Background to the Invention The production of hydrogen from hydrocarbon fuels is an industrially important process. Known methods for hydrogen production include thermal decomposition, which is also known as cracking or pyrolysis, and steam reforming.

Hydrocarbon fuels are cracked via the application of high temperatures to produce hydrogen and carbon- containing by-products. The cracking reaction is highly endothermic and, therefore, generally requires a long reaction time. This drawback can be alleviated to a certain extent by adding water or oxygen to the hydrocarbon fuel feed. In general, however, thermal decomposition is a very energy-intensive process and has the additional disadvantage that large amounts of soot and heavy hydrocarbon by-products are produced, which create problems during the recovery of the desired product (s).

In addition, thermal decomposition is usually performed in the presence of a catalyst which can become poisoned by the particulate and heavier hydrocarbon by- products and by the impurities that may be present in the fuel. Thus, catalyst wastage and replacement and disposal of the spent catalyst increases the overall cost of this type of hydrogen-producing process.

Steam reforming involves the reaction of hydrocarbon fuel and water to produce carbon monoxide and hydrogen.

Like thermal decomposition, steam reforming is also very energy-intensive, requiring high operational temperatures. Also, like thermal decomposition, steam reforming is typically performed in the presence of a catalyst. Therefore, the disadvantages mentioned above

in relation to the use of catalysts in cracking apply equally to their use in steam reforming processes.

An alternative to thermal decomposition or steam reforming of hydrocarbon fuels is partial oxidation of the fuel in an oxygen-deficient environment such that the reaction products will include hydrogen and carbon monoxide in addition to the products of complete combustion, which are carbon dioxide and water. Indeed in extreme cases, partial oxidation may result in all of the carbon in the hydrocarbon fuel being transformed into carbon monoxide and all of the hydrogen being transformed into hydrogen gas.

Like thermal decomposition and steam reforming processes, however, partial oxidation processes require high operating temperatures. The operating temperature may be lowered via use of a catalyst but, as previously discussed, the use of catalysts has drawbacks, in particular in relation to the overall cost of the process. An additional disadvantage of partial oxidation processes is the requirement for feeding oxygen to the reaction. This requirement adds cost to the overall process because either pure oxygen must be supplied, which is very expensive, or, if air is used as the source of oxygen, energy is required to be fed to the process which also adds to the overall cost of the process.

However, of the three above-described processes for the production of hydrogen, partial oxidation is the only exothermic reaction. Therefore, it is advantageous over the use of thermal decomposition or steam reforming as the amount of energy input required is lower than that required in thermal decomposition or steam reforming processes. Accordingly, the cost of hydrogen production via partial oxidation is proportionally lower.

Partial oxidation may be achieved via use of superadiabatic combustion (SAC) technology. One instance in which this will occur is when a gaseous mixture of a fuel and an oxidant is fed to a porous material which may be non-catalytic and which has a high thermal capacity.

When the gaseous mixture is ignited, the energy produced

via combustion of the mixture is retained within the porous material via heat exchange between the gaseous mixture and the surfaces of the material. This stored energy enables attainment of flame temperatures which are substantially higher than the adiabatic temperature of the gaseous mixture fed to the system. Accordingly, the combustion of fuel subsequently fed into the porous material may be achieved without the need for substantial, additional external heating of the porous material.

US 2002/0020113 Al discloses an SAC process which may be used to produce hydrogen from fuel-rich mixtures.

The process comprises creating a transient thermal combustion wave which is confined and controlled within a porous fixed bed reactor by reversing or counter-flowing the flow of gaseous reactant mixture through the reactor.

By necessity, this involves somewhat complex methodology.

In addition, the reactor disclosed in US 2002/0020113 Al is bulky, not least because of the ducting required to facilitate reversal or counter-flow of the reactant mixture through the reactor. Therefore, it would be desirable to take advantage of SAC technology but in an overall more straightforward manner than has been achieved in the art. In particular, it would be desirable to provide a compact SAC device for the production of hydrogen from fuel-rich mixtures, which could therefore be used in a variety of applications, including for hydrogen enrichment purposed in vehicles or gas turbines, for example.

Summary of the Invention According to a first aspect of the present invention, a method of producing hydrogen from a fuel- rich mixture comprising oxygen and a hydrocarbon fuel comprises providing a device comprising a first porous ceramic layer and a second porous ceramic layer; feeding the mixture into the first porous layer of the device; and combusting the mixture within the device to produce hydrogen, wherein the average pore size of the first porous layer is smaller than the average pore size of the

second porous layer, and the thickness of the second porous layer, in the direction of feeding the mixture, is at least equal to the thickness of the first porous layer.

According to a second aspect of the present invention, a method of generating power comprises producing hydrogen by a method as defined above and feeding the hydrogen so-produced to a fuel cell.

According to a further aspect of the present invention, a power generation unit comprises a device as defined above, connected to a fuel cell which in turn generates power through combustion of hydrogen. Such power generation units may be used to provide power in a wide variety of applications including powering vehicles and also domestic or industrial premises such as buildings, plants and sites.

According to a further aspect of the present invention, a hydrogen-enrichment unit comprises a device as defined above, connected to an internal combustion engine or gas turbine to enrich the fuel with hydrogen and hence improve the combustion characteristics of the engine or gas turbine. Such techniques may be used in power generation or mobile applications.

The present invention therefore provides a straightforward, efficient and cost-effective method for the production of hydrogen via partial oxidation of a fuel-rich mixture comprising oxygen and a hydrocarbon fuel. Furthermore, the method of the invention does not require the combustion zone within its SAC device to be confined and maintained via reversal or counter-flow of the fuel mixture into the device, but rather achieves this through the structure of the device itself.

Description of the Invention The method according to the present invention utilises a relatively simple device composed of porous ceramic material (s), to produce hydrogen from a variety of fuel-rich mixtures comprising oxygen and a hydrocarbon fuel. Through the choice of appropriate ceramic materials, and specifically their pore size, combustion

occurs within the device, with the flame stabilised around the interface between the first and second porous layers.

In the context of the present invention, the term "hydrocarbon fuel"means a compound or mixture of compounds which may be burned as a fuel and which contain at least carbon and hydrogen in their molecular make-up.

Examples of suitable hydrocarbon fuels include straight and branched chain hydrocarbons such as alkanes and alkenes containing 1-40 carbon atoms, specific examples of which include methane, heptanes and octanes; straight and branched chain alcohols containing 1-40 carbon atoms, such as methanol; and hydrocarbon mixtures such as liquefied petroleum gas (LPG), gasoline, kerosene, naphtha, petrol and diesel oil, for example those commercially available from petrol stations.

Additional examples of hydrocarbon fuels useful in the present invention include biologically-derived fuels.

These fuels are derived from renewable sources, for instance animal and/or vegetable oils. One example of such a fuel is available under the trade name Biodiesel, and is composed of mono-alkyl esters of long chain fatty acids. The biologically-derived fuels may be used alone or blended with more conventional hydrocarbon fuels.

Typically, the liquid hydrocarbon fuels will be vaporised prior to mixing with oxygen.

Furthermore, in the context of the present invention, the terms"rich"or"fuel-rich", as applied to a mixture comprising hydrocarbon fuel and oxygen, mean a mixture in which the oxygen is present in an amount which is insufficient to bring about complete combustion of all of the fuel in the mixture to produce carbon dioxide and water. Similarly, the terms"lean"or"fuel-lean", as applied to a mixture comprising hydrocarbon fuel and oxygen, mean a mixture in which the oxygen is present in an amount in excess of that required to bring about complete combustion of all of the fuel in the mixture.

The term"stoichiometric"as applied to a mixture comprising a hydrocarbon fuel and oxygen, means a mixture comprising fuel and the exact amount of oxygen required to bring out complete combustion of all of the fuel in the mixture. The stoichiometry of a fuel-containing mixture may also be defined in terms of an equivalence ratio which is typically identified in the art by the symbol . The equivalence ratio equates to the amount of oxygen required for complete combustion of a particular fuel divided by the amount of oxygen actually present in the combustion system. Thus, in a fuel-lean environment, will be less than 1, in a stoichiometric environment will be equal to 1 and in a fuel-rich environment will be greater than 1.

Turning to the device itself, this comprises first and second ceramic porous layers, each of which may take a variety of different forms. For instance, each layer may independently be selected from single, monolithic structure such as a ceramic foam or a ceramic honeycomb- structured material, or a plurality of discrete ceramic elements, such as ceramic beads, fibres or mats of fibres, positioned within the respective layer such that pores are created therein. For example, if ceramic beads are employed, they are preferably close-packed.

Preferably, however, the porous layers independently comprise a ceramic foam or ceramic beads, and more preferably both layers comprise either a ceramic foam or ceramic beads.

Furthermore, each"layer"may itself be composed of a number of sub-layers, for instance according to the commercially-available materials.

In order to stabilise the flame in the device at around the interface between the first and second porous layers, and thus achieve efficient SAC of the fuel-rich mixture, the pore size of the layers must be different and in particular the average pore size of the first layer should be smaller than the average pore size of the

second layer. In this context, the average pore size is the average size of the voids comprised in the layer, and all references to an average should be understood to be the mean.

Suitable pore sizes of the first and second layers may be achieved in a number of ways. For instance, if the first and second porous layers each comprise a ceramic foam, the average diameter of the pores in the foam making up the first layer is preferably less than or equal to 0.5 mm, and the average diameter of the pores in the foam making up the second layer is preferably in the range 1-4 mm.

Alternatively, if the first and second porous layers comprise ceramic beads, the average diameter of the beads in the first layer is preferably in the range 1-3 mm, while the average diameter of the beads forming the second layer is preferably in the range 3-5 mm.

The porosities of the layers are not critical, but are typically similar to one another. The porosity of both layers is preferably higher than about 25%, and is more preferably in the range 50% to 95%. In this context, porosity is defined by the relative densities of the layers and, more specifically, by the percentage volume of voids comprised in the total volume of each layer.

Accordingly, the greater the porosity of a layer, the greater its percentage volume of void and the lower its density.

The porous layers may comprise any ceramic material which is known in the art to be suitable for use in high temperature applications such as fuel combustion processes. Examples of suitable ceramic materials include alumina, mullite, zirconia-toughened mullite, cordierite, silicon carbide, zirconia, yttria-stabilised zirconia, and mixtures thereof. Preferably, the porous layers comprise alumina or cordierite, and most preferably both layers comprise alumina.

The first and second porous layers in the device are preferably in contact with one another, achieved for instance by placing one layer on top of or adjacent the

other. Other suitable means of connecting the layers may be envisaged.

Typically, the layers of the device are retained or mounted in an open-ended structure, such as a tube, made from a material which can withstand the temperatures normally employed in fuel combustion, for example quartz.

The perimeter of the tube provides a close fit around the perimeter of the layers. If required, the layers may be held tightly in position within the tubular structure via means of a high temperature-resistant deformable material, such as alumina film.

In use, the device is generally positioned such that the fuel-rich mixture flows in a horizontal or vertical direction into and through the first porous layer of the device, and the thicknesses of the different porous layers are defined in this direction. More specifically, the thickness of the first porous layer in the device is at least equal to the thickness of the second porous layer. Preferably, however, the thickness of the second porous layer is greater than the thickness of the first porous layer, and more preferably the thickness of the second porous layer is at least twice the thickness of the first porous layer.

The actual thicknesses of the layers are not crucial as long as they are at least equal. However, the thickness of the first layer may typically be in the range 0.5-5 cm, and preferably in the range 1-2 cm, while the thickness of the second layer may generally be at least 0.5 cm, preferably in the range 1-10 cm and more preferably in the range 2-8 cm. Most preferably, the thickness of the first layer is approximately 2 cm and the thickness of the second layer is approximately 4 cm.

During operation of the method of the invention, a mixture comprising oxygen and a hydrocarbon fuel is fed into the device through its first porous layer, and a flame is then ignited at the surface of the second porous layer not in contact with the first porous layer.

Alternatively, ignition can occur via igniters located at the interface of the two porous layers.

Because the velocity of the fuel-rich mixture is low if SAC is not achieved, this warm-up phase is important to create SAC combustion of fuel-rich mixtures and leads to combustion occurring at the interface. The relative movement of the flame and the fuel-rich mixture within the device is regulated and optimised through the choice of material used to form the porous layers, as described above.

The mixture is fed into the device typically at a velocity in the range 0.1-10 m/s, and preferably in the range 0.1-1 m/s.

In a preferred embodiment, the equivalence ratio of the mixture of fuel and oxygen initially fed into the device is approximately 1, ie. a stoichiometric fuel/oxygen mixture, or lower, ie. a fuel-lean mixture.

This also aids establishment of a stable flame within the device at the interface of the two layers. Once the flame has stabilised, which typically occurs after approximately 30 seconds to 5 minutes, the fuel/oxygen equivalence ratio is switched to rich, ie. becomes greater than 1.

Ultimately, the hydrogen produced and any by- products, such as carbon monoxide, carbon dioxide, water vapour and lower hydrocarbons such as methane, acetylene, ethylene or butane, exit the device via the second layer.

The amount of hydrogen produced can be regulated by the velocity of the mixture fed into the device, the pressure and, in particular, the cross-sectional area of the device. Typically, the cross-sectional area of the device is in the range 0.001 to 1 m2, and more preferably in the range 0.001 to 0.01 m2.

The device is typically operated at atmospheric pressure, but may be operated at super-atmospheric pressures such as 1-10 bar. Temperatures within the device during hydrogen production according to the invention are typically in the range 800-2500 °C, and preferably in the range 1000-1700 °C.

Typically, the device is capable of generating between 1,000 and 15,000 kW/m2 of the cross-sectional area of the device.

The hydrogen produced may be used in a variety of applications, for example it may be fed to a fuel cell.

Fuel cells are, of course, found in a wide variety of devices such as vehicles engines and domestic-or industrial-scale power generation units. It is envisaged that the method of the present invention, when the above- described device is connected to a fuel cell, will find particular use in vehicle engines or domestic heat and/or electricity generators in a variety of locations including domestic and industrial premises such as buildings, plants and sites, for example houses, flats and factories.

At present, a significant amount of worldwide research is being carried out into the possibility of powering cars and other vehicles using fuel comprising at least some hydrogen via so-called"on-board hydrogen generation"technology. Hydrogen fuel, when combusted, of course creates much less noxious by-products than conventional hydrocarbon fuels, as the main product of hydrogen combustion is water. Therefore, there is a real incentive to facilitate the use of hydrogen fuel in all types of combustion systems in order to reduce the environmentally unfriendly emissions of conventional combustion systems, which predominately use hydrocarbon and fossil fuels.

Accordingly, hydrogen produced by the method of the invention may be used to enrich fuel fed to an internal combustion engine or gas turbine thus improving their combustion characteristics. This is achieved via use of a hydrogen-enrichment unit comprising the device previously described connected to the internal combustion engine or gas turbine.

The present invention is now described with particular reference to the accompanying drawings and the following examples.

Description of Figures Figures 1 and 2 are cross-sectional views through representations of devices employed in the present invention.

Figure 1 shows a first porous ceramic layer 1 in contact with a second porous ceramic layer 2, the first layer having a smaller average pore size than the second layer. Both layers comprise alumina foams, and are tightly held within a quartz tube 3 via means of alumina film 4, which is wound around the circumference of the layers.

Figure 2 shows a first porous ceramic layer 11 which comprises alumina beads of smaller diameter and which is in contact with a second ceramic porous layer 12, which comprises alumina beads of a larger diameter. The layers are held within a quartz tube 13 via means of alumina film 14, which is positioned around the perimeter of layers 11 and 12. The beads are additionally kept in position via means of a metal wire mesh 16.

The direction of flow of the fuel/oxygen-containing mixture through the devices of Figures 1 and 2 is indicated by the arrows in the Figures. In use, SAC of a fuel-rich mixture comprising oxygen and a hydrocarbon fuel is achieved within the device as previously described, with the flame stabilised around the interface 5,15 between the two porous layers.

Examples The devices of Figures 1 and 2, identified as Device 1 and Device 2 respectively, were assembled as summarised in the following table and then run under the conditions given in the table to produce hydrogen. DEVICE 1 DEVICE 2 Type of porous medium Alumina Packed bed foam with alumina beads Pores per inch of thickness, 60 first layer Pores per inch of thickness, 20- second layer Bead diameter, first layer-2-3 (mm) Bead diameter, second layer-4-5 (mm) Burner inner diameter (mm) 30.4 30.4 Length of first layer (mm) 20 20 Length of second layer (mm) 40 80 Fuel Methanol Methanol Equivalence ratio, 4 6 Gas Feed velocity (m/s) 0. 12 0. 15 Volumetric composition of products after ceramics (%) : H2 24 35 CO 21 20 C02 3 5

The alumina foams were manufactured by Selee Corporation Inc. (USA) and sold under the name"ceramic foams"and the beads were manufactured by Fluka Chemika Gmbh, product"Aluminium oxide", code 06400.