Carbon nanostructures are fullerene-type structures and typically have at least one dimension in the nanometre range. They may be, but are not limited to, tubular structures, fibres, powders or a mixture of these morphologies. In all these structures the basic building blocks of the structure are six and five-membered carbon rings, joined in various configurations. Carbon nanotubes were first reported by Iijima (Nature 354, (1991), 56). They have a tubular structure, and may be single-walled or multi-walled.

Fibres may or may not be hollow, and the basal graphitic plane may be oriented along the length of the fibre or across the fibre, and even stacked conically. Many of the nanostructures and the factors controlling their formation and properties are discussed by Harris (Carbon Nanostructures and Related Structures, Cambridge 1999) and Dresselhaus et al (Science of Fullerenes and Carbon Nanotubes, Academic, 1996).

Carbon nanostructures can be synthesised by methods such as arc discharge, laser ablation and chemical vapour deposition (CVD). Many of the synthetic methods use a catalyst to promote the growth of the nanostructures. In CVD methods, a carbonaceous gas such as C2H2, CH4, or CO is decomposed by a catalyst. The synthesis of nanostructures by CVD methods is a promising mass production method because the temperature required for nanostructure growth is around 700°C, which is much lower than in arc discharge or laser ablation. Takiwara et al (Jpn. J. appl. Phys. Vol. 39 (2000), 5177-5179) disclose a CVD method that uses nickel and zinc deposited on copper substrates to produce carbon"nanocoils". Wei et al (J. Vac. Sci. Technol. B, 18 (6), 3586-3589) disclose the growth of carbon nanotubes wherein the catalyst comprises a silica substrate, a thin (lOnm) titanium layer and iron particles deposited by electron-gun evaporation.

Li et al. (Chemical Physics Letters, 335, (2001), 141-149) disclose a CVD method for the controlled growth of carbon nanotubes wherein the catalyst comprises graphite foil coated with a stainless steel film. The film is applied using magnetron

sputtering and is 2-lOOnm thick. Annealing in hydrogen converts the stainless steel film to stainless steel particles.

Dong et al. (Jn of Zhejiang Univ., 34, (2000) pp. 511-514, English abstract) describe the growth of carbon nanotubes on nano-metallic materials including stainless steel. Such materials are difficult to produce and expensive.

There are numerous potential applications of carbon nanostructures. They have considerable strength and flexibility and are potentially useful as reinforcing elements in composites. Additionally, they are electrically conductive and may be useful in nano-scale electronic circuits. Finally, they have interesting hydrogen storage properties, and may find application as hydrogen storage media (Seung Mi Lee et al, Synthesis Metals 113, (2000), 209-216). It is possible that an ideal structure for hydrogen storage has many basal plane edges at the surface because these may be suitable sites for attachment of hydrogen. Therefore a conically stacked nanofibre may be a suitable structure for hydrogen storage.

There is a need for processes for the synthesis of carbon nanostructures that can be applied on an industrial scale. The processes should produce large yields of nanostructures with reproducible properties. To be economically attractive, the processes should be simple and should not require extremes of pressure or temperature, or expensive reagents.

The present inventors have devised an improved process for the synthesis of carbon nanostructures. In accordance with the present invention, a process for the synthesis of carbon nanostructures comprises contacting a carbon source with a catalyst such that carbon nanostructures are grown on the catalyst, characterised in that the catalyst comprises stainless steel flake.

In the context of the present invention, the term"catalyst"is used to describe a finely divided material on which reactions can take place and where nucleation and growth can proceed. It is not intended that the term'catalyst'be solely used in the conventional sense, that is, there is no requirement in the present invention that the catalyst be unchanged during the process of the invention.

The term"stainless steel flake"is used to describe an iron alloy in particulate form where the particles have one physical dimension (minor dimension) which is substantially smaller than the other two physical dimensions. The flake suitably has a minor dimension (or thickness) of 0.1-1 urn, preferably about 0. 6 urn. As is known in the art, stainless steels are those alloy steels which include appreciable amounts of chromium, for example 12% and above. Typically, the alloys may contain up to 25% chromium, up to 15% nickel, up to 1.5% carbon, and small amounts of other elements such as niobium, titanium, aluminium and copper, with the balance being iron. A preferred composition the stainless steel flake is designated 316L, which is a common grade of stainless steel of low carbon content. Alternative designations of stainless steel may also be used as can mixtures of two or more types of stainless steel. Suitable 316L stainless steel can be purchased from Novamet Speciality Products Corporation in the form of flakes with a minor dimension of ca. 0.6 llm, the other two dimensions of the flake being in the range 10-100 urn.

To form the supported stainless steel catalyst as reported by Li et al. , it is necessary to sputter commercial stainless steel onto a graphite foil substrate.

The sputtering is carried out in a vacuum chamber at 300°C. Following the sputtering step, the stainless steel film is reduced at 660°C in 50Torr of flowing hydrogen and nitrogen for two hours. It is stated that this step is needed to obtain catalyst particles and enhance the catalytic activity. This catalyst is unlikely to be used for mass production of carbon nanostructures due to the multiple steps and combination of low pressure and high temperature required to provide the catalyst. By contrast, the process of the present invention does not require multiple steps or careful pressure and temperature control to provide a catalyst that will promote nanostructure growth. Stainless steel flake is advantageously employed to provide the catalyst.

Preferably, the carbon source comprises a carbonaceous gas, a carbonaceous liquid or any mixture thereof. Some non-limiting examples include carbon monoxide, methane, natural gas and other gaseous or liquid hydrocarbons. The carbon source may be provided substantially pure or be diluted or admixed with an inert or reactive carrier.

Several methods can be used to carry out the process of the present invention. In a first method, a small quantity of stainless steel flake is placed in a tube, preferably a quartz tube. The tube is rotated to distribute the flake on the surface of the tube. The quantity of stainless steel flake used will depend on factors such as the size and shape of the tube and the gas flow rate. A carbon source gas e. g., CO, CH4 or similar hydrocarbon is passed over the flake at temperatures between 475°C and 1000°C and carbon nanostructures are grown on the flake. The reaction time is dependant on gas flow rate and temperature. Preferably, the carbon source gas is diluted with hydrogen.

In a second method, the stainless steel flake is fluidised in fluidised bed system.

A carbon source gas and optionally hydrogen are passed into the bed. The carbon nanostructures grow on the flake. It is possible that some of the nanostructures may become detached from the flake during the reaction and be contained within the gas flow from the bed.

In a third method the stainless steel flake is admixed with a carbonaceous liquid and the mixture contacted with a heat source. For example, the flake may be blended into an organic liquid such as xylene, and sprayed into the hot zone of a furnace. The organic liquid acts as a carbon source so it is not necessary to add a carbon source gas.

Several factors will influence the nature of the carbon nanostructures produced, including the process method used, the selection of the carbon source and the temperature. Advantageously, the process of the present invention may be carried out under substantially atmospheric pressure. This simplifies the process and leads to further cost reductions when compared to e. g. CVD methods for nano-structure synthesis.

Although not as preferred, high and low pressures can also be used if desired.

In an alternative embodiment of the invention the stainless steel flake is supported on a support. The supported stainless steel catalyst as reported by Li et al is produced by sputtering commercial stainless steel onto a graphite foil substrate. By contrast, stainless steel flake can be deposited on a support by simply bringing the flake into contact with the support. The support may be a heat resistant support such as a honeycomb support made of, for example, cordierite. Alternatively, the heat resistant support may be a heat exchanger, which may be made of surface oxidised metal or high

temperature plastic. If nanostructures are grown on the stainless steel flake deposited on the support, then the support may be a useful hydrogen storage device. Therefore the present invention also provides a process for producing a hydrogen storage device comprising the steps of depositing stainless steel flake on a support and synthesising carbon nanostructures on the flake.

Stainless steel is supplied to the process of the invention as stainless steel flake.

However, it is possible that the form of the stainless steel will change during the process and the catalyst may not actually be stainless steel in the form of the original flakes, but may be stainless steel in a different form. Some change of shape is likely to occur particularly at more elevated temperatures during processing as the flake reduces its surface energy.

The carbon nanostructures grow on the surface of the stainless steel catalyst.

Depending on the intended application of the carbon nanostructures, it may be necessary to remove the nanostructures from the catalyst. One suitable method for the removal of the nanostructures from the catalyst is boiling in hydrochloric acid, although other methods will be apparent to those skilled in the art. For most applications however, it is unlikely to be necessary to remove the nanostructures from the catalyst. Stainless steel is inert in many chemical environments, so the presence of stainless steel is unlikely to interfere with the functioning of the nanostructures.

In a further aspect, the present invention provides carbon nanostructures as produced by the novel process according to the invention.

In a yet further aspect, the present invention provides the use of carbon nanostructures as produced by the novel process according to the invention, as hydrogen storage media.

The invention will now be described by way of example only, which is not intended to be limiting thereof.

Example 0.020g of 316L stainless steel flake from Novamet Speciality Products Corporation (0.6 um thick, other dimensions 40-100 um) was placed in a quartz tube.

The tube was rotated to distribute the flake on the surface of the tube. The tube was heated to a temperature of between 475°C and 800°C and a gas mixture of a carbon source and hydrogen was passed through the tube at a flow rate of approximately 600ml per hour for 24 hours (unless otherwise stated). At the end of this period the tube was cooled first under hydrogen and then under nitrogen.

Samples of carbon nanostructures were analysed by Transmission Electron Microscopy (TEM) using a Philips EM400T electron microscope operating at an accelerating voltage of lOOkV. Table 1 shows the results of several experiments using different gas mixtures and different temperatures.

Significant yields of carbon nanostructures can be produced by the method of the present invention, particularly when CO is used as the carbon source. The choice of conditions affects the size and morphology of the nanostructures that are produced.

Table 1 Gas Mix Gas Ratio Temperature Yield (g) Observations by TEM (°C) CO: H2 1: 1 550 0.653 20um helical/straight closed ended fibres. 2nm inner diameter; 30nm outer diameter. CO: H2 1: 3 550 0.232 6 urn hollow fibres. 3mn inner diameter; lOOnm outer diameter. CO: H2 1: 4 550 0.484 Fibres, closed and open ended. 5nm outer diameter. CH4: H2 1 : 1 800 0. 063 Mass of clusters which are mainly graphitised carbon. Some nanotubes. CH4: H2 2 : 1 800 0. 075 Some thick walled nanotubes. Mainly clusters of graphitised carbon. CH4: H2 1 : 3 550 0 No data CH4 : H2 1: 3 800 0.096 No data Natural 1: 3 800 0.117 Coarse irregular tubes with gas: H2 outer diameter from 24mn to 180nm and internal diameter 4 to l6nm. Concertina like fibres with diameters from 60mn to 210nm. Smooth straight tubes with outer diameter of 35non and inner diameter of 20nm tCO : H2 1 : 2 550 0.89 Long open and closed ended carbon nanotubes, 5ntn inner diameter. The tube walls appeared to have collapsed forming cavities along the length of the carbon fibre. CO : H2 1: 3 550 1.225 6u. m fibres with some close- ended nanotubes. 3nm inner diameter and 80nm outer diameter. tCH4-H2 1: 3 475 0 No data 1-2 700 0. 75 Carbon nanotubes with 4nm inner diameter and 70nm outer diameter. t Reaction for 48 hours, 2-Reaction for 18 hours.