WO1993001128A1

WO1993001128A1 - Preparation and separation of fullerenes

Info

Publication number: WO1993001128A1
Application number: PCT/AU1992/000345
Authority: WO
Inventors: Michael Amos Wilson; Anthony Michael Vassallo; Louis Sing Kim Pang; Andrew John Palmisano
Original assignee: Commonwealth Scientific And Industrial Research Organisation
Priority date: 1991-07-10
Filing date: 1992-07-10
Publication date: 1993-01-21

Abstract

Fullerenes may be produced from coal by heating the coal to render it electrically conductive. An electric current is then passed through this material to volatilise at least a part of the coal in an inert atmosphere substantially free of oxygen. The condensed soot recovered contains fullerenes. The mixture of fullerenes so obtained may be separated by extraction into a suitable organic solvent and adsorbed onto a solid carbonaceous material. The individual fullerenes may be sequentially eluted from the carbonaceous material.

Description

PREPARATION AND SEPARATION OF FULLERENES Field of the Invention

The present invention relates to a process for producing buckminsterfullerene and other fullerenes. In particular, the present invention relates to a process for producing buckminsterfullerene and other fullerenes from coal or like carbonaceous materials. The present invention further relates to the separation of a mixture of fullerenes into its component f llerenes. Background Art

A new form of carbon was identified in 1985 (Kroto et al. Nature 1985, 318, 162-163) which has subsequently been isolated (Haufler et al Journal of Physical Chemistry 1990, 94, 8634-8636). This compound was termed buckminsterfullerene.

Its structure was confirmed as a spherical ball molecule consisting of a series of six and five membered rings in the structure of a soccerball. The molecule generally has sixty carbon atoms and hereinafter this form of the molecule will be referred to as Cg_Q. Other structures collectively termed fullerenes, such as fullerenes C__n and C_fi. with similar structures but containing extra six membered rings were also isolated. All of these structures have become collectively known as fullerenes. Alternative names for these structures include buckyball or soccerball carbon.

Reported processes for producing buckminster- fullerenes and the other fullerenes in macroscopic quantities have generally relied upon graphite as a raw material. The graphite is heated in a furnace or with an electrical arc under an atmosphere of argon or other suitable inert gas such as helium. Such processes have been reported by Kratschmer, Nature, 347,354, Aje et al. Journal of Physical Chemistry, 1990, 94, 8630; Taylor et al, Journal of the Chemical Society Chemical communications 1990, 1423; Haufler et al, Journal of Physical Chemistry, 1990, 94, 8634, Allemand et al, Journal of the American Chemical Society 1991, 113, 1050. It has also been reported that microscopic amounts of fullerenes may be produced by the laser heating of coal (Greenwood et al, Organic Mass Spectrometry 1990, 25, 353-362). It has subsequently been reported that partial combustion of benzene in an argon-oxygen mixture produced fullerenes (Howard et al, Nature, 1991, 352, 139-141).

One problem with preparing large quantities of C_fin and other forms of fullerenes on an industrial scale is the cost of graphite that is used as a raw material. So far, fullerenes are uniquely prepared in macroscopic quantities from the heating or arcing of graphite. In order to produced fullerenes economically, it is imperative to find a cheap source material since graphite is already a value-added carbon, costing between US$1,000-$5,000 per tonne.

Another problem with preparing large quantities of C_gn and other forms of fullerenes on an industrial scale is the difficulty of separating the different fullerenes from each other. Following generation of the fullerenes, it has been found that several different forms of the fullerenes are intimately mixed in the product. Separation (as described in the prior art) has been carried out by extraction with an organic solvent preferably toluene and then the crude product is placed on a chromatography column of alumina or silica gel and eluted with hexane. This later process is extremely time consuming and is unsuitable for an industrial process. Disclosure of the Invention

In contrast to the prior art, the present invention seeks to generate fullerenes by using coal or like carbonaceous materials as a source material and/or to use a carbonaceous material such as coal or graphite, as an adsorption medium to assist in the separation of individual fullerenes from the mixture.

Accordingly, in a first aspect the present invention provides a process for producing fullerenes from coal or like carbonaceous material which comprises: a) treating said coal or like carbonaceous material under conditions such that at least a portion of said coal or like carbonaceous material becomes electrically conductive; and b) causing an electric current to pass through the treated coal or like carbonaceous material to thereby volatilise at least a part of said coal or like carbonaceous material, step (b) being carried out in an inert, substantially oxygen free, atmosphere. Preferably, step (b) is carried out in an inert atmosphere comprising helium, neon, argon, krypton or mixtures thereof, said atmosphere being substantially free of oxygen.

The present invention is distinguished from known prior art in that it enables coal or other like carbonaceous materials to be used as a raw material. Coal is not carbon but a complex mixture of organic compounds comprising 1-8% hydrogen, 0-5% nitrogen, 0-5% sulphur and up to 50% oxygen. In contrast to the present invention, the prior art processes for the production of macroscopic quantities of fullerenes have required the use of pure carbon forms, especially graphite, as a raw material.

In the process of the present invention, an electric current may be caused to pass through the treated coal or like carbonaceous material by one of the following methods: i) The treated coal or like carbonaceous material may form an electrode of a system having at least two electrodes. The electrodes may be spaced apart from each other. A voltage is applied across the electrodes and this causes an electric arc to be generated between the electrodes, resulting in an electric current passing through the coal or like carbonaceous material. This causes volatilisation of at least part of the coal or like carbonaceous material. ii) The treated coal or like carbonaceous material may be subjected to electro-induction heating. In this method, a wire is wrapped around the coal or like carbonaceous material. High frequency current is passed through the wire which causes current to pass through the coal or like carbonaceous material, and volatilisation of part of the coal consequently occurs. iii) A single rod of treated coal or like carbonaceous material may be resistively heated by passing an electrical current through the coal or like carbonaceous material.

In each of (i), (ii) and (iii) above, the volatilised material is collected as a soot and fullerenes may be recovered therefrom.

As used in respect of this first aspect of the present invention the term "coal or similar carbonaceous material" means coal; another naturally occurring carbonaceous material such as peat, oil, or wood; or impure carbonaceous products derived from one of the foregoing at temperatures of less than 2000 C. Suitable impure carbonaceous products include pitch, charcoal, petroleum tars, and vacuum bottom tars.

The treatment carried out in step (a) of the present invention ensures that at least a portion of the treated coal or like carbonaceous material becomes electrically conductive which, of course, then enables the coal or like carbonaceous material to support the electric current of step (b) of the present invention. Following step (a), the treated coal or like carbonaceous material should have an electrical resistivity sufficiently low such that the treated coal or like carbonaceous material is able to conduct an electric current sufficient to produce arcing from the coal or like carbonaceous material, or to carry sufficient current to allow electrothermal heating to volatilise at least part of the treated coal or like carbonaceous material.

In step (a) the coal or like carbonaceous material is preferably treated under conditions such that at least a portion of the coal or like carbonaceous material is converted to coke. For example, the coal or like carbonaceous material may be heated to a temperature of from 350°C to 1800°C, preferably 1000°C to 1800°C in an inert atmosphere. The inert atmosphere may be helium, neon, argon or mixtures thereof. The coal or like carbonaceous material may be heated for any desired period of time, with a heating time of 0.25 - 48 hours being preferred, 1-25 hours being more preferred.

For brown coals and other non-coking coals, step (a) preferably comprises mixing the coal with a binder, such as pitch, prior to heating to form an electrically conductive rod. The heating step should be sufficient to at least partially carbonise the material in the rod.

The coal or like carbonaceous material is preferably finely ground and packed into a convenient shape, e.g. a rod or cube, prior to step (a) . However, it is possible to produce fullerenes from lumps of coal.

The process of the present invention may be carried out using coals of any rank, although brown and sub-bituminous coals are preferred. Metallurgical coke and the liquid crystal material called mesophase (which is intermediate between coke and coal) are also suitable for use in the present invention. It has been found that mesophase forms the highest yields of fullerenes and will produce fullerenes under mild conditions at which graphite is insufficiently active. Pitch may also be used as a starting material for the process .

Once the sample has been treated in step (a), the sample has a degree of electrical conductivity. For example if the sample is in the shape of a rod having dimensions of diameter 15-20mm and length 30-50mm, the final coke rod preferably has an electrical resistance of from 0.1 to 5.0 ohms across its length.

As stated above, the soot resulting from step (b) of the invention is collected and fullerenes may be recovered from the soot. The fullerenes may be conveniently recovered from the soot by dissolving the fullerenes in a suitable solvent. A preferred solvent is toluene. Other solvents that may be used include xylenes, 1,3,5-trimethylbenzene, 1,2,4-trichlorobenzene, 1-methylnapthalene, quinoline, benzene, pyridine, 1,2,3,5-tetramethylbenzene, hexane, heptane and mixtures thereof.

The fullerenes may be separated from the solvent by simply filtering the solution from the insoluble soot and then evaporating the solvent from the filtrate, leaving the fullerenes as a solid residue.

According to a second aspect, the present invention comprises a method for separating a mixture comprising at least a portion of first fullerenes and a portion of second fullerenes, which process comprises extracting said mixture into an organic solvent, contacting said organic solvent and extracted fullerenes with a solid carbonaceous material to adsorb said fullerenes on said carbonaceous material, and eluting at least part of said portion of first fullerenes from said solid carbonaceous material.

The method of the present invention may further comprise eluting said portion of first fullerenes and subsequently eluting said portion of second fullerenes. In a preferred embodiment, the carbonaceous material used in this second aspect of the present invention is coal, with anthracite coal or semi-anthracite coal being especially preferred. In a most preferred embodiment, the carbonaceous material is graphite. The carbonaceous material is preferably placed in a column.

The mixture of fullerenes may comprise a mixture of C_β0 and C-₀ fullerenes or indeed further include other fullerenes.

The organic solvent may be any organic solvent capable of extracting fullerenes. Hexane and toluene mixtures have been found to be especially suitable. Other solvents that may be used include toluene, xylenes, 1,3,5-trimethylbenzene, 1,2,4-trichlorobenzene, 1-methylnapthalene, quinoline, benzene, pyridine, 1,2,3,5-tetramethylbenzene, heptane and mixtures thereof. Once the solvent and extracted fullerenes have been loaded onto the carbonaceous materials, each portion of the fullerenes may be eluted by passing an eluting agent through the carbonaceous material. Hexane and toluene mixtures are preferred as the eluting agent. Brief Description of the Drawings

Preferred embodiments of the invention will now be described with reference to the following examples and Figures. In the Figures; Figure 1 shows a schematic diagram of the apparatus used to generate fullerenes;

Figure 2 shows a more detailed form of the apparatus of Figure 1;

Figure 3 shows an infra-red spectrum of fullerene produced from Goonyella coke; and

Figure 4 shows an infra-red spectrum of fullerene produced from Norwich Park coal being (a) crude fullerenes and (b) purified C_gfi.

Best Method for Carrying Out the Invention The coal is finely ground and heated in an inert argon atmosphere for 24 hours at 395 C to form a rod of approximately 18 mm diameter x 45 mm in length. Higher temperatures up to 1500°C are also suitable. Temperatures as high as 2500 C are not required since this results in the formation of graphite. The initial heating step causes partial coke formation and gives the rod sufficient strength to remain intact. The rod is subject to further coke formation by further heating at 1200°C in Argon (or neon or helium) for 5 hours although shorter and longer reaction times are also suitable. This second heating step is used to form coke having suitable conductivity for subsequently generating fullerenes. Of course, the twό-stage heating process to promote coke formation described in this example may be replaced by an appropriate single-stage heating process. The final coke rod should have an electrical resistance of 1.5 to 3 ohms across its length.

Apparatus used for the electrical arcing of the coke rods to produce fullerenes is shown diagramatically in Figure 1. The apparatus comprises a reaction vessel 10 containing coke rods 12 and 14. The coke rods 12 and 14 are respectively connected to an electrical source 17 through leads 16 and 18. The reaction vessel 10 is shown in more detail in figure 2. The reaction vessel 10 comprises a stainless steel cylindrical jacket 19 and stainless steel end flanges 21. A water cooled cylindrical copper jacket 22 is provided inside the vessel 10 and surrounding the coke rods 12 and 14. Water inlets to the jacket 22 through pipe 23 and outlets through pipe 24.

The vessel 10 is connected to a rotary vacuum pump (not shown) through duct 25 which is equipped with a vacuum valve 26 and a pressure transducer 27. Helium is introduced into the vessel 10 through inlet pipe 28. The coke rods 12 and 14 are held in copper electrodes 29 - St ¬

and 31 which are each cooled by water flowing in through duct 32 and out through duct 33. The electrode 31 is sealingly engaged with its associated one of the end flanges 21 by an O-ring 34. The electrode 29 holding the coke rod 12 is fixed and is connected to the negative pole of the electricity source when it is d.c. The electrode 31 holding the coke rod is movable longitudinally by a screw threaded moving mechanism 35 in a steel support frame 36. In a typical experiment, two coke rods are subject to electrical arcing in a 250 torr helium atmosphere and 24V a.c. at 105-110A. It has been found that it is essential that substantially all oxygen be removed from the reactor for adequate generation of fullerenes. Table 1 lists the yields of toluene soluble products and the duration of arcing for the different coals used. Infra-red spectroscopy of the toluene extracted products shows peaks characteristic of C_βn and C₇₀ in similar ratios to the products obtained from graphite. By carefully varying reaction conditions it is possible to promote the formation of C-_Q over C_β0. Nuclear magnetic resonance spectroscopy confirms the presence of C_fin and C_₀ as described in the prior art. The presence of mineral matter in the coal does not inhibit fullerene formation and certain minerals comprising phosphates, borates and sulphates and copper, lead or iron compounds can alter the distribution of fullerenes and other products in the resultant soot. The use of a d.c. rather than a.c. arc promotes fullerene formation and the use of higher currents (up to 1000A have been investigated) is also beneficial. Table 1. Experimental conditions and yields of crude fullerenes from coals

Arcing Toluene Time (h) Solubles (g)

4 0.25 4 0.15 4 0.18 4 0.10 4 0.24

4 0.26

Example 2

In this preparation, the primary requirement for the preparation of fullerenes from coal is the preparation of self supporting, electrically conductive rods. These rods are prepared from both coking and non-coking coals. A coking coal from Goonyella seam, Australia was finely ground and demineralised according to published procedures. Three fractions of this coal with mineral matter contents of 0.7, 2.9 and 7.9% (as received) respectively were obtained. Each coal fraction was ground again, placed in an aluminum mould and heated in an argon flow for 24 hours at 395°C to form a rod of approximately 18 mm diameter by 45 mm length. The rod was subjected to coke formation by further heating at 1200°C in an argon flow for 5 hours. The final coke rod has an electrical resistance of 1.5-2.4 ohms along its length. Conductive rods for the other coking or non-coking coals were prepared as follows. The coals were Coalcliff middlings, Newvale vitrinite, Yarrabee semi-anthracite and Loy Yang brown coal. Each coal was mixed with 20 wt. % pitch (Kopper's pencil, an oil derived pitch) as a "binder". Each coal-pitch composite and neat pitch, after being finely ground, was packed in a Swagelok type stainless steel tube of size 12mm i.d. by 152mm length and sealed. These were heated at 500°C for 20 hours to form a rod. This rod was then further heated at 1200 C for 5 hours in argon which was sufficient to carbonise the Coalcliff, Newvale and semi-anthracite composite so that they became electrically conductive. For Loy Yang composite and neat pitch, further heating at 1200 C for 10 hours and 1300°C for 1 hour in argon was required for sufficient electrical conductivity to be induced. Five oven cokes were supplied by the Australian Coal

Industries Research Laboratories (ACIRL) . These cokes were derived from Goonyella, Blackwater, Riverside,

Gregory and Norwich Park coking coals. They were prepared by heating coals of below 3.35mm size in a movable wall

3 coke oven of 0.42m volume under sealed condition at

1010°C for 16-19 h. Samples of cokes were cut into rods of about 10mm by 120mm size for fullerene production.

In a typical experiment, two coke rods were subjected to electrical arcing in a 250 torr helium atmosphere in a stainless steel chamber as shown in Figures 1 and 2. Both a.c. and d.c. currents were used. A Lincoln arc welding supply capable of delivering up to 1000 A at 65 V was used for d.c. experiments while a Commonwealth Industrial Gas arc welder supply capable of delivering up to 200 A at 36 V was used for a.c. experiments. In the case of d.c. current, graphite was used as the cathode and the coke or pitch to be vaporised made the anode. A voltage of 23-30 V was used. During d.c. arcing, the anode was vaporised but the cathode was not. For a.c. experiments, two coke rods were used and a voltage of 24 V was applied. A current in the range of 80 to 130 A was passed through the electrodes for all experiments. After some hours of operation, soot was deposited on the condenser and other parts of the chamber. The soot collected was Soxhlet extracted with toulene. The C,-_n and C__n fullerenes were separated by established chromatography techniques. Table 2 lists the carbon content and ash yield for the different cokes and carbonised pitch. Table 3 lists the yield data of the toluene soluble product (crude fullerene) and the duration of arcing for the different cokes and pitch used. The infrared spectrum of the toluene extracted product from the arcing of Goonyella coke (Figure 3) and Norwich Park Coke (Figure 4(a)) shows absorption peaks characteristic of C_gQ and C_₀ fullerenes, and in similar ratios (ca. 10:1 for C_fin : C_?n) to that reported for graphite as a source material. The ratios have been confirmed by solid-state

13 C nuclear magnetic resonance spectroscopy.

All cokes tested so far produced fullerenes. The presence of mineral matter in the coal does not inhibit fullerene formation from coke, however its presence does result in a reduced yield of fullerene. An optimum fullerene yield of 8.6% was obtained from superclean Goonyella Coke. This compares favourably with the yield of 9.3% obtained from graphite under identical conditions. Carbonised pitch only yielded 2% fullerenes, therefore in the composite with coal in which pitch was only present at 20 wt%, the fullerenes were predominantly produced from the coal, since much higher yields were obtained from these composites. Interestingly, Loy Yang brown coal composite produced 7.7% yield which on a coal only basis is at least equivalent to, if not higher than graphite at 5.8%. However, brown coal could encounter difficulties in large scale fullerene production. Firstly, because of its high oxygen content, brown coal does not undergo thermoplasticity on heating, thus making it resistant to coking. Here, this was overcome by adding pitch as a binder to form a rod. Indeed, additional carbonisation time and temperature was need to obtain conductivity in brown coal. Secondly, the high oxygen content in brown coal (e.g. 25 wt% daf for Loy Yang coal) might oxidise any fullerenes during their formation in the arc. The data in Table 2 shows that after carbonisation, the Loy Yang coke contains less than 1% of heteroatoms which should have little effect on fullerene production. Although coking occurs readily at 1200 C, graphitisation does not usually occur below 2500 C on these time scales. These experiments show that graphitisation is not necessary for the production of fullerenes from coal. The use of d.c. rather than a.c. appears to markedly increase the yield of fullerenes from both graphite and coal. In conclusion, all cokes and pitch tested so far produce fullerenes. With the present cost of coke at about $500 per tonne or less, its use as an industrial source of fullerenes may greatly improve the economics of production.

Table 2. Carbon Content and Ash Yield for Coke^a Carbon Sample %C %ash yield %0+H+S+N^D

a wt % dry basis b by difference c denotes the level of demineralisation of the coal before coking

Table 3. Experimental Conditions and Yields of Crude Fullerenes from Coke and Graphite

* the fullerene generator was modified for these experiments, and the yield of fullerenes was reduced significantly but consistently. The second aspect of the present invention will now be described with reference to the following example. Example 1

Crude fullerenes (30g) in toluene were placed on a graphite powder (4kg) natural graphite 98-99% (Hopkins and Williams, Essex, England) and packed into a 60cm x 2cm column. The column was eluted with a toluene added to the graphite at a flow rate of lcc/min. C_fif) was isolated in the first 8 litres of solvent eluted. C_7f) was isolated after 12 litres was eluted. The yields were 26g and 3g respectively. Example 2

An Australian semi-anthracite coal from the Yarrabee mine in Queensland (H/C 0.74, 0/C 0.013) was ground, and size separated. The fraction that passed through a 125μm sieve was retained by a 63μm mesh sieve was separated and extracted in a Soxhlet for three days with toluene to remove any soluble material.

A glass column of 1.2 x 58cm dimensions was packed with the 63-125μm fraction using hexane and crude fullerene (15mg) was dissolved in toluene (3ml) and absorbed on the head of the column. After elution with hexane (800ml) at a flow rate of 2ml/min, 8.7 mg of solid was obtained from the eluate. This was identified as pure C_fir) by infra-red spectroscopy. Further elution with hexane or 10% toluene 90% hexane mixtures (80ml) did not yield distinctive magenta coloured solutions of C_βQ, and further elution was continued with toluene. Initial elution with toluene (200ml fraction) yielded C_βn, C_7n mixtures, but enriched in C_7n relative to the crude fullerenes. Further elution with toluene (200ml) produced fractions enriched in C_fin, suggesting at least two different binding sites for C_gQ. Recovery of fullerene was quantitative within experimental error. The used coal can be employed for further purification of _fi0/C₇₀ mixtures. Figures 4(a) and 4(b) show the effect of purifying f llerenes derived from Norwich Park Coal. The infra-red spectra show the characteristic absorption pattern of C_β0 with peaks at 528, 577, 1185 and 1429. Without wishing to be bound by theory, it is postulated that the ability of various coals to separate fullerenes may be because they consist of disordered sheets of aromatic condensed rings. Either, the pores are appropriately sized for separating fullerenes or there is

2 some interaction between sp hybridised structures in coal and fullerene. It is noteworthy that the ring current magnetic susceptibility of C_gfi and C_?0 differ considerably and C_7Q should behave more like a conventional aromatic molecule. Thus we might predict that C_7fi would be preferentially retained by the coal, as is observed in the initial stages of the separation.

The above results show that coal may be used to both produce fullerene and to separate pure fractions of the various fullerenes. Accordingly, an integrated process for the production and purification of fullerenes using coal is possible. The coal can first be used to produce fullerenes and also to separate the fullerenes into component fractions. Moreover, any byproducts or spent separation medium can be recycled to generate more fullerenes.

Those skilled in the art will appreciate that the invention described herein is susceptible to variations and modifications other than those specifically described. It is therefore to be understood that the invention includes all such variations and modifications which fall within its spirit and scope.

Claims

CLAIMS:-

1. A process for producing fullerenes from coal or like carbonaceous material which comprises: a) treating said coal or like carbonaceous material under conditions such that at least a portion of said coal or like carbonaceous material becomes electrically conductive; and b) causing an electric current to pass through the treated coal or like carbonaceous material to thereby volatilise at least a part of said coal or like carbonaceous material, step (b) being carried out in an inert, substantially oxygen free, atmosphere.

2. A process as claimed in claim 1 in which the inert atmosphere comprises helium, neon, argon, krypton or mixtures thereof.

3. A process as claimed in claim 1 in which the electric current is caused to pass through the treated coal or like carbonaceous material by causing the treated coal or like carbonaceous material to form an electrode in a system having at least two electrodes and applying a voltage across the electrodes to cause an electric arc to be generated between the electrodes.

4. A process as claimed in claim 1 in which the electric current is caused to pass through the treated coal or like carbonaceous material by subjecting it to electro-inductive heating.

5. A process as claimed in claim 1 in which the electric current is caused to pass through the treated coal or like carbonaceous material by passing an electric current directly through it.

6. A process as claimed in claim 1 in which the coal or like carbonaceous material is coal.

7. A process as claimed in claim 1 in which the coal or like carbonaceous material is selected from the group comprising mesophase, pitch, charcoal, petroleum tar and vacuum bottom tar.

8. A process as claimed in claim 6 in which the coal is selected from the group comprising brown coal and sub-bituminous coals.

9. A process as claimed in claim 1 in which the coal or like carbonaceous material is rendered electrically conductive by being heated to a temperature of from 350°C to 1800°C.

10. A process as claimed in claim 9 in which the temperature is from 1000°C to 1800°C.

11. A process as claimed in claim 9 or claim 10 in which the heating is carried on for a period of from 0.25 to 48 hours.

12. A process as claimed in claim 11 in which the period is from 1 to 25 hours.

13. A process as claimed in claim 1 in which soot condensed from the volatilised coal or like carbonaceous material is recovered and extracted with a suitable organic solvent to recover the fullerenes produced by the process.

14. A process as claimed in claim 13 in which the solvent is selected from the group comprising toluene, xylenes, 1,3,5-trimethylbenzene, 1,2,4-trichlorobenzene, 1-methylnapthalene, quinoline, benzene, pyridine, 1,2,3,5-tetramethylbenzene, hexane, heptane and mixtures thereof.

15. A method for separating a mixture comprising at least a portion of first fullerenes and a portion of second fullerenes, which process comprises extracting said mixture into an organic solvent, contacting said organic solvent and extracted fullerenes with a solid carbonaceous material to adsorb said fullerenes on said carbonaceous material, and eluting at least part of said portion of first fullerenes from said solid carbonaceous material.

16. A process as claimed in claim 15 in which the carbonaceous material is firstly eluted of the portion of first fullerenes and is then eluted of the portion of second fullerenes.

17. A process as claimed in claim 15 in which the carbonaceous material is selected from the group comprising coal and graphite.

18. A process as claimed in claim 17 in which the carbonaceous material is an anthracite or semi-anthracite coal.

19. A process as claimed in claim 15 in which the carbonaceous material is placed in a column.

20. A process as claimed in claim 15 in which the organic solvent is selected from the group comprising toluene, xylenes, 1,3,5-trimethylbenzene, 1,2,4-trichlorobenzene, 1-methylnapthalene, quinoline, benzene, pyridine,

1,2,3,5-tetramethylbenzene, hexane, heptane and mixtures thereof.

21. A process as claimed in claim 15 in which the fullerenes are eluted from the carbonaceous material using hexane and toluene mixtures as a solvent.

22. Fullerenes produced and/or separated by a method as claimed in any one of claims 1 to 21.

23. Fullerenes as claimed in claim 22 in which the fullerene is at least principally buckminsterfullerene.