Improvements Relating To Graphene Nanomaterials

Abstract

A process for preparing a graphene nanomaterial product, the process comprising: cavitating a liquid medium comprising a diaromatic hydrocarbon component to synthesise from the diaromatic hydrocarbon component a dispersion of graphene nanomaterial in the liquid medium; and obtaining a graphene nanomaterial product from the dispersion.

Claims

1. A process for preparing a graphene nanomaterial product, the process comprising: cavitating a liquid medium comprising a diaromatic component to synthesise graphene nanomaterial from the diaromatic component and form a dispersion of the graphene nanomaterial in the liquid medium; and obtaining a graphene nanomaterial product from the dispersion.

2. The process of claim 1, wherein the nanomaterial product comprises at least 10 ppm graphene nanomaterial synthesised from the diaromatic component.

3. The process of claim 1 or claim 2, wherein the nanomaterial product consists of graphene nanomaterial synthesised from the diaromatic component.

4. The process of any preceding claim, wherein the nanomaterial product comprises graphene quantum dots, graphene nanoflakes, graphene nanoribbons, graphene nanosheets, or combinations thereof.

5. The process of any preceding claim, wherein the nanomaterial product comprises one or more heteroatom impurities.

6. The process of any preceding claim, wherein the graphene nanomaterial product is oxidised.

7. The process of any preceding claim, wherein the graphene nanomaterial product is functionalised.

8. The process of any preceding claim, wherein the diaromatic component comprises optionally substituted fused or linked diaromatic hydrocarbons or heterocycles.

9. The process of any preceding claim, wherein the diaromatic component is a diaromatic hydrocarbon component consisting of one or more optionally substituted diaromatic hydrocarbons.

10. The process of any preceding claim, wherein the diaromatic component comprises one or more compounds of Formula A or Formula B, or heterocyclic variants thereof, optionally substituted with one or more moieties at one or more of the numbered positions: ##STR00006##

11. The process of claim 10, wherein the one or more moieties are selected from alkyl, alkenyl or alkynyl substituents, and halides.

12. The process of claim 10 or claim 11, wherein the one or more moieties are selected from methyl, ethyl, and halides.

13. The process of any preceding claim, wherein the diaromatic component comprises methylnaphthalene or ethylnaphtalene, and optionally naphthalene.

14. The process of any preceding claim, wherein the liquid medium comprises a stabilising component for stabilising the dispersion of graphene nanomaterial to be formed.

15. The process of claim 14, wherein the stabilising component comprises a solvent which on addition to the liquid medium, is capable of reducing the distance in Hansen space between the predicted Hansen Solubility Parameters of the liquid medium (HSPs: .sub.Ds, .sub.Ps and .sub.Hs) and of graphene, graphene oxide, or reduced graphene oxide (HSPs: .sub.Dg, .sub.Pg and .sub.Hg), such that it reduces R, where R.sup.2=4 (.sub.Dg.sub.Ds).sup.2+(.sub.Pg.sub.Ps).sup.2+(.sub.Hg.sub.Hs).sup.2.

16. The process of claim 14 or claim 15, wherein the stabilising component comprises N-Methyl-2-pyrrolidone (NMP).

17. The process of any preceding claim, wherein the liquid medium consists of the diaromatic component and a balancing amount of stabilising component, in particular NMP.

18. The process of any preceding claim, wherein the liquid medium comprises a functionalising component and a functionalised graphene nanomaterial is synthesised.

19. The process of claim 18, wherein the functionalising component is selected to enhance dispersion stability of the graphene nanomaterial.

20. The process of claim 18 or claim 19, wherein the functionalising component comprises one or more compounds capable of taking part in a nucleophilic substitution, electrophilic substitution, condensation reaction or addition reaction.

21. The process of any one of claims 18 to 20, wherein the functionalising component comprises or consists of molecules comprising an amine group, molecules comprising a hydroxyl or peroxide group, molecules comprising a carboxylic acid group, or combinations thereof.

22. The process of any one of claims 18 to 21, wherein the functionalising component comprises aromatic molecules, optionally substituted with one or more groups as defined in claim 21.

23. The process of any one of claims 18 to 22 wherein the liquid medium comprises an emulsion of the functionalising component in the diaromatic component, the emulsion optionally being kinetically stable or thermodynamically stable.

24. The process of any one of claims 18 to 23 comprising cavitating the liquid medium in the presence of oxygen or another heteroatom impurity.

25. The process of any preceding claim, wherein the liquid medium has an initial boiling point in the range of from 180 to 300 C.

26. The process of any preceding claim comprising treating the liquid medium, prior to cavitation, to reduce air or oxygen dissolved therein.

27. The process of any preceding claim wherein cavitation of the liquid is effected by subjecting the liquid medium to ultrasound.

28. The process of claim 26 wherein the ultrasound has a frequency in the range of from 20 to 100 kHz.

29. The process of any preceding claim, wherein cavitation is carried out under an inert atmosphere.

30. The process of any preceding claim, comprising stabilising the dispersion and obtaining the graphene nanomaterial product from the stabilised dispersion.

31. The process of claim 30, wherein stabilising the dispersion comprises forming a colloid of the dispersed graphene nanomaterial in the liquid medium, such that particles of the dispersed graphene nanomaterial do not settle.

32. The process of any preceding claim comprising adding a stabilising component to the liquid medium during and/or after cavitation of the liquid medium.

33. The process of claim 32 wherein the added stabilising component comprises a solvent which on addition to the liquid medium, is capable of reducing the distance in Hansen space between the predicted Hansen Solubility Parameters of the liquid medium (HSPs: .sub.Ds, .sub.Ps and .sub.Hs) and of graphene, graphene oxide or reduced graphene oxide (HSPs: .sub.Dg, .sub.Pg and .sub.Hg), such that it reduces R, where R.sup.2=4 (.sub.Dg.sub.Ds).sup.2+(.sub.Pg.sub.Ps).sup.2+(.sub.Hg.sub.Hs).sup.2.

34. The process of claim 33, wherein the added stabilising component comprises N-Methyl-2-pyrrolidone (NMP).

35. The process of any preceding claim wherein the dispersion of graphene nanomaterial is stabilised by a stabilising component and constitutes the graphene nanomaterial product.

36. The process of any preceding claim, wherein the graphene nanomaterial product comprises a subset of the dispersion and is obtained by one or more separation steps.

37. The process of any preceding claim, wherein turbostratic particles comprising graphene layers are separated from the dispersion to provide a purified dispersion constituting the graphene nanomaterial product.

38. The process of any preceding claim comprising allowing the graphene nanomaterial in the dispersion to agglomerate and settle out of the dispersion to form turbostratic particles, re-dispersing graphene nanomaterial from the turbostratic particles to form a re-dispersion, and obtaining the graphene nanomaterial product from the re-dispersion.

39. The process of claim 38, wherein the re-dispersion of graphene is stabilised by a stabilising component and constitutes the graphene nanomaterial product.

40. The process of claim 38 or claim 39, wherein re-dispersing the graphene nanomaterial comprises exposing the turbostratic particles to a re-dispersing liquid comprising a stabilising component, and optionally agitating the particles.

41. A graphene nanomaterial product obtainable by a process according to any preceding claim.

42. A graphene nanomaterial product comprising a graphene nanomaterial dispersed in a liquid medium comprising a diaromatic hydrocarbon or heterocycle and a stabilising component.

43. The graphene nanomaterial product of claim 41 or claim 42 comprising a graphene nanomaterial dispersed in a liquid medium comprising 1-methylnaphthalene and NMP.

Description

DETAILED DESCRIPTION

[0112] The present invention will now be further described with reference to the following non-limiting examples and the accompanying drawings.

Experimental Methods

[0113] In a typical experiment 50 mL of a liquid medium was sonicated, during which a darkening occurred due to the formation of a dispersion of particles in the liquid medium.

Sonication

[0114] Sonication was carried out with a VCX 750 (750 W) ultra-sonic processor (ex Sonics Materials Inc.) and a 13 mm extender horn, which delivered ultrasound to a 50 mL sample of hydrocarbon contained within a jacketed glass beaker. Cold water (10 C.) was passed through the jacket to keep the liquid hydrocarbon below its flash point. A PTFE lid was used to prevent splashing whilst compressed air or nitrogen was blown over the surface of the hydrocarbon to inhibit condensation inside the reaction vessel. The whole apparatus was housed inside a box to reduce acoustic noise.

[0115] Ultrasound was produced at a frequency of 20 kHz and when the amplitude of the processor was set to 65% the delivery of power to the ultrasound probe was 72 W. The probe is made of titanium alloy (Ti 6Al-4V) and consists of 90% titanium, 6% aluminium and 4% vanadium. This material is susceptible to cavitation erosion and becomes tarnished during use. The probe was polished on silicon carbide papers (P400 and P1000) between each experiment to maintain a smooth and shiny tip surface.

Chemicals

[0116] Ex Acros Organics: Activated charcoal (SA 2 decolounsing).

[0117] Ex Alfa Aesar: Biphenyl (99% purity); 1-Methylnaphthalene (96% purity), N-Methyl-2-pyrrolidone (HPLC grade); Naphthalene (99+% purity).

[0118] Ex Sigma Aldrich: Aniline (99.5% purity); Benzoic acid (99.5% purity); Dodecylamine (98% purity); Formic acid (95% purity); Hydrogen peroxide solution (30 wt %); 1-Methylnaphthalene (95% purity); Octylamine (99% purity); Phenol (99% purity); Quinoline (97% purity), Toluene (anhydrous, 99.8% purity); Tri(ethylene glycol) monoethyl ether (technical grade).

[0119] Ex VWR Chemicals: Acetic acid glacial; n-Heptane (HiPerSolv chromanorm for HPLC, filtered at 0.2 m).

[0120] Abbreviations

[0121] A, Absorbance; l, length; 1-MN, 1-methylnaphthalene; NMP, N-methyl-2-pyrrolidone; NAP, naphthalene; N, particle number density.

UV-Vis Spectrophotometry

[0122] UV-vis spectrophotometry was carried out using a UV-1699PC VWR Spectrophotometer (VWR International, Radnor, Pa.).

Laser Particle Counter

[0123] Particle analysis was carried out using a Spectrex LPC-2200 laser particle counter (Spectrex Corporation, Redwood City, Calif.), which makes measurements based on the principle of near-angle light scattering. A revolving laser beam is passed through the walls of a glass container; any particles present in the sample cause the beam to scatter. The extent of scattering is proportional to the number and size of the particles, which are reliably counted in the 1-100 m size range. Samples were gently swirled before being left to stand to allow any air bubbles to settle. Number density counts were based on an average of ten consecutive measurements.

[0124] When the number concentration (N) is higher than 1000 cm.sup.3 there is a risk of overlap between particles in the third dimension (i.e. closer to or further from the detector), which may lead to two or more small particles being counted as a single large particle. Sample dilution with chromatography grade n-heptane, which has a low background particle count (<20 cm.sup.3), prevents this artefact from occurring.

[0125] 1-MN and NMP were pre-filtered (1.0 m PTFE Acrodisc membrane ex Sigma Aldrich) before irradiation with ultrasound. The background count for all solvents was <50 cm.sup.3.

Transmission Electron Microscopy (TEM)

[0126] Dispersions were filtered onto a holey carbon film 300 mesh copper TEM grid. Imaging was performed in transmission mode using a JEOL 2100 TEM (JEOL Ltd, Tokyo) at 200 kV or 160 kV beam voltage.

X-Ray Photoelectron Spectroscopy (XPS)

[0127] XPS was carried out using a purpose-built ultra-high vacuum system equipped with a Specs PHOIBOS 150 electron energy analyser and Specs FOCUS 500 monochromated Al Ka X-ray source (Specs GmbH, Berlin, Germany). Samples were prepared for analysis by filtering dispersions through unsupported alumina membranes with a 0.2 m pore size followed by washing with iso-propanol. The filters were cut to size (>10 mm10 mm) and attached to standard Ni XPS sample holders using conducting double-sided vacuum-compatible adhesive pads. Survey and narrow scans were acquired over the binding energy range between 0 and 1100 eV using a pass energy of 50 eV and high resolution scans were made over individual photoelectron lines using a pass energy of 15 eV. Data processing and curve fitting were carried out using CasaXPS software v2.3.16, with quantification carried out using Scofield cross-sections corrected for the energy dependence of the analyser transmission and the effective electron attenuation lengths.

Raman Spectroscopy

[0128] Renishaw InVia Raman Microscope (Renishaw plc., Wolton-upon-Edge, UK) with a 532 nm laser and 2400 l/mm (vis) grating.

Example 1Gravimetric Analysis (Samples A-G)

[0129] For each of samples A to G set out in Table 1 below, 50 mL of liquid medium consisting of a diaromatic component was sonicated to a pre-set level of energy measured with the in-built meter on the ultrasonic processor. Darkening occurred due to the formation of a dispersion of black particles. The colloid was not stable and flocculation followed by sedimentation was visible after 1-2 days.

[0130] Gravimetric analysis of the sediment material was achieved by standing a sample for ca. 1-2 weeks before centrifugation (3500 rmin.sup.1 for 20 minutes) and filtering (0.7 m Whatman glass microfiber filter, Grade GF/F). The sediment was washed with n-heptane to remove residual high boiling hydrocarbons.

[0131] The filtrate mixture of 1-MN (and NAP where included) and n-heptane was left to stand, during which time a second, brown sediment formed. Filtering, washing and weighing in the same manner as the first sediment was performed for gravimetric analysis.

[0132] FIG. 1 shows the first (A.1-G1) and second (A.2-G.2) sediment after filtration. Both sediments increased as a function of exposure time to ultrasound. The black sediment (A.1-G.1) was always produced in a higher yield than the brown sediment (A.2-G.2), which showed a darkened colour with increasing mass.

[0133] Calorimetric measurements were used to determine that the transfer of ultrasonic power to the sonochemical reaction was 39 W at a transfer efficiency of 53%. The power intensity of the 13 mm extender probe (diameter 1.27 cm) was therefore 30 Wcm.sup.2, which is sufficient to produce transient cavitation bubbles. FIG. 2 shows that total sediment mass increased linearly with the level of ultrasound energy (powertime) applied.

[0134] For samples E and F, white crystals of NAP were dissolved in 1-MN up to a volume fraction of 0.2: at higher levels complete solubility becomes an issue unless special measures are taken to aid dissolution. Introduction of the non-alkylated diaromatic (NAP) did not have a significant impact on the level of sediment that was produced (Table 1 and FIG. 2).

TABLE-US-00001 TABLE 1 Sedimentation experiments. First Sediment Second Sediment Liquid Standing Standing Sample medium Energy.sup.a time Mass time Mass i.d. .sub.1-MN:.sub.NAP (kJ) (days) (mg) (days) (mg) A 1.0:0.0 86 52 2.9 23 1.2 B 1.0:0.0 200 14 4.8 9 2.0 C 1.0:0.0 300 13 7.7 9 1.2 D 1.0:0.0 300 14 7.0 8 2.1 E 0.9:0.1 300 14 7.5 8 1.9 F 0.8:0.2 300 7 6.2 10 1.9 G 1.0:0.0 600 7 12.4 10 5.4 .sup.ameasured on the in-built meter of the ultrasonic processor

Example 2Colloid Stabilisation (Samples H-J)

[0135] The sonication of diaromatic hydrocarbons (samples H to J detailed in Table 2 below) produced dispersions that showed an absorbance across all wavelengths of visible light. FIG. 3 shows transmission spectra of diaromatic components before and after treatment with ultrasound.

[0136] The diaromatic hydrocarbons used as starting materials are relatively transparent in the 400-1000 nm range. In FIG. 3, 1-MN/NAP represents a diaromatic hydrocarbon component with .sub.1-MN=0.8, .sub.NAP=0.2 by volume. 1-MN/NAP/NMP represents a diaromatic hydrocarbon component with .sub.1-MN=0.4, .sub.NAP=0.1, .sub.NMP=0.5 by volume.

[0137] In FIG. 3, H, I and J are as set out in Table 2 below. Absorbance at 660 nm was used as an indicator of nanosheet concentration (see also Table 2)..sup.1

[0138] After sonication, each sample (H.sub.1-MN, I.sub.1-MN and J.sub.1-MN) had a high count of 1 m particles (N>110.sup.6 cm.sup.3), which increased with ultrasound exposure time. The irradiated samples were split in two (225 mL). NMP was added to one of each of the sub-samples (=1:1) and the resulting mixtures (H.sub.1-MN/NMP, I.sub.1-MN/NMP and J.sub.1-MN/NMP) treated with ultrasound for a further 10 minutes. This decreased N (0.2-0.310.sup.6 cm.sup.3) as the 1 m particles were reduced to a size below the detection limit of the counter (<1 m). These dispersions were stable and did not undergo flocculation or sedimentation.

[0139] FIG. 4 shows particle number density of sonicated diaromatic mixtures, before (H.sub.1-MN, I.sub.1-MN& J.sub.1-MN) and after (H.sub.1-MN/NMP, I.sub.1-MN/NMP & J.sub.1-MN/NMP) addition of NMP. FIG. 5 shows the change in particle number density as a function of ultrasound exposure time.

[0140] The measurable particles remaining in the stabilised dispersions had diameters of 1-12 m. FIG. 6 shows the size distribution of aggregate particles remaining in stabilised colloids.

TABLE-US-00002 TABLE 2 Summary of colloid stabilisation experiments. Sedi- NMP Absorb- men- Liquid added ance/ N tation Sample medium Energy.sup.a post- length (10.sup.6 after id. .sub.1-MN:.sub.NAP (kJ) sonication (m.sup.1) cm.sup.3 standing H.sub.1-MN 0.9:0.1 100 No 55 1.39 Yes H.sub.1-MN/NMP Yes 28 0.21 No I.sub.1-MN 0.8:0.2 300 No 173 3.06 Yes I.sub.1-MN/NMP Yes 81 0.33 No J.sub.1-MN 0.8:0.2 600 No 325 5.28 Yes J.sub.1-MN/NMP Yes 156 0.30 No .sup.ameasured on the in-built meter of the ultrasonic processor

[0141] Extending sonication times beyond 10 minutes or centrifugation at 1500 rmin.sup.1 did not remove this aggregate material (Table 3).

TABLE-US-00003 TABLE 3 Impact of sonication and centrifugation on the aggregate particle number density of sample J.sub.1-MN/NMP. Ultrasound Centrifugation @ 20 kHz @ 1500 r min.sup.1 N (min) (min) (10.sup.6 cm.sup.3) 10 0 0.30 0.03 20 0 0.30 0.04 30 0 0.29 0.04 10 20 0.30 0.03 10 80 0.28 0.04

[0142] Filtration of the colloids through different grades of Whatman glass microfiber filters produced some reduction in particle count but with relatively low filter efficiencies: 2% with GF/D (pore size 2.7 m), 30% with GF/A (pore size 1.6 m) and 36% with GF/F (pore size 0.7 m).

Example 3Sedimentation (Samples H-J)

[0143] The particle number density in a freshly sonicated (300 kJ) sample of 1-MN was initially relatively low (N=0.310.sup.6 cm.sup.3), however upon standing thermally driven collisions took place between small particles (<1 m) leading to an increase in the count of 1 m particles. FIG. 7 shows the increase of 1 m particles due to thermally driven collisions of sub-micron particles in the sonicated sample. An upper limit to particle number density was established over a period of ca. 6 hours (FIG. 7). N=2.3510.sup.6 cm.sup.3 after 10 days. Samples H-J were shaken before pipetting into UV-Vis quartz cuvettes and measuring the absorbance at 660 nm over 14 days. FIG. 8 shows the absorbance at 660 nm of samples H-J during 14 days of standing.

[0144] Each of the unstabilised dispersions (H.sub.1-MN, I.sub.1-MN & J.sub.1-MN) showed a drop in absorbance as flocculation followed by sedimentation occurred. The rate of change in absorbance increased as a function of the initial concentration of material present in the dispersion. There is also evidence that the rate of this process increased during the first two days of standing. FIG. 9 shows the increase in the rate of sedimentation during the first 2 days of standing. This is consistent with the growth of small particles by perikinetic agglomeration followed by flocculation, leading to even faster rates of sedimentation. In contrast, the absorbance at 660 nm of each of the stabilized dispersions (H.sub.1-MN/NMP, I.sub.1-MN/NMP & J.sub.1-MN/NMP) remained relatively stable over the 14 days (FIG. 8).

Example 4Rate of Particle Formation (Samples K-T)

[0145] 1-MN (ex Alfa Aesar) contains heteroatom species (typically ca 0.6 wt. % sulfur), which are responsible for at least some of the yellow colouration of the hydrocarbon. Some purification was achieved by adding activated charcoal (20 g) to 1-MN (200 mL) and stirring for 7 hours. Sonication of mixtures K-O (Table 4) was performed whilst samples were taken at 100 kJ intervals to measure the change in absorbance with time.

[0146] FIG. 10 shows the UV-visible spectra of 1-MN before and after treatment with activated charcoal. FIG. 11 shows the rate of formation of particles (absorbancelength.sup.1time.sup.1) as monitored by absorbance of 660 nm light.

[0147] The formation rate decreased slightly (2-3%) after 1-MN was treated with activated charcoal (K v. L and M v. N). The addition of NAP produced a 9-10% (K v. M and L v. N) increase in formation rate. The introduction of NMP during sonication resulted in a stable colloid with a low particle number density (N=0.1410.sup.6 cm.sup.3) and a rate of formation that was reduced by a factor of ten (O v. M).

TABLE-US-00004 TABLE 4 Experiments to assess rates of formation. Sample Liquid medium Energy.sup.a 1-MN pre-treated A l.sup.1 t.sup.1 Sedimentation i.d. .sub.1-MN:.sub.NAP:.sub.NMP (kJ) with activated C (m.sup.1 min.sup.1) after standing K 1.0:0.0:0.0 600 No 2.27 Yes L 1.0:0.0:0.0 600 Yes 2.21 Yes M 0.8:0.2:0.0 600 No 2.49 Yes N 0.8:0.2:0.0 600 Yes 2.43 Yes O 0.4:0.1:0.5 600 No 0.25 No .sup.ameasured on the in-built meter of the ultrasonic processor

[0148] The variation in the rate of formation of particles from three different batches of 1-MN (ex Alfa Aesar) is given in Table 5. White crystals of biphenyl can be dissolved in 1-MN up to a volume fraction of 0.35. FIG. 12 shows that this results in an 8% increase in the formation rate. Combining quinoline (a liquid nitrogen-containing diaromatic) with 1-MN at a volume fraction of 0.5 produces an even greater (38%) increase in formation rate. The cavitation of neat quinoline has a reaction rate over twice that of the rate of particle formation of 1-MN. Data in FIG. 12 and Table 5 also show that when toluene (a mono aromatic) is cavitated a very low rate of reaction is observed.

TABLE-US-00005 TABLE 5 Further experiments to assess rates of formation. Sample Energy.sup.a A l.sup.1t.sup.1 i.d. Liquid medium (kJ) (m.sup.1 min.sup.1) P(1) 1-MN (batch 1) 600 2.16 P(2) 1-MN (batch 2) 600 2.10 P(3) 1-MN (batch 3) 600 2.09 Q Quinoline 300 4.54 R 1-MN & Quinoline 400 2.93 (.sub.1-MN:.sub.Quinoline = 0.5:0.5) S 1-MN & Biphenyl 600 2.29 (.sub.1-MN:.sub.Biphenyl = 0.65:0.35) T Toluene 400 0.14 .sup.ameasured on the in-built meter of the ultrasonic processor

Example 5Impact of Nitrogen Sparging (Samples U-W)

[0149] The conditions used to produce sample N (.sub.1-MN=0.8, .sub.NAP=0.2) were also employed for a sample U, with the exception that nitrogen gas was bubbled into the hydrocarbon through a fine needle for 10 minutes before commencing sonication. The ultrasound treatment (600 kJ) was then carried out under a flowing atmosphere of nitrogen. Sample U was left to stand without addition of NMP and after a week no visible flocculation or sedimentation had occurred. The number density of 1 m particles was lower (N=1.2210.sup.6 cm.sup.3) than previously observed for samples exposed to ultrasound for the same length of time (5.2810.sup.6 cm.sup.3, Table 3). Centrifugation at 3500 r min.sup.1 for 60 minutes resulted in a small amount of black sediment (U.1=1.5 mg). The supernatant remained as a stable colloid until n-heptane (50 mL) was added, which produced sedimentation of black material (U.2=10.7 mg).

[0150] Sonication of 1-MN (300 kJ) was also conducted whilst continuously bubbling gas (air or N.sub.2) into the hydrocarbon during irradiation with ultrasound (V and W). Sedimentation after sonication was observed when air was bubbled; this black material was collected by centrifugation and filtration (V.1=6.2 mg). The addition of n-heptane to the supernatant produced brown sediment (V.2=3.0 mg).

[0151] When N.sub.2 was continuously bubbled sedimentation did not occur, although some black material was isolated by centrifugation (W.1=2.7 mg). Addition of n-heptane produced black sediment (W.2=2.4 mg). These results are summarized in FIG. 13. This shows the change in sedimentation mass when employing a pre-treatment or continuous N.sub.2 or air sparge (U.1, V.1 and W.1first stage sediment after sonication, U.2, V.2 and W.2second stage sediment after n-heptane addition).

Example 6Characterisation of Material Formed During Cavitation (Samples X-Z)

[0152] Samples X, Y and Z from the cavitation of 1-MN/NAP (.sub.1-MN=0.8, .sub.NAP=0.2) were prepared for analysis by TEM, XPS and Raman spectroscopy as detailed in Table 6. The dispersions produced under air (X and Y) were not stable and after standing for seven days underwent centrifugation at 3500 r min.sup.1 for 20 minutes. The brown supernatant (X.2.sub.1-MN and Y.2.sub.1-MN) was removed by pipette and the remaining black sediment added to NMP (50 mL), followed by a short period (10 kJ) of sonication to produce a black dispersion (X.1.sub.NMP and Y.1.sub.NMP). A third sample (Z) was prepared by bubbling N.sub.2 through the hydrocarbon prior to sonication and then flowing N.sub.2 to maintain an oxygen free atmosphere during irradiation. The resulting dispersion was combined with NMP (=1:1) and a further 10 minutes of ultrasound applied under a N.sub.2 atmosphere to obtain a stable dispersion.

TABLE-US-00006 TABLE 6 Summary of samples prepared for analysis. 1-MN Sedimen- Liquid pre-treated Air tation Samples for Sample medium Energy.sup.a with or after TEM, XPS i.d. .sub.1-MN:.sub.NAP (kJ) act. C N.sub.2 standing & Raman X 0.8:0.2 600 Yes Air Yes X.1.sub.NMP X.2.sub.1-MN Y 0.8:0.2 300 Yes Air Yes Y.1.sub.NMP Y.2.sub.1-MN Z 0.8:0.2 600 Yes N2 No Z.sub.1-MN/NMP .sup.ameasured on the in-built meter of the ultrasonic processor

[0153] FIG. 14 shows transmission spectra of a) X, b) Y, c) Z and particle number density measurements of d) X, e) Y and f) Z. (Abs. at 660 nm/l (m.sup.1): X.1.sub.NMP 308, X.2.sub.1-MN 34, Y.1.sub.NMP 150, Y.2.sub.1-MN 19, Z.sub.1-MN/NMP 144. N (10.sup.6 cm.sup.3): X.1.sub.NMP 0.27, X.2.sub.1-MN 0.24, Y.1.sub.NMP 0.12, Y.2.sub.1-MN 0.23, Z.sub.1-MN/NMP 0.30).

[0154] The transmission spectra and absorbance at 660 nm reflect the separation of black and brown materials in sample X and Y. All samples had particle counts 5 0.310.sup.6 cm.sup.3), consistent with stabilised colloids (c.f. H.sub.1-MN/NMP, I.sub.1-MN/NMP and J.sub.1-MN/NMP in Table 2).

[0155] TEM analysis was carried out on X.1.sub.NMP, Y.2.sub.1-MN and Z.sub.1-MN/NMP. FIG. 15 shows the TEM of the brown supernatant Y.2.sub.1-MN, which is a low concentration suspension comprising of material that is not finely dispersed but agglomerated into >1 m sized structures which are electron opaque, and therefore relatively thick.

[0156] FIG. 16a shows the TEM of the black colloid X.1.sub.NMP, which is comprised of thin sheets of variable dimensions (10-200 nm) that are seen scattered across the surface of the holey carbon grid. Occasional larger (200-1000 nm) dark particles, which appear to be made of multiple overlapping sheets, are also present (FIG. 16b). These are consistent with the residual aggregate material identified during analysis by laser particle counter.

[0157] FIG. 17 shows the TEM of Z.sub.1-MN/NMP, which contains small (<50 nm) irregular shaped particles, which appear to consist of clusters of primary circular structures with diameters of ca. 15 nm. The N.sub.2 sparge and atmosphere during sonication therefore results in some alteration of morphology and the formation of smaller particles.

[0158] Thin films were prepared by vacuum filtration of dispersions onto alumina membranes (0.2 m Whatman Anodisc inorganic unsupported filter) mounted on a fritted glass holder. Films were washed with iso-propanol (15 mL) and dried in an oven (60 C.) for two days. X.1N.sub.NMP produced a grey/black film in contrast to the film from X.2.sub.1-MN, which was yellow-brown. Sample Z.sub.1-MN/NMP had a dark brown colouration. FIG. 18 shows thin films on 0.2 m alumina filters from dispersions a) X.1.sub.NMP, b) X.2.sub.1-MN and c) Z.sub.1-MN/NMP.

[0159] XPS spectra showed the film composition was primarily carbon and oxygen. Aluminium was also present in significant levels (4-6 atom %) as a result of fractures in the thin films producing a paving of 50 m fragments with 10-20 m gaps. The data was corrected based on the assumption of clean Al.sub.2O.sub.3 in areas not covered by carbonaceous material. High levels of carbon (89-92 atom %) are inversely correlated with lower levels of oxygen (6-9 atom %).

[0160] Low to trace levels of nitrogen, phosphorus and sulphur were also present and in one case a trace level of sodium was detected (Table 7).

TABLE-US-00007 TABLE 7 Elemental composition (atom %) of thin films prepared from samples X and Z. Element X.1.sub.NMP X.2.sub.1-MN Z.2.sub.1-MN/NMP O 1s 9.11 7.66 6.00 N 1s 0.71 0.15 0.85 C 1s 88.94 91.13 91.85 P 2s 0.97 0.69 1.02 S 2p 0.26 0.36 0.27 Na 1s 0.02

[0161] FIG. 19 shows the C1s XPS spectrum of thin films from a) X.1.sub.NMP, b) X.2.sub.1-MN and c) Z.sub.1-MN/NMP. The C 1s spectra showed a main peak, which was charge-referenced to an approximate binding energy of 285 eV. This feature can be deconvoluted and fitted with five peaks: CC (285.0 eV), CO (286.5 eV), CO (287.8 eV), OCO (289.5 eV) and a -* shake up, as seen in aromatics (291.5 eV). Groups who have synthesised graphene have also reported these five features,.sup.2 although here the ratio of oxygenated carbon is somewhat higher and closer to levels typically reported for reduced graphene oxide (rGO) (Table 8)..sup.3,4

TABLE-US-00008 TABLE 8 Composition of C 1s XPS signal attributed to carbon in different environments for samples in this work compared against some reported values for graphene, reduced graphene oxide (rGO) and graphene oxide (GO). X. X. Graph- Group 1.sub.NMP 2.sub.1-MN Z.sub.1-MN/NMP ene.sup.2 rGO.sup.3 rGO.sup.4 GO.sup.3 GO.sup.4 CC 63% 77% 68% 79% 70% 69% 44% 49% CO 22% 14% 14% 10% 23% 18% 45% 45% CO 10% 6% 13% 3% 14% 14% 8% 7% C(O)O 3% 2% 3% 6% 1% 2%

[0162] Oxygen spectra were broad and curve-fitted with two main components at 533 eV and 534-535 eV. However, strong overlap with oxygen in Al.sub.2O.sub.3 is expected and it is therefore difficult to be specific about the chemical groups present.

[0163] The N is line was not sufficiently intense on sample X.2.sub.1-MN to enable any useful acquisition at higher resolution. Other spectra had poor signal-to-noise and curve fitting was therefore subject to statistical error. A component at 400 eV can be attributed to nitrogen in a relatively neutral electronic environment e.g. in amine bonds or similar. A second weak component at 402-403 eV is also seen which would normally be attributed to protonated nitrogen e.g. in a quaternary ammonium ion. There is some weak evidence for a third component at approximately 406.4 eV in X.1.sub.NMP which would correspond to oxidised nitrogen in the form of nitrite-like or nitrate-like groups.

[0164] Phosphorus was strongly correlation with the aluminium signal and is therefore assumed to be a low level impurity of the alumina filter. The low sulphur levels correlated with carbon levels but not the aluminium signal. Close inspection of the survey scans showed that sulphur, when detected with a sufficient signal-to-noise ratio, was present in a fully oxidised state e.g. as sulphate SO.sub.4.sup.2.

[0165] FIG. 20 shows the Raman spectra of thin films from X.1.sub.NMP, X.2.sub.1-MN, Y.1.sub.NMP and Z.sub.1-MN/NMP. The Raman spectra show both G and D bands for all samples although these features were very weak in X.2.sub.1-MN, presumably due to the low levels of material captured on the alumina filter. The G band is shifted to a higher wavenumber (1590 cm.sup.1) than in graphite or graphene (1580 cm.sup.1), reflecting the degree of oxidation and the presence of sp.sup.3 carbon atoms. The D band (1360 cm.sup.1) also has a higher intensity than in pure forms of graphite and graphene, which can be attributed to defects and disorder at both the edge and on the basal plane of nanosheet structures. The intensity of the 2D band is low and seen in combination with other overtone bands (D+G and 2G)again consistent with oxidized forms of graphene..sup.5 The intensity ratio of these bands (I.sub.D/I.sub.G0.83) and their full width at half maximum (FWHM.sub.G 87 cm.sup.1) are consistent with values reported for nanosheets of graphene with a degree of oxidation..sup.5

Example 7Functionalisation of Graphene Nanomaterial (Samples AA-AQ)

[0166] Graphene is known to have poor colloidal stability in most common solvents. However dispersion can be aided by the functionalisation of partially oxidised graphene in which the oxygen-containing groups that are present can be used to attach different functional groups. This can be done via sonochemical reactions in which an additional molecule is introduced into the liquid medium comprising diaromatic hydrocarbon whilst it undergoes cavitation. By using functional groups with different polarities it becomes possible to stabilise graphene colloids across a range of different solvents.

[0167] Octylamine (bp 178 C., mp 1 C.) is miscible with 1-MN and when present at 10 vol % produces a significant reduction in the rate of particle formation (Table 9). Functionalisation with the alkylamine results in an improved colloidal stability in 1-MN when it is used at 0.2, 2 and 10 vol % (FIG. 21).

[0168] Dodecylamine (bp 247 C., mp 27 C.) forms a cloudy macroemulsion when shaken with 1-MN. When ultrasound is applied the emulsion droplets become smaller leading to the formation of a transparent nanoemulsion. The rate of particle formation during cavitation is not so severely affected as with octylamine (Table 9). The longer alkyl chain length in the molecule leads to more effective colloid stabilisation (FIG. 22), although at 10 vol % the kinetic stability of the nanoemulsion formed during sonication is sufficiently poor to cause some inconsistencies in the stability of the colloidal graphene.

TABLE-US-00009 TABLE 9 Functionalisation with alkylamines. Sample Energy.sup.a A l.sup.1t.sup.1 i.d. Liquid composition (kJ) (m.sup.1 min.sup.1) P(1) 1-MN (batch 1) 600 2.16 P(2) 1-MN (batch 2) 600 2.10 P(3) 1-MN (batch 3) 600 2.09 AA 0.2 vol % Octylamine.sup.b 600 2.10 AB 2 vol % Octylamine.sup.b 600 1.86 AC 10 vol % Octylamine.sup.b 600 0.97 AD 0.2 vol % Dodecylamine.sup.b 600 2.08 AE 2 vol % Dodecylamine.sup.b 600 1.75 AF 10 vol % Dodecylamine.sup.b 600 1.68 .sup.ameasured on the in-built meter of the ultrasonic processor .sup.bin 1-MN

[0169] Water has a very low solubility in 1-MN and for additions above 200 ppmv cloudy water-in-oil emulsions are formed. Again the application of ultrasound reduces the size of the water droplets in these emulsions until transparent nanoemulsion formation occurs. Such nanoemulsions are kinetically stable and result in small reductions in the rate of particle formation (Table 10). The introduction of water at 200 ppmv, 0.2 vol % and 2 vol % leads to colloids that are less stable than when just 1-MN undergoes cavitation (FIG. 23). Similarly hydrogen peroxide can be used as a functionalising group and produces a similar level of colloidal stability as when just water is used (FIG. 24).

[0170] Acetic acid (bp 118 C., mp 16 C.) and tri(ethylene glycol) monoethyl ether (bp 256 C.) are both miscible with 1-MN and when introduced at 2 vol % produced a small reduction in the rate of particle formation (Table 10). Functionalisation with these molecules leads to a slight decrease in the colloidal stability of graphene nanosheets in 1-MN (FIG. 25).

TABLE-US-00010 TABLE 10 Functionalisation with H.sub.2O, HOOH, acetic acid and tri(ethylene glycol) monoethyl ether. Sample Energy.sup.a A l.sup.1t.sup.1 i.d. Liquid composition (kJ) (m.sup.1 min.sup.1) P(1) 1-MN (batch 1) 600 2.16 P(2) 1-MN (batch 2) 600 2.10 P(3) 1-MN (batch 3) 600 2.09 AG 200 ppm(v) H.sub.2O.sup.b 600 2.12 AH 0.2 vol % H.sub.2O.sup.b 600 1.96 AI 2 vol % H.sub.2O.sup.b 600 1.75 AJ 0.2 vol % HOOH in H.sub.2O.sup.b 600 1.79 AK 2 vol % Acetic acid.sup.b 600 1.84 AL 2 vol % Tri(ethylene glycol) 600 1.85 monoethyl ether.sup.b .sup.ameasured on the in-built meter of the ultrasonic processor .sup.bin 1-MN

[0171] Aniline (bp 184 C., mp 6 C.), benzoic acid (bp 249 C., mp 122 C.) and phenol (bp 182 C., mp 41 C.) are either miscible with or soluble in 1-MN. Each of these aromatic molecules was added to 1-MN (ex Sigma Aldrich) at 2 vol %. Formic acid (bp 101 C., mp 8 C.) is not miscible with 1-MN and at 2 vol % forms a macroemulsion when shaken with 1-MN. Again when ultrasound is applied the macroemulsion becomes a transparent nanoemulsion. The addition of these molecules resulted in no significant changes in the rate of particle formation during cavitation (Table 11). Formic acid produced a less stable colloid in 1-MN whereas the three other molecules increased colloidal stability (FIG. 26). The effectiveness of colloidal stabilisation follows the order benzoic acid<aniline<phenol.

TABLE-US-00011 TABLE 11 Functionalisation with aniline, benzoic acid, phenol and formic acid Sample Energy.sup.a A l.sup.1t.sup.1 i.d. Liquid composition (kJ) (m.sup.1 min.sup.1) AM 1-MN (ex. Sigma Aldrich) 600 1.96 AN 2 vol % Aniline.sup.b 600 1.93 AO 2 vol % Benzoic acid.sup.b 600 2.02 AP 2 vol % Phenol.sup.b 600 2.03 AQ 2 vol % Formic acid.sup.b 600 1.99 .sup.ameasured on the in-built meter of the ultrasonic processor .sup.bin 1-MN

Discussion

[0172] The compression and rarefaction of sound waves when passed through a liquid can produce bubbles; the formation, growth and collapse of which is known as cavitation. These bubbles, which are comprised of vapour and dissolved gases, shrink and expand under the influence of the acoustic field. Individual bubbles experience interference from their surroundings and consequently expansion to an unstable size can be followed by an implosive collapse. This produces localised hot spots (>5000 K), which are characterised by very rapid heating and cooling rates (>10.sup.9 K s.sup.1).sup.6-10

[0173] During sonication the maximum radius of a bubble (R.sub.max) and the time that it takes to collapse () is defined by:.sup.11

R.sub.max=4/(3w.sub.a)(P.sub.AP.sub.h)(2/(P.sub.A).sup.1/2[1+2(P.sub.AP.sub.h)/3P.sub.h].sup.1/3[1]

=0.915R.sub.max(/P.sub.m)1/2(1+P/P.sub.m)[2]

Where w.sub.a is the applied circular frequency (2f.sub.a) (s.sup.1), P.sub.A is the amplitude of the oscillating acoustic pressure (Nm.sup.2), P.sub.h is the hydrostatic pressure (Nm.sup.2), is the density (kgm.sup.3) of the liquid being irradiated, P.sub.m is the pressure (Nm.sup.2) in the liquid at the moment of collapse (which is the sum of the ultrasound amplitude during sonication and the ambient liquid pressure, (P.sub.A+P.sub.h) and P is the vapour pressure (Nm.sup.2) inside the collapsing bubble (typically assumed to be the vapour pressure of the liquid). If >.sup.th of the cycle time of the acoustic pressure (10 s for ultrasound at 20 kHz) then insufficient time is available for bubbles to undergo complete collapse..sup.11 Consequently R.sub.max is limited for the implosive bubble collapse of diaromatic hydrocarbons during cavitation induced with ultrasound.

[0174] Some of the first reports of non-aqueous sonochemistry were published over fifty years ago and suggest that the sonication of aromatic and heterocyclic compounds produces ring cleavage and acetylene production..sup.12-14 Subsequently Suslick and co-workers found that alkanes undergo sonochemical reactions, which are similar to high temperature (>1200 C.) pyrolysis..sup.15 For example, the products of n-decane sonolysis are hydrogen, methane, acetylene and a series of alkenes including ethylene, propylene, butene, pentene, etc. This is consistent with the operation of a radical chain Rice mechanism..sup.16 Misik and Riesz in related work trapped and identified the radicals produced during the sonolysis of a number of different organic liquids, including n-alkanes. Their results were also consistent with a pyrolysis mechanism..sup.17-18 Cataldo found that the prolonged sonication of benzene, toluene, styrene, decalin and tetralin produced insoluble dark matter..sup.19 Infrared spectroscopy of the material from benzene sonolysis suggested that the product contained a cross-linked structure similar to radiation-damaged polystyrene. Decalin and tetralin sonication caused aromatisation reactions, although decalin was also cracked to o-xylene and ethylene. Somewhat related studies have also been carried out on middle distillate hydrocarbons (C.sub.8-C.sub.26), in particular for diesel fuels where there is an interest in understanding cavitation-induced fuel degradation in the high-pressure fuel systems of vehicles..sup.20-23

[0175] Katoh et al. irradiated benzene (bp 80 C.) with ultrasound and observed a slow formation of solid carbon particles (yellow colouration of the benzene after 1 hour and black sooty material formed after 12 hours)the formation of C.sub.60 in low yield (ca. 1 pg) was seen after 1 hour..sup.24 The same group also irradiated chlorobenzene (bp 131 C.) and 1,2-dichlorobenzene (bp 179 C.) in the absence and presence of a variety of metal particles (ZnCl.sub.2, Zn, Ni, NiCl.sub.2 and ZnO) to produce either carbon nanotubes or graphitic particles..sup.25 Graphitisation was not observed in the absence of these metals. They concluded that polymerisation reactions proceeded in the vapour phase to form disordered carbon (primary sonochemistry)..sup.26, 27 Annealing into more ordered structures occurred when metal particles were present through collisions that were induced by the turbulent flow and shockwaves produced by collapsing bubbles.

[0176] Assuming an adiabatic compression allows the maximum pressure and temperature that occur inside a collapsing bubble to be calculated:.sup.28, 29

T.sub.max=T.sub.o[P.sub.m(1)/P][3]

P.sub.max=P[P.sub.m(1)/P].sup.(-1)[4]

Where T.sub.0 is the ambient temperature (K) of the liquid, is the ratio of heat capacity of the bubble gas at constant pressure (C.sub.p) and volume (C.sub.v). The relatively high vapour pressure of mono aromatic hydrocarbons compared to diaromatic hydrocarbons means the temperatures produced inside bubbles are not sufficiently high to generate significant and rapid particle formation. For example if identical conditions (T.sub.o and P.sub.m) are assumed for different liquids then from equation [3] it follows that T.sub.max (1-MN)/T.sub.max (toluene) 200 and T.sub.max(1-MN)/T.sub.max(benzene) 500. Temperatures inside the bubbles generated by the cavitation of water have been experimentally determined at 4300200 K..sup.30 Equation [3] gives T.sub.max(1MN)/T.sub.max(H.sub.2O) 50 and this would suggest that accurate estimates of T.sub.max for diaromatic hydrocarbons from this equation are not possible. However it is experimentally observed that only the low vapour pressures of diaromatic hydrocarbons (e.g. 1-MN and NAP) are uniquely able to produce the conditions required for homolytic fission of CH bonds and rapid formation of particles in the vapour phase of the bubbles.

[0177] Cyclodehydrogenation reactions enable the growth of two-dimensional carbon sheets..sup.31-33 If a reaction takes place between the -positions (1,8 or 4,5) on two adjacent naphthalene molecules then perylene (the second molecule in the homologous rylene series) is produced by a peri-condensation along the zig-zag edge of the molecules.

##STR00002##

[0178] An armchair feature is established by this reaction, which then allows for a second type of cyclodehydrogenation that produces a cata-condenstaion involving adjacent -positions (2,3 or 6,7).

##STR00003##

[0179] A perfect sheet of graphene is grown if only , and , CC bonds are formed. The repeating unit cells established by the two reactions are shown below.

##STR00004##

[0180] However if condensation reactions occur between the and position of two adjacent molecules, perfect tessellation breaks down and holes (see dots in 8) will appear within the growing sheet.

##STR00005##

[0181] The temperature of the diaromatic hydrocarbons rapidly equilibrated during sonication to 40 C. Hydrocarbons contain small amount of dissolved air: 100-130 mg/L between 10 and 40 C. (determined from Ostwald coefficients reported as a function of temperature for hydrocarbons)..sup.34 Some of this air is removed during the early stages of sonication, with the degree of degassing being dependant on the level of power being used to induce cavitation..sup.35 Sonication of diaromatic hydrocarbons at the input power used in these experiments produced material with a C/O atom ratio of 10 (X.1.sub.NMP). This is consistent with naphthalene and oxygen reacting in a 2:1 mole ratio in the collapsing bubbles. C/O atom ratio of the product increased to 15 (Z.sub.1-MN/NMP) when nitrogen was bubbled through the reaction medium, consistent with 3 moles of naphthalene reacting with 1 mole of oxygen. Further removal of dissolved air will lead to even higher C/O atom ratios and the production of purer forms of graphene nanomaterial. Freeze-pump-thaw, bubbling of inert gasses and inducing cavitation all offers means of reducing the presence of dissolved oxygen in the diaromatic hydrocarbons and decreasing the level of oxidation in the graphene produced.

[0182] When NMP (bp 202 C.) is added to the reaction mixture (Sample 0 in Table 4) then the vapour pressure of this component (P=133 Nm.sup.2 cf. P=24 Nm.sup.2 for 1-MN and P=99 Nm.sup.2 for NAP).sup.36,37 means that the mole fraction of diaromatic hydrocarbons in the bubble is reduced. Additionally T.sub.max decreases as the total vapour pressure of the bubble is raised. This results in a ten-fold reduction in the rate of nanosheet formation, which is consistent with the sonochemistry of particle formation taking place in the vapour phase of collapsing bubbles (primary sonochemistry).

[0183] The partially oxidised graphene sheets produced during cavitation undergo perikinetic agglomeration to produce particles of a measurable size (1 m). This material can undergo flocculation to eventually produce black sediments. However some of the sheets produced remain dispersed (due to variations in composition and/or size) until n-heptane is added and they then form brown sediment. The brown colouration is consistent with sheets of smaller dimensions as found for GQDs..sup.38,39. Sparging the reaction mixture with nitrogen before cavitation increases the C/O atom ratio in the product from 10 to 15, consistent with a partial removal of dissolved oxygen from the diaromatic hydrocarbons. This shift in composition and possibly change in the distribution of sheet sizes to smaller dimensions means that although some agglomeration still occurs this does not result in flocculation and sedimentation. FIG. 27 shows the consequences of diaromatic hydrocarbon cavitation produced by irradiation with ultrasound (poGNpartially oxidised graphene nanosheets).

[0184] Based on reported Hansen Solubility Parameters (HSPs) for graphene (G),.sup.40 reduced graphene oxide (rGO) and graphene oxide (GO).sup.41 it is possible to assess how readily these two-dimensional materials might disperse in mixtures of 1-MN, NAP and NMP by calculating distances in Hansen solubility space:.sup.42

R.sup.2=4(.sub.D1.sub.D2).sup.2+(.sub.P1.sub.P2).sup.2+(.sub.H1.sub.H2).sup.2

More highly oxidized forms of graphene are more likely to agglomerate in diaromatic solvents. FIG. 28 shows changes in Hansen solubility distance (R) between 2-dimensional materials (graphene (G), reduced graphene oxide (rGO) and graphene oxide (GO)) and di-aromatic hydrocarbons (.sub.1-MN=0.8, .sub.NAP=0.2) upon addition of NMP. The HSPs used to calculate the Hansen solubility distance in FIG. 28 are listed in Table 12.

TABLE-US-00012 TABLE 12 Hansen solubility parameters. .sub.D1 .sub.P .sub.H Material (MPa.sup.1/2) (MPa.sup.1/2) (MPa.sup.1/2) 1-Methylnaphthalene (1-MN).sup.43 20.6 0.8 4.7 Naphthalene (NAP).sup.43 19.2 2.0 5.9 N-methyl-2-pyrrolidone (NMP).sup.43 18.0 12.3 7.2 Graphene (G).sup.40 18.0 9.3 7.7 Reduced graphene oxide (rGO).sup.41 17.9 7.9 10.1 Graphene oxide (GO).sup.41 17.1 10.0 15.7

[0185] The sheets of graphene produced in this work have levels of oxidation similar to those reported for rGO and the addition of NMP (.sub.NMP=0.5) after sonication is therefore sufficient to establish stable dispersions of nanosheets. Indeed the Hansen solubility distance could be used to determine other mixtures and solvents that would produce stable colloidal suspensions. The turbostratic structure of nanosheets that forms through agglomeration is disordered and allows sediment to be dispersed into stable colloids of graphene nanosheets by choosing a solvent with an appropriate set of HSPs. The presence of some residual aggregate particles in dispersed colloids may be a consequence of inter-particle collisions driven by the shockwaves generated during the implosion of bubbles forcing sheets together into more graphitic-like material. These aggregates are therefore more ordered structures, which are resistant to dispersion back into nanosheets. However the complete separation of residual aggregate from dispersed nanosheets should be achievable by standard separation techniques.

[0186] Making full use of the properties of graphene will require its dispersion in a broad range of solvents. Adjusting levels of oxidation (GO v rGO v G) can increase the range of solvents that may be used to achieve stable colloids but such oxidation has an impact on the physical properties of the 2-dimensional sheets. Chemical functionalisation of graphene provides another means of achieving greater dispersibility..sup.44 This can be done via non-covalent modifications in which - or cation- interactions are used to adsorb molecules on to the graphene surface. Alternatively surfactants and particles have also been employed to make graphene more dispersible. Covalent modification of graphene typically employs the enhanced chemical reactivity of GO or rGO to produce functionalised graphene. GO and rGO contain a range of epoxide, ether, aldehyde, ketone, alcohol and carboxylic acid groups which provide reactive sites on the sheets at which covalent functionalization can be achieved. Nucleophilic substitutions (e.g. the reaction of the amine functionality of organic modifiers at epoxy groups on the sheet), electrophilic substitutions (e.g. the grafting of aryl diazonium salts to the surface of graphene), condensation reactions (e.g. the reaction of carboxylic acid and hydroxyl groups in GO with isocyanate functionalised hydrocarbons to form amide and carbamate ester linkages) and addition reactions (e.g. the reaction of alkyl azides with graphene via a biradical or [2+1] cycloaddition pathway to form functionalised graphene sheets) have all been the subject of review..sup.45

[0187] Example 7 illustrates how covalent functionalization may be achieved during the production of graphene nanosheets by the cavitation of diaromatic hydrocarbons. Alkylamines are known to produce nucleophilic ring opening of epoxides and the grafting of alkylamine onto the sheet surface..sup.46-48. The volatility of octylamine (bp 178 C.) is sufficiently high to result in a significant impact on the composition of the collapsing bubbles (Table 9). Indeed when introduced into the reaction mixture at =0.1 bubble vapour pressure is significantly increased, resulting in a reduction in the rate of particle formation. The longer alkyl chain in dodecylamine (bp 247 C.) results in a much-reduced impact on bubble composition (Table 13). The ability of both alkylamines to produce colloid stabilisation in 1-MN would suggest that the reaction of the alkylamine with the partially oxidised graphene nanosheets takes place, at least in part, in the liquid phase outside of the collapsing bubble (secondary sonochemistry).

TABLE-US-00013 TABLE 13 Impact of alkylamine on bubble composition. Material .sub.RNH2.sup.a x.sub.RNH2.sup.b Octylamine 0.002 0.025 0.020 0.210 0.100 0.591 Dodecylamine 0.002 0.000 0.020 0.003 0.100 0.018 .sup.aliquid fraction in reaction mixture .sup.bcestimated mole fraction of RNH.sub.2 in the bubble assuming complete degassing

[0188] Aromatic organic molecules with different functional groups (NH.sub.2, OH, COOH) can be used to produce functionalisation. Importantly materials that are not miscible or soluble in diaromatic hydrocarbons can also be used to produce functionalisation. For example the formation of nanoemulsions allows water and formic acid to react with nanosheets, making them more polar (via the incorporation of OH and O(O)CH groups) and less stable in 1-MN. It therefore becomes possible to combine solid molecules in a range of different solvents or liquid molecules that are not miscible with diaromatic hydrocarbons and produce functionalisation during the production of graphene nanomaterials by cavitation.

CONCLUSION

[0189] Cavitation of diaromatic hydrocarbons provides a means of generating the conditions necessary for a rapid formation of graphene nanomaterial in the vapour phase of imploding bubbles. The level of dissolved air in the liquid reaction medium has an impact on the degree of partial oxidation and for untreated diaromatic hydrocarbons material similar in composition to reduced forms of graphene oxide (rGO) is produced. Reducing the level of dissolved air in the reaction mixture can produce material closer to pristine forms of graphene.

[0190] Colloid dispersions of nanomaterial particles that are partially oxidised are prone to agglomeration when dispersed in the starting reaction medium. Larger agglomerates flocculate and then form sediments when dispersions are left to stand. However stable dispersions of sheets can be obtained by adjusting the Hansen Solubility Parameters of the reaction medium before, during or after cavitation. Alternatively sediments may be removed from the reaction medium and re-dispersed into an appropriate solvent to obtain stable dispersion of nanosheets.

[0191] The stability of colloid dispersion can also be adjusted by decreasing the level of dissolved air present in the reaction medium during cavitation. Colloidal stability may also be achieved for a range of different solvents by functionalising the material produced during cavitation by introducing appropriate molecules into the reaction medium. These molecules may or may not be soluble in or miscible with diaromatic hydrocarbons.

[0192] This bottom up synthesis has potential to be modified to allow for the production of graphene in forms required for different applications and also controlled for the yield of sheet dimensions down to the scales required for the fabrication of graphene quantum dots.

REFERENCES

[0193] 1. K. R. Paton et al., Nature Materials, 2014, 13, 624-630. [0194] 2. C. R. Herron, K. S. Coleman, R. S. Edwards and B. G. Mendis, J. Mater. Chem., 2011, 21, 3378-3383. [0195] 3. J. Kauppila, Graphene from graphite by chemical and physical techniques, Ph.D. thesis, University of Turku, 2014. [0196] 4. Sobon et al., Optics Express, 2012, 20, 19464-73. [0197] 5. K. Krishnamoorthy, M. Veerapandian, K. Yun and S.-J. Kim, Carbon, 2013, 53, 38-49. [0198] 6. K. S. Suslick, and D. J. Flannigan, Annu. Rev. Phys. Chem., 2008, 59, 659-83. [0199] 7. K. S. Suslick, D. A Hammerton and R. E. Cline, J. Am. Chem. Soc., 1986, 108, 5641-2. [0200] 8. E. B. Flint and K. S. Suslick, Science, 1991, 253, 1397-9. [0201] 9. Y. T. Didenko, W. B. McNamara and K. S. Suslick, J. Phys. Chem. A, 1999, 103, 10783-8. [0202] 10. W. B. McNamara, Y. T. Didenko and K. S. Suslick, Nature, 1999, 401, 772-5. [0203] 11. T. J. Mason and J. P. Lorimer, Applied sonochemistry: the use of power ultrasound in chemistry and processing, Weinham: Wiley-VCH; 2002. [0204] 12. L. Zechmeister and L. Wallcave, J. Am. Chem. Soc., 1955, 77, 2853-5. [0205] 13. L. Zechmeister and E. F. Magoon, J. Am. Chem. Soc., 1956, 78, 2149-50. [0206] 14. D. L. Currell and L. Zechmeister, J. Am. Chem. Soc., 1958, 80, 205-8. [0207] 15. K. S. Suslick, J. J. Gawienowski, P. F. Schubert and H. H. Wang, J. Phys. Chem., 1983, 87, 2299-301. [0208] 16. P. Riesz, D. Berdahl and C. L. Christman, Environ. Health Perspect., 1985, 64, 233-52. [0209] 17. V. Misik and P. Riesz, J. Phys. Chem., 1994, 98, 1634-40. [0210] 18. V. Misik and P. Riesz, Ultrason. Sonochem., 1996, 3, 25-37. [0211] 19. F. Cataldo, Ultrason. Sonochem., 2000, 7, 35-43. [0212] 20. G. J. Price and M. McCollom, Fuel, 1995, 74, 1394-7. [0213] 21. G. J. Price and M. McCollom, Ultrason. Sonochem., 1995, 2, S67-70. [0214] 22. R. D. Lockett and M. Jeshani, Int. J. Eng. Res., 2013, 6, 606-21. [0215] 23. R. J. Price, D. Blazina, G. C. Smith and T. J. Davies, Fuel, 2015, 156, 30-39. [0216] 24. R. Katoh, E. Yanase, H. Yokoi H, S. Usuba, Y. Kakudate and S. Fujwara, Ultrason. Sonochem., 1998, 5, 37-8. [0217] 25. R. Katoh, Y. Tasaka, E. Sekreta, M. Yumara, F. Ikazaki, Y. Kakudate and S. Fujjiwara, Ultrason. Sonochem., 1999, 6, 185-7. [0218] 26. R. Katoh, H. Yokoi, S. Usuba, Y. Kakudate and S. Fujiwara, Ultrason. Sonochem., 1998, 5, 69-72. [0219] 27. H. Xu, B. W. Zeiger and K. S. Suslick, Chem. Soc. Rev., 2013, 42, 2555-2567. [0220] 28. B. E. Noltingk and E. A. Neppiras, Proc. Phys. Soc. 8, 1950, 638, 674-85 (London). [0221] 29. E. A. Neppiras, Phys. Rep. Rev. Sec. Phys. Lett., 1980, 61, 159-251. [0222] 30. Y. Didenko, W. B. McNamara III and K. S. Suslick, J. Phys. Chem. A., 1999, 103, 10783-10788. [0223] 31. M. Murakami, Synthetic Metals, 1987, 18, 531-536 [0224] 32. H. Kamo, M. Yudasaka, S. Kurita, T. Matsui, R. Kikuchi, Y. Ohki and S. Yoshimura, Synthetic Metals, 1994, 68, 61-63. [0225] 33. M. Treier, C. A. Pignedoli, T. Laino, R. Rieger, K. Mullen, D. Passerone and R. Fasel, Nature Chemistry, 2011, 3, 61-67. [0226] 34. Handbook of Fluid Dynamics and Fluid Machinery: Fundamentals of Fluid Dynamics, Volume I, Ed. by J. A. Schetz and A. E. Fuhs, Chichester: John Wiley & Sons, Inc; 1996, p.sub.161-162. [0227] 35. T. Vencel, J. Donovalova, A. Gaplovsky, T. Kimura and S. Toma, Chem. Pap., 2005, 59(4), 271-274. [0228] 36. NIST Chemistry WebBook, http://webbook.nist.gov/chemistry/ [0229] 37. Dortmund Data Bank, DDBST GmbH (Oldenburg), http://ddbonline.ddbst.com/AntoineCalculation/AntoineCalculationCGI.exe [0230] 38. P. He, J. Sun, S. Tian, S. Yang, S. Ding, G. Ding, X. Xie and M. Jiang, Chem. Mater., 2015, 27(1), 218-226. [0231] 39. S. Kim, D. H. Shin, C. O. Kim, S. S Kang, S. S. Joo, S-J. Choi, S. W. Hwang and C. Stone, Appl. Phys. Let., 2013, 102, 053108. [0232] 40. Y. Hernandez, M. Lotya, D. Rickard, S D. Bergin and J. N. Coleman, Langmuir, 2010, 26(5), 3208-3213. [0233] 41. D. Konios, M. M. Stylianakis, E. Stratakis and E. Kymakis, Journal of Colloid and Interface Science, 2014, 430, 108-112. [0234] 42. M. Yi, Z. Shen, X. Zhang and S. Ma, J. Phys. D: Appl. Phys., 2013, 46, 025301. [0235] 43. C. M. Hansen, Hansen Solubility Parameters: A User's Handbook, 2.sup.nd Edition, Boca Raton: CRC Press; 2007. [0236] 44. D. W. Johnson, B. P. Dobson and K. S. Coleman, Current Opinion in Colloid & Interface Science, 2015, 20, 367-382. [0237] 45. T. Kuila, S. Bose, A. K. Mishra, P. Khanra, N. H. Kim and J. H. Lee, Progress in Materials Science, 2012, 57, 1061-1105. [0238] 46. X. Yang, T. Mei, J. Yang, C. Zhang, M. Lv and X. Wang, Applied Surface Science, 2014, 305, 725-731. [0239] 47. J. Jang, V. H. Pham, S. Hyun, J. S. Chung, Journal of Colloid and Interface Science, 2014, 42, 62-66. [0240] 48. P. Feicht, D. A. Kunz, A. Lerf and J. Breu, Carbon, 2014, 80, 229-234.