, a "liquid" is taken to be any substance which is in the liquid state at the temperature at which the method is conducted. Thus for example if a dispersion in naphthalene were to be required, the method could be conducted at a temperature above the melting point of naphthalene (about 81 °C). The resulting suspension of molecular layers may then be cooled if required to provide a solid dispersion of the molecular layers.

[00051 ] The ratio of the aggregate to the electron-deficient aromatic compound may be from about 1 : 10 (i.e. 0.1) to about 1000: 1 (i.e. 1000), depending on the nature of the aggregate and the electron-deficient aromatic compound. The ratio of the aggregate to the electron-deficient aromatic compound may be from about 0.1 to about 1000, or 0.1 to 500, 0.1 to 200, 0.1 to 100,

0.1 to 50, 0.1 to 20, 0.1 to 10, 0.1 to 5, 0.1 to 2, 0.1 to 1 , 0.1 to 0.5, 1 to 1000, 2 to 1000, 5 to 1000, 10 to 1000, 50 to 1000, 100 to 1000, 200 to 1000, 500 to 1000, 1 to 100, 1 to 10, 10 to 100, 10 to 500, 50 to 500, 100 to 500 or 50 to 100, e.g. about 0.1 , 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9,

1 , 2, 3, 4, 5, 6, 7, 8, 9, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 150, 200, 250, 300, 350, 400, 450, 500, 600, 700, 800, 900 or 1000. For graphite, a common ratio would be from about 0.1 to about 10, or about 0.1 to 5, 0.1 to 2, 0.1 to 0.1, 0.1 to 0.5, 0.5 to 10, 1 to 10, 2 to 10, 5 to 10 or 1 to 5, e.g. about 0.1 , 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1 , 2, 3, 4, 5, 6, 7, 8, 9 or 10. For CNTs a more common ratio would be about 100 to about 1000, or about 100 to 500, 100 to 200, 200 to 1000, 500 to 1000 or 200 to 500, e.g. about 100, 150, 200, 250, 300, 350, 400, 450, 500, 600, 700, 800, 900 or 1000.

[00052] An important aspect of the invention is the electron-deficient aromatic compound. This is generally an aromatic ring, or aromatic ring system, having electron -withdrawing groups attached to it. It may have more than one electron withdrawing groups attached, e.g. 2, 3, 4 or 5. Suitable electron-withdrawing groups include carbonyl-containing functional groups such as amides, esters, ketones, aldehydes, anhydrides, imides etc. Other electron withdrawing groups such as halides, nitro groups etc. may present additionally or alternatively. The electron- deficient aromatic compound may have one or more hydrophobic chains, e.g. 1 , 2, 3, 4 or 5, attached thereto. It is thought that these may stabilise the molecular layers in the carrier liquid relative to the aggregated form by reducing the intermolecular and/or surface forces between the layers of the aggregate. This may particularly be the case when the carrier liquid is hydrophobic or non-polar. Suitable hydrophobic chains include hydrocarbon chains (optionally substituted, e.g. with aromatic groups), polyoxypropylene chains ((OCH ₂ CH(CH ₃ )) _n ) etc. Alternatively, if the carrier liquid is a polar or hydrophilic liquid, the chains may be hydrophilic, e.g. polyoxyethylene chains. The chain length may be from about 6 to about 20 atoms in length, or about 6 to 10, 10 to 20 or 8 to 18 carbons in length, e.g. about 6, 7, 8, 9, 10, 1 1 , 12, 13, 14, 15, 16, 17, 18, 19 or 20 carbons in length. The chain may be straight chain or branched. It may contain aromatic groups and/or alicyclic groups. Particularly suitable electron-deficient aromatic compounds include bisimides. These are suitable as they are readily synthesised and can incorporate a large range of chains. In some instances the electron-deficient aromatic compound may contain one or more chiral centres. In this instance it may be present as a racemate, or may contain the different optical isomers and/or diastereomers in different ratios. It is thought that the presence of these different isomers may inhibit crystallisation of the electron-deficient aromatic compound and thereby improve its solubility in the carrier liquid.

[00053] The method of the invention may be conducted without addition of surfactant, or without addition of an amphiphile. The electron-deficient aromatic compound in this case is not regarded as an amphiphile or a surfactant but may act as a compatibliser between the carrier liquid and the molecular layers. Thus the method may be conducted without addition of surfactant, or without addition of an amphiphile, or without the addition of a compatibliser, other than the electron-deficient aromatic compound.

[00054] The method may be conducted at any suitable temperature, preferably above the melting point of the carrier liquid. It may be conducted at a temperature between about 0 and about 100°C, or between about 0 and 80, 0 and 60, 0 and 40, 0 and 20, 20 and 100, 50 and 100, 10 and 50, 10 and 20, 20 and 50 or 30 and 70°C, e.g. about 0, 10, 20, 30, 40, 50, 60, 70, 80, 90 or 100°C, or may be at some other suitable temperature.

[00055] The dispersions produced above may be filtered in order to provide films of the molecular layers. The exfoliated molecular layers of this film have the electron-deficient aromatic compound on a surface, optionally on the majority or substantially all of the surfaces, thereof. The electron-deficient aromatic compound and the exfoliated sp ² carbon network may form a charge transfer complex. This may render the film semiconducting. The film may have a conductivity of at least about 5000 S/m (i.e. 50S/cm). It may have a conductivity of at least about 100, 150, 200, 250, 300, 350, 400, 450, 500, 600 or 700S/cm, or from about 50 to about lOOOS/cm, or about 50 to 800, 50 to 700, 50 to 500, 50 to 300, 50 to 100, 100 to 1000, 200 to 1000, 500 to 1000, 500 to 800, 100 to 500, 100 to 200, 200 to 500 or 500 to 700S/cm, e.g. about 50, 60, 70, 80, 90, 100, 150, 200, 250, 300, 350, 400, 450, 500, 550, 600, 650, 700, 750, 800, 850, 900, 950 or l OOOS/cm.

[00056] Filtration of the dispersion may be microfiltration or may be ultrafiltration. It may be through a filter of pore size below that of the maximum dimension of the molecular layers. It may be through a filter of pore size less than about 50nm, or less than about 40, 30, 20 or lOnm, or from about 10 to about 50nm or about 10 to 30, 30 to 50 or 15 to 30nm, e.g. about 10, 20, 30, 40 or 50nm. This may be a maximum, mean or nominal pore size. The filtration may be facilitated by a vacuum on the downstream side of the filter and/or by increased pressure on the upstream side of the filter. The filter may be capable of withstanding the applied pressure difference across it. It may be a sintered metal filter. It may be an aluminium filter. Alternatively it may be a filter with the pore size described above supported on a support filter, said support filter having a larger pore size than the filter and being capable of withstanding the applied pressure difference across it.

[00057] The present invention relates in certain embodiments to a novel method for the production of pristine graphene sheets through exfoliation of graphi te in the presence of electron accepting molecules. Suitable electron acceptors have conjugated aromatic structures linked to electron withdrawing chemical species. Branching of the pendant group has been found to improve the exfoliation ability of the acceptors. It is thought that delamination of graphene sheets occurs to some extent in the solid state, although dispersions in solvents allow improved exfoliation. The solvents that can be used to prepare stable suspensions include aprotic polar solvents (e.g. NMP, DMF, DMAc), halogenated solvents (e.g. chloroform, dichloromethane, dichlorobenzene), ketones (e.g. acetone, cyclohexanone) and ethers (anisole). The use of electron acceptors results in a significant increase in concentrations of the graphene suspension. Furthermore, very little energy is required for exfoliation. By contrast, in other methods, typically extended sonication or high shear apparatus are necessary.

[00058] The present method therefore results in sheets with greater lateral dimensions and fewer defects, leading to a great improvement in the aspect ratio and electrical conductivity of the graphene sheets. In addition to these vastly improved properties, it appears that the acceptors are effectively doping the graphene particles, leading to semi-conducting properties. This enables many applications of the product, including composites, battery electrodes, conducting films, lubrication, etc. Higher concentrations of graphene than previously described have been achieved in a broad range of non-aqueous solvents. The range of solvents in which significant graphene concentrations can be prepared has been significantly expanded, providing novel potential uses.

[00059] Most commercially available "graphene" on the market is based on graphene oxide or reduced graphene oxide. The material generated by the invention described here can be described as pristine, that is, defect free.

[00060] Higher concentrations of the dispersions allow greater amounts of graphene to be incorporated into products. For example previous methods for dispersion of graphene in NMP achieves only about 5g/300L. The presently described method reduces the quantity of solvent (i.e. increases the concentration) by at least an order of magnitude. Significantly lower production costs are a result. Flexibility of solvent systems provides improved compatibility with existing production processes, leading to quicker uptake of graphene technology in to existing markets. Additionally, the present process reduces the energy needed to exfoliate graphene substantially. As a result, the present method substantially reduces costs of preparing graphene and enables use of solvent systems that are more environmentally friendly than currently used solvents.

Examples

Example 1: Exfoliation of graphene

[00061 ] Herein is described the successful exfoliation of graphene in organic solvents stabilized by a non- covalent electron donor (graphene) - electron deficient acceptor (A ^' l and A2) interaction with low power, short sonication times required. The yield of graphene was greatly improved (over tenfold) in most organic solvents relative to that without acceptor. The acceptor is considered to serve a dual role in that it both weakens the attractions between lamellae in graphite and also stabilizes the exfoliated graphene resulting in stable, high quality graphene dispersions after exfoliation. The graphene concentration achieved in NMP was 0.14 mg/mL, 0.1 mg/mL in CHCI ₃ and 0.08 mg/mL in DMF after 30 minutes sonication compared to 0.01 mg/mL without acceptor.

[00062] Relatively weak Van der Waals forces (π-π stacking) in graphite play a key role in holding the graphene layers together. This attractive force between adjacent layers can be weakened by charge transfer (CT) between graphene and an external molecule, resulting in exfoliation. This can be facilitated by sonication, which imparts shear that aids in overcoming the forces between the layers. In the present experiments the inventors explore the introduction of an external electron deficient acceptor (A1/A2) with mild sonication to induce interlayer separation in graphite due to donor-acceptor (CT complex) interaction, which serves a dual role in exfoliation and stabilizing the exfoliated graphene (Fig. 1 ).

[00063] Graphene was produced using a batch liquid phase exfoliation technique. This procedure has the advantage of producing liquid phase graphene without oxidation or substantial defect formation. Synthetic graphite, 2-ethylhexylamine, pyromellitic dianhydride (PMDA), naphthalene-l ,4,5,8-tetracarboxylic acid dianhydride, organic solvents: NMP (N-methyl-2- pyrrolidone), DMF (dimethyl formamide), chloroform, DCM (dichloromethane), acetone, anisole, 1 ,2-DCB (1,2-dichlorobenzene), CH-ONE (cyclohexanone)), hexane, toluene, iso- propanol, dim ethyl sulph oxide (DMSO)) and epoxy resin (bisphenol A diglycidyl ether and methacrylic acid (MAA) were used as received from Sigma Aldrich. Cyrene was obtained from Circa Group Pty Ltd.

[00064] Compounds Al and A2 were synthesized as described below.

Synthesis of Acceptor Al

A1

[00065] 4365 mg (20 mmol) of pyromellitic dianhydride was degassed and 44 mL of dehydrated DMF was added. 5685 mg (44 mmol) of 2-ethylhexylamine was added and the mixture heated at 150 °C at reflux under nitrogen atmosphere for 24 hours. The reaction mixture was cooled, the organic layer washed with water then extracted with dichloromethane. The organic layer was dried using magnesium sulphate and filtered and the solvents evaporated. The crude product was purified using column chromatography on silica gel; eluent dichloromethane/hexane (2: 1). The yield was 58% (5200 mg). The 1H NMR spectrum (42.5 MHz, CDCL ₃ r. t.) of Al is shown in Fig. 12. Synthesis of Acceptor A2

[00066] 2.7 g (10.0 mmol) of naphthalene-l,4,5,8-tetracarboxylic acid dianhydride was degassed in a round bottomed flask and 50 mL of DMF (dehydrated) was added. 2.9 g (22.0 mmol) of 2-ethylhexylamine was then added and the resulting mixture heated to 150 °C at reflux under nitrogen gas for 24 hours. The reaction mixture was cooled, poured into water and the organic layer extracted with dichloromethane and again washed with water. The organic layer was dried using magnesium sulphate and filtered and the solvents evaporated under reduced pressure. The crude product was purified using column chromatography silica gel; eluent dichloromethane. The yield was 84 % (4000 mg). The 1H NMR spectrum (42.5 MHz, CDCI ₃ , r. t.) of A2 is shown in Fig. 13.

Exfoliation of graphene

[00067] 50 mg of graphite powder (5mg/mL) was ground with A1/A2 (10/100 mg) followed by addition of 10 mL of organic solvent, polymer or monomer. Identical suspensions were prepared in solvent only without A1/A2 as a standard. The suspensions were left for at least one week to maximise donor-acceptor interaction followed by sonication for 30 minutes (except when using acetone and DCM in which case sonication was for 1 hr) using a low power bath sonicator (Unisonics, Australia, 50 Hz). Sonication with a relatively high power (Qsonica© Q700 operating at approximately 100W) microtip was also investigated for feasibility in industrial process. The suspension was centrifuged using Eba 20, Hettich, Zentrifugen centrifuge at 3000 rpm for 30 minutes to sediment larger, non-exfoliated graphite materials and the top half pippeted off, giving stock graphene suspension. The graphene dispersions were) then used to prepare conducting films by vacuum filtration onto a porous alumina membrane 20 nm (0.02 μηι) pore size for analysis.

[00068] Ή NMR measurements of the acceptors was recorded on Spinsolve carbon (Magritek, SPA409) bench top NMR at 42.5 MHz at room temperature. X-ray diffraction (XRD) patterns of the acceptor ground with graphite were recorded on a Rigaku RAD-3C diffractometer (35 kV, 20 m A, Japan) with CuKa radiation (k = 1.548 A) at a scan rate of 2/min for 1 hr with angles ranging from 10 to 70 degrees. Exfoliated graphite in organic solvents, polymers and monomers was characterized using UV-Vis spectrophotometry with a UV-2600, Shimadzu Spectrophotometer. Graphene suspensions with A1/A2 were diluted for UV measurements. The solution was transferred to a 5 mL UV quartz cell (0.01 m optical path) where measurements were done over the wavelength range of 200-800 nm. By measuring the absorbance of graphene dispersion at 660 nm, concentration of suspended graphene was obtained from Lambert-Beer law using extinction coefficient of graphene (a ₆₆ o=2460 mL/mg/m). Electrical conductivity measurements were done using a 4 point probe (Jandel, model RM 3000). Raman spectra were taken on a confocal Raman microscope (alpha 300R, WITec) with a 532.1 nm wavelength incident laser light. The morphology and size of graphene were measured using a field emission scanning electron microscope (FE-SEM, ZEISS SUPRA 40VP) at 3 kV under 35,000 x and 60,000x.

[00069] CT interaction requires close contact and can occur even in the absence of solvent. The acceptors were thus ground with graphite prior to solvent addition. Immediately, a smooth, dark paste was formed which could be easily used to write or draw marks on the glass vial (Fig. 2b). The friction of graphite during grinding was also found to be dependent on the amount of acceptor added to graphite. Lower acceptor concentrations resulted in higher friction and vice versa.

[00070] XRD patterns of the reflection peaks of the samples of graphite ground with Al and A2 (markings on glass vial in Fig. 2) are shown in (Fig. 3b and c). The intensity of the ground sample with Al and A2 was significantly decreased compared to the graphite sample. This is evidence of solid phase delamination of graphite into a few layers of graphite (<10 layers).

[00071 ] Since NMP and DMF have been shown to be good solvents for graphite exfoliation due to their matched surface energies, these solvents were added to the sediment of graphite ground with acceptor. Chloroform, a poor solvent for graphite exfoliation due to poor matched surface energy, was also tested in order to afford a better insight into the effect of acceptor on the solvent-graphite surface energy. [00072] Immediately after addition of solvent to the ground acceptor-graphitic sediment, dark dispersions were formed even prior to sonication. The change in solubility after acceptor addition indicated that graphite became more stable in suspension upon addition of acceptor. In NMP for example, concentrations of 0.02 and 0.03 mg/mL were obtained without sonication even after 30 minutes centrifuge (3000 rpm) using Al and A2 respectively. This was over two times that without acceptor after 30 minutes of sonication. (Fig. 6). The yield of graphene produced in all solvents increased further after only 30 minutes sonication (visually shown in Fig. 4b,c and d by the darkened suspensions in comparison to sonication without addition of acceptors. The measured concentrations of graphene in the solvents are summarized in Figure 6) For NMP, suspension concentrations of 0.15 and 0.18 mg/mL in Al and A2 respectively were achieved compared to O.Olmg/mL without acceptor).

[00073] XRD of films from the centrifuged dispersions confirmed complete exfoliation of graphite into pristine single and few layered graphene through absence of XRD (002) diffraction peak in Al and almost diminished peak in A2 (Fig. 3 d and e).

[00074] These results also indicate that the acceptor is bound to graphite through CT complexation in the solid phase resulting in weakening of the π-π attraction between graphene sheets in graphite. In some areas the CT interaction is more dominant resulting in sheet separation and hence exfoliation even without an external force. In most areas however the CT interaction is less dominant but all the same present, resulting in a relatively weakened π-π interaction between the graphene sheets which is completely overcome during sonication. The results presented herein suggest the importance of using a combination of an external force with acceptor in aiding and increasing exfoliation yield. This process thus shows promise as a low cost industrial method for graphene production in organic solvents with minimal energy input.

The solvent also plays a significant role even though it may have a limited ability on its own to disperse and stabilize exfoliated graphene, with most sedimenting during centrifugation. Most importantly, solvents with a good matching surface energy such as NMP lead to higher concentrations of graphene after only acceptor addition. An interesting observation was made in the case of chloroform with relati vely poor matching surface energy. In this case graphene could be effectively exfoliated only after addition of acceptor and clear solutions were obtained without acceptor after centrifugation, suggesting that an external additive that can interact with graphite can result in exfoliation even with a poor solvent as is the case in water/surfactant graphene solutions.

[00075] In all samples the yield of graphene also increased with acceptor concentration up to a critical acceptor concentration, and thereafter the yield decreased (Fig. 5). However, there was a large difference in the critical concentration of Al compared to that of A2. A2 showed a critical weight concentration at 0.05 g whereas Al was at 0.075 g. Furthermore, even though the acceptor concentration was lower in A2 than Al, the concentration of graphene was much higher in A2 (0.14 mg/mL) compared to Al (0.13 mg/mL).

[00076] This difference in critical concentrations between Al and A2 may be explained from the association constant (K _a ) between the graphene and Al and A2. Previous studies have reported K _a for pyrene groups and Al (0.74 M ^"1 ) and A2 (4.23 M ^"1 ) using 1,2-dichloroethane. A linearity of the Benesi-Hildebrand plots and Job's plots implied the formation of a CT complex with 1 : 1 donor-acceptor complexation. Therefore, the use of acceptor Al with a lower association constant may be expected to require a larger acceptor concentration than A2 with a higher association constant. These results also suggest that a critical degree of association exists. Another important factor is acceptor solubility limitation at higher concentrations. In NMP for example, A2 has poor solubility and saturation is reached at lower weight concentration compared to Al, and thus higher weight concentrations could not be tested (Fig. 5).

[00077] Apart from solubility limitations at higher concentrations, the inventors hypothesize that during exfoliation there are three interactions to consider: graphene-acceptor, graphene- solvent and solvent-acceptor. At moderate concentrations graphene easily interacts with both acceptor and solvent resulting in exfoliation. At higher acceptor concentrations there are two possibilities. Solvent-acceptor interaction could be dominant to graphene-acceptor and graphene-solvent interaction resulting into decreased exfoliation and yield of the graphene. Alternatively graphene-acceptor interaction increases, resulting in a folded structure which may cause aggregation. Thus it is important to maintain the right balance between graphene-acceptor interaction and solvent-acceptor interaction, even though the surface energies may be matched to that of graphene, as is the case of NMP.

[00078] Stability tests of exfoliated graphite (initial graphite concentration - 5 mg/mL) were conducted with or without Al (0.075 mg) in a wide range of organic solvents, polymer and monomer. For this test, dark dispersions obtained after centrifuge (CF) were determined to be stable (good solvent) whereas clear solutions with graphitic sedimentation were labelled to be unstable ("US": poor solvent) (Table 1 ).

Solvent BP (°C) 5 _D (MPa) ^1/2 C(;(mg/mL) CG+ΛΙ (mg/mL)

Cyrene 203 18.8 0.05 0.17

Epoxy - - 0.03 0.15

NMP(N-methyl-2-pyrrolidone) 202 18 (15-21) 0.01 0.13

Chloroform 61 17.8 0.01 0.07

DMF(Dimethylformamide) 152 17.4 0.01 0.07

l,2DCB(l,2Dichlorobenzene) 180 19.2 0.05 0.07

MAA 161 - 0.02 0.07

Acetone 57 15.5 US 0.06

Anisole 154 - 0.01 0.05

DCM (Dichloromethane) 40 18.2 US 0.01

CH-ONE(Cyclohexanone)) 155 17.8 US 0.01

DMSO 189 18.4 US US

Ethanol 78 15.8 US us

Hexane 69 15.3 us us

Toluene 1 1 1 18.0 us us

Iso-propanol 83 15.8 us us

Graphene - 18(15-21) - -

Table 1 : Effect of acceptor (Al) on stability of exfoliated graphene in organic solvent, polymer and monomer. (BP = boiling point, 5D = solubility parameter, Co = concentration of graphene achieved in the solvent without the addition of the A 1 acceptor and QJ ₊ AI ⁼ concentration of graphene in the solvent after addition of A).

[00079] The concentrations of graphene dispersions in all organic solvents, epoxy resin and MAA in which the suspension was stable or slightly stable, with or without addition of acceptor (Al, A2) are shown in Figure 6.

[00080] It was concluded that the best solvent for graphite exfoliation of those tested was cyrene. In cyrene using A2, graphene had a concentration of 0.27 mg/mL compared to 0.18 mg/mL in NM P for the same concentration.

[00081 ] In the case of 1 ,2-DCB the solvent has two electron withdrawing chlorine groups which make 1 ,2-DCB act as an electron acceptor, which can therefore compete with Al. This may explain the high exfoliation before, and limited changes after, addition of Al . Furthermore many studies on graphene-composite formation have been limited to graphene oxide due to its good dispersibility in most organic solvents especially DMF, and NMP. The first step during this composite formation is to disperse graphene oxide in the organic solvent then solution process the dispersion into the polymer. While effective, the process is quite laborious and it would be advantageous to exfoliate directly into the polymer. In this study higher concentrations of graphene were achieved in both the epoxy resin and methacrylic acid after addition of acceptor.

[00082] The results presented herein confirm that strong interlayer graphene π-π interaction exists in all solvents. Therefore, an electron acceptor may be added to overcome and interrupt the π-π interaction and hence enable exfoliation, and stable dispersions after centrifuge.

[00083] It therefore appears that criteria for efficient exfoliation and stabilization of graphene are: a) the solvent should have a good or mediocre surface energy relative to that of graphene b) addition of an external effector e.g. electron acceptor is useful to interrupt the π-π interaction between the graphene layers in graphite.

Characterisation of graphene

[00084] UV/Vis spectra were measured in order to demonstrate that there was indeed non- covalent interaction of graphene with acceptor through CT complexation, resulting in doped graphene. The electron acceptor is therefore viewed as a dopant to the graphene. Charge transfer is determined by the relative position of the density of states (DOS) of the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO) of the dopant and Fermi level of graphene. If the LUMO is below the Fermi level then there is charge transfer from graphene to the dopant, i.e. electron acceptor. Therefore, molecules with electron withdrawing groups on the surface of graphene will lead to p-type doping of graphene. The strength of an electron acceptor is important for charge transfer. Weak electron acceptors such as fullerene (Ceo) do not result in significant charge transfer and would therefore not be efficient as a p-type dopant. The appearance of a charge transfer (CT) band at λ _ιηα χ 380 nm in the UV/Vis absorption spectrum provided evidence of non-covalent functionalization of graphene. (Fig. 7).

[00085] SEM analysis of all samples showed large, thin, few layered graphitic flakes when Al or A2 was added (Fig. 8 b and c respectively) compared to without acceptor (Fig. 8 a). Some flakes are also partially transparent to the electron beam suggesting them to be very thin. The flakes lie flat on top of one another with a lateral size from 6μτη to approximately 7 μπι. This may yield higher electrical conductivity than smaller sheets due to the lower proportion of edge defects. Large graphene sheets are highly desirable in improving the electrical and mechanical properties of composites.

[00086] Raman spectroscopy was used to study and characterize the quality of exfoliated graphite flakes. From the obtained spectra (Fig. 9) three vibrational bands typical of graphene were clearly obtained. D band at around 1350 cm ^"] ,G at 1580 cm ^"1 and 2D at 2700 cm ^"1 . The D band is indicative of the flake edge and basal plane degree of functionalization. Since the Raman excitation beam has a spot size comparable to that of the sizes of most graphene flakes, the beam can always see a large quantity of graphene flake edges. All D bands are much broader compared to the raw material graphite (Fig. 8b) due to increased flake edges as a result of exfoliation. The shape of the 2D band is indicative of the number of monolayers per flake. This band is slightly angled suggesting slight aggregation during film formation. Due to the high concentrations of the dispersions with added acceptor, thicker films were obtained compared to without acceptor. However this band shows that while some degree of aggregation is present, the films consist of a few layered graphene. For more than 5 layers the 2D band resembles that of graphite. None of the present spectra of exfoliated graphite bear significant resemblance to the raw material. Finally the G band is an in-plane vibrational mode of the sp ^" hybridized carbon atoms that make up graphene sheets. Therefore the G band position is very sensitive to chemical doping. In the acceptor exfoliated graphite the G band is slightly shifted to lower wavelengths, providing evidence of non-covalent doping of graphene.

[00087] In order to test the electrical properties of the films, the sheet resistance (Rs) of all the films was measured after diying the films overnight at 70 °C. Since thermoresponsivity was previously shown with polymers bearing pyrene groups when electron acceptors were added, DC conductivity tests with increasing temperature were conducted on films placed directly on a heated hot plate to see whether the graphene-acceptor interaction was thermoresponsive and reversible. The DC conductivity was then calculated from (Rs) and the film thickness (f) obtained from SEM, as shown below. High temperature conductivity measurements were conducted on films placed directly on a heated hot plate.

Resistivity (ohms cm) = Rs (ohms square) x t (cm) Conductivity S/cm = 1 / Resistivity (ohm cm)

Conductivity S/m = 1 /Resistivity (ohm cm) x 100

[00088] DC conductivities for NMP exfoliated graphite films are shown in Table 2 below. For graphite films exfoliated in chloroform, the results are shown in Tables 3 and 4.

Rs Resistivity Conductivity Conductivity

(ohms (ohms S/cm S/m square) cm)

NMP only exf. G. (BH,

NMP only exf.

G.(AH,400 "C) ^{0 00005 27} ' ⁶⁶⁰

1.383 0.723 72

Al in NMP exf. G. (BH,

_, , 0.00005 343

Rtp)

0.01717 58.24 5,824

A, , NMP e _X , C ₍₂ 00 ^ ^ ^ ^ ^

An _{n N} MP e, _f . G. ,AC, ^ ^ ^ ^

A2 in NMP exf. G. (BH,

, 0.00006 133.32

Rtp)

0.008 125.01 12,501

A2 in NMP exf. G. (200

o 0.00006 90.96

0.0055 183.23 18,323

A2 in NMP exf. G. (AC,

n , 0.00006 109.08

Rtp)

0.0065 152.79 15,279

Table 2. Electrical properties with increase in temperature of graphite films exfoliated in NMP (0.05g A1/A2, 0.1 Graphite, 15 mL NMP, 30 minutes sonication, 1 minute heating on hot plate then cooled; AC = after cooling, BH = before heating) t (cm) Rs Resistivity Conductivity Conductivity (ohms (ohms S/cm S/m

square) cm)

0.00005 1944

Rtp)

0.0972 10.29 1029

CHCI ₃ only exf. G.

0.00005 1574

(150 °C))

0.0787 12.71 1271

CHCI ₃ only exf. G. (AC,

0.00005 1771

Rtp)

0.0886 1 1 .29 1 129

0.00005 392.95

(BH, Rtp)

0.0196 50.89 5089 0.0153 65.26 6526

Al in CHCI ₃ exf. G.

0.00005 386.77

(AC, Rtp) 0.0193 51.71 5171

A2 in CHCI ₃ exf. G.

0.00005 45.04

(BH, Rtp)

0.0023 444.04 44,405

A2 in CHCI ₃ exf. G.

0.00005 28.51

(150 °C)

0.0014 701.51 70,151

A2 in CHCI ₃ exf. G.

0.00005 38.16

(AC, Rtp)

0.00191 524.1 1 52,41 1

Table 3. Electrical properties with increase in temperature of films exfoliated in CHCI ₃ (0.075 Al/ 0.05g A2, 0.15 Graphite, 20 mL solvent, 15 minutes sonication, 30 minute heating on hot plate; AC = after cooling, BH = before heating)

Rs Resistivity Conductivity Conductivity

(ohms (ohms S/cm S/m square) cm)

A l in CHCl

0.00005 3254

(BH, Rtp)

0.1627 6.15 615

Al in CHCI ₃ exf. G. _

(150 °C) ^{0 00005 20%} 0.1048 9.54 954

Al in CHCl exf. G

0.00005 2120

A2 in CHCI ₃ exf. G.

0.00005 1446

^οπ, ινψ, _{13 g3}

A2 in CHC1 ₃ exf. G. 0.00005 986.32 n 05 2028 (200 °C)

A2 in CHC1 ₃ exf. G.

. _m . 0.00005

(AC, Rtp)

0.06 16.79 1679

Table 4. Electrical properties with increase in temperature of films exfoliated in CHCI ₃ (0.05g A 1/A2, 0.1 Graphite, 15 mL solvent, 30 minutes sonication, 30 minute heating on hot plate; AC = after cooling, BH = before heating)

[00089] For all acceptor exfoliated graphene films, the conductivity was higher than without acceptor for both NMP and chlorofonn systems. This is considered to be due to larger graphene sheets (evidenced by SEM) in acceptor exfoliated graphite samples and therefore more surface area and sheet overlap for electrical contact. However, the films in A2 exfoliated graphite in NMP had lower conductivity of 1 1,794 S/m compared to 71 ,000 S/m in chloroform. It is difficult to completely remove NMP even after annealing even at 1000 °C, and this may lead to lower achievable conductivities. The films of graphite exfoliated in chloroform had effectively no residual solvent due to the low boiling point of chlorofonn, and therefore had higher conductivity.

[00090] The acceptor exfoliated films showed an increased conductivity at higher temperatures. This is thought to be because the interaction between the acceptors and graphene is non- covalent, thus higher temperatures are expected to result in cleavage of this interaction. All films were observed to revert to nearly their original, but slightly higher, conductivity on cooling. This may be due to the loss of solvent and/or to re-interaction of acceptors with graphene. This reversibility is very important, especially in sensor and energy storage applications. The conductivity, and hence the electronic properties, of graphene is also dependent on the acceptor concentration (Tables 3 and 4) and thus by varying the amount of acceptor one may be able to switch from graphene insulator to semiconductor, which is also desirable for many applications.

[00091 ] The inventors have demonstrated high yield, low energy exfoliation of graphite into pristine non-covalently doped graphene using CT interaction. The graphene exfoliated in the presence of an electron acceptor has improved properties compared to their undoped

counterparts.

Example 2: dispersion of carbon nanotubes [00092] This experiment relates to dispersion of MWCNTs in organic solvents (Cyrene, DMF, NMP and chloroform) stabilized by a non-covalent electron donor (MWCNT) - electron deficient acceptor (Al or A2) interaction with low power, short sonication times. The M WCNTs dispersions in NMP for instance (after only 30 minutes sonication) were stable even after centrifuging (2.0 mg/mL in Al, 0.20 mg/mL in A2 relative to <0.01 mg/mL without acceptor). The acceptor is thought to perform a dual role in that it not only reduces attractive π-π interaction between the individual MWCNTs but stabilizes the dispersed MWCNTs resulting in stable, high quality dispersions.

[00093] Carbon nanotubes (CNTs), like graphene, have attracted considerable attention due to their extraordinary physical and chemical properties. Example 1 , above, disclosed the successful exfoliation of graphite in organic solvents stabilized by a non-covalent electron donor (graphene) - electron deficient acceptor (Al or A2) interaction with low power, short sonication times required. The yield of graphene was substantially improved (almost twentyfold) in most organic solvents relative to that without acceptor.

[00094] This experiment extends a similar strategy to that used in exfoliation of graphene to realize increased dispersibility of CNTS in organic solvents. In this experiment the inventors explore the introduction of an external electron deficient acceptor (A1/A2) with mild sonication to induce deaggregation and solubility in CNTS due to donor-acceptor (CT complex) interaction which serves a dual role in dispersing and stabilizing the CNTs in a similar manner as for graphite/graphene (see Fig. 1).

Dispersion of CNTs

[00095] MWCNTs were purchased from Carbon Allotropes, Australia. NMP (N-methyl-2- pyrrolidone), Cyrene, DMF and Chloroform was used as received. Al and A2 were synthesized as described in Example 1 . In a typical experiment, 50 mg of MWCNTs (5mg/mL) was ground with Al or A2 (0.1 mg, 0.075 mg respectively) followed by addition of 10 mL of organic solvent. In the case of DMF due to poor acceptor solubility the weight of Al and A2 used was 0.05 and 0.01 g respectively)

[00096] Identical suspensions were prepared in solvent but without Al or A2, as a standard. The suspensions were left for at least one week to enable CT interaction and were then sonicated for 30 minutes using a low power 50Hz bath sonicator. The suspension was centrifuged at 3000 rpm for 30 minutes to sediment non-dispersed CNTs and the top half was pippeted off to provide a stock CNT suspension. The concentrations of the suspensions were characterized using UV-Vis spectrophotometry with a UV-2600 Shimadzu Spectrophotometer. CNT suspensions using Al or A2 were diluted for UV measurements: 0.5 mL of graphene suspension diluted with solvent to l OOmL and lOmL respectively. The solution was transferred to a 5 mL UV quartz cell where measurements were conducted over the wavelength range of 200-800 nm. The absorbance at 660 nm was used as a measure of concentration of suspended CNTs from Lambert-Beer law, using the extinction coefficient of graphene (a=2460 mL/mg/m).

[00097] In NMP the addition of Al or A2 resulted in stable MWCNTs dispersions (Fig. 1 1 ). As observed with graphene (Example 1), the interaction between the MWCNTs appeared to commence in the solid phase. The concentration achieved for M WCNTs in NMP with 2.0 and 0.2 mg/mL in Al and A2 respectively, compared to no stabilized dispersion after centrifuge in the sample where the acceptor was not used are shown in Fig. 10, together with data for dispersion in DMF, cyrene and chloroform.

Industrial applicability

[00098] Graphene has many uses. The presently described technology, which enables production of highly concentrated dispersions of pristine graphene sheets with large lateral dimension, enables a number of important applications. The most promising solvents such as NMP are used extensively in the production of anodes for Li ion batteries. Other solvents allow incorporation into a range of application areas, including the use of graphene in mechanical reinforcement applications, electrical and thermal coatings. Other applications include as a novel solid lubricant and in semi-conductor applications.