PATHWAY

The invention relates to a method of forming a coating within an internal pathway.

There are many known applications where it is desirable to form a coating within an internal pathway. In one such application, a porous solid coating is formed on an internal surface of a tube so as to make a column that may be used for chromatography. In another such application, a catalytic coating is formed on an internal surface of a channel so that chemical reactions may be performed within the channel. In yet another such application, a catalytic coating may be formed on an internal surface of a porous body so as to allow chemical reactions to be performed within the porous body.

A first known method for forming a coating within an internal pathway of a body involves deposition of solid particles from a colloidal sol which fills the internal pathway. The liquid continuous phase is evaporated by heating the body in an oven leaving the solid particles which form the coating. However, this method is not satisfactory. The rate of evaporation is generally very slow. The internal pathway may become blocked by plugs of the solid particles. Also, ejection of the sol may occur if the liquid continuous phase boils.

A second known method is disclosed in US 2004/0033319 Al. This document teaches a method of forming a coating on the internal surface of a tube by providing a thin film of liquid covering the internal surface. The liquid contains an organic metal compound which decomposes to form the coating on heating. As the liquid just covers the internal surface, rather than filling the internal pathway, the surface area of the liquid is increased and the rate of evaporation is increased. In addition, as the internal pathway is not filled, the problem of ejection of liquid caused by boiling of the liquid is reduced. Nevertheless, this method still suffers from drawbacks. For example, the coating may be uneven if the liquid from which the coating is formed does not cover the internal surface evenly.

In accordance with an aspect of the current invention, there is provided a method of forming a coating within an internal pathway, comprising: providing a body having an inlet and an outlet and an internal surface which defines an internal pathway extending within the body between the inlet and the outlet; streaming a mixture of a gas and a fluid along at least a part of a length of the internal pathway, the fluid comprising one or more substances for forming a solid coating on the internal surface, the fluid being a liquid solution of said one or more substances in a solvent or being a dispersion with at least one of said one or more substances being solid particles dispersed in a liquid continuous phase; during said streaming of the mixture, applying localised heat progressively along said at least a part of the length of the internal pathway; and wherein said progressive application of localised heat causes, within said at least a part of the length of the internal pathway, formation from the one or more substances of a solid coating on the internal surface.

While the current invention should not be interpreted as necessarily overcoming the defects of the prior art, one or more advantages may be achieved as discussed in the following description.

The streaming of the mixture of the gas and the fluid may advantageously reduce or prevent the formation of an insulating vapour layer (i.e. the Leidenfrost effect), which can occur in prior art methods on heating, for example, between an internal surface and a sol filling an internal pathway, or between an internal surface and a layer of liquid covering the internal surface. Reduction or prevention of the Leidenfrost effect may advantageously increase heat transfer from the body to the one or more substances which form the coating.

The current coating method may proceed relatively rapidly and the ejection of liquid caused by boiling may be reduced or prevented. Formation of solid plugs in the internal pathway may also be reduced or prevented.

Moreover in preferred embodiments, the streaming of the mixture of the gas and the fluid, and the application of localised heat progressively along the length of the internal pathway, may interact synergistically to give a coating which tends to have more constant properties (such as thickness and composition) both along the length of the internal passageway and also in the cross-sectional direction of the internal passageway. If, hypothetically, the heat was applied to the whole length of the internal pathway simultaneously, the one or more substances which form the coating would become depleted as the mixture progressed along the internal pathway. This could, for example, lead to a coating that was thicker upstream as compared to downstream, or which varied in composition along the length. By application of localised heat progressively along the length of the internal pathway, there may be an improved homogeneity of the mixture of the gas and the fluid, at all locations along the length of the internal pathway, when compared at the respective points in time when they are heated. In other words, when any location along the internal pathway is being subjected to the localised heating, that location is preferably exposed to a portion of the mixture of the gas and the fluid that has not been substantially depleted of coating substance (or substances) by previous upstream heating of the mixture. In addition, the use of the mixture of the gas and the fluid may lead to an improved homogeneity of the coating in the cross- sectional direction of the internal pathway, especially if the streaming of the mixture reduces or eliminates the Leidenfrost effect and/or ejection of liquid due to boiling and/or uneven covering caused by movement of liquid under gravitational pull.

A substance for forming a coating may preferably be present in the fluid at a concentration of between 0.001 to 70wt%, and more preferably at a concentration of between 0.1 to 30wt %.

In many instances, the current method can be used to prepare relatively thick coatings. Preferably, the coating has a thickness of at least 1 pm, and more preferably at least 2 μητι, or at least 5 pm, or at least 10 pm, or at least 15 pm or at least 20 pm, in the direction of the cross-section of the internal pathway. In addition, the current invention may allow formation of a coating extending along a relatively long internal pathway. Preferably, the internal pathway is coated over a length of at least 0.1m. More preferably, the internal pathway is coated over a length of at least 0.2m, or at least 0.5m, or at least lm, or at least 5m, or at least 10m.

In one embodiment, the internal pathway is a channel and the internal surface forms the perimeter of the channel. For example, the body may be a tube with the inlet and the outlet being provided at opposite ends of the tube. The channel is the internal channel or lumen of the tube and the internal surface is the internal surface of the tube.

Alternatively, the body may have a plurality of channels extending through it between the inlet and the outlet. In this case, the body may have a plurality of internal surfaces, with each internal surface forming the perimeter of a respective one of the channels. In the case that there is a plurality of channels, the inlet may be a formation which connects the channels, such as a manifold or chamber. Alternatively, the inlet may simply be the openings of the individual channels when considered collectively. Likewise the outlet may be a formation which connects the channels, or simply channel openings when considered collectively.

Channels may be rectilinear but do not need to be.

A channel may have any practical cross-sectional shape. For example, a channel may be circular in cross-section. In a body which has a plurality of channels, the channels do not need to have the same cross-sectional shape and/or size as one another. The cross-sectional shape and/or size may vary along the length of a channel.

A channel will have a maximum cross-sectional dimension. This is simply the largest dimension extending in a straight line across the channel in a cross-sectional direction. Hence, when the channel has a circular cross-section, the maximum cross-sectional dimension will be a diameter. When the channel has a square cross-section, the maximum cross-sectional dimension will be a diagonal extending from one corner of the square to the opposite corner. If the cross-section of the channel is constant along the length of the channel, then the maximum cross-sectional dimension will be present all along the length of the channel. Alternatively, if the channel cross-section varies along the length of the channel, then the maximum cross-sectional dimension may only occur at a certain point or points along the length of the channel. Preferably, a channel has a maximum cross-sectional dimension of less than 10 mm. In this case, the maximum cross-sectional dimension may be less than 5 mm, less than 4 mm, less than 3 mm, less than 2 mm, less than 1 mm, less than 0.5 mm, less than 0.4 mm, less than 0.3 mm, less than 0.2 mm or less than 0.1 mm.

In another embodiment, the body is porous and the internal pathway is formed by a plurality of interconnecting internal spaces (that is to say the pores) within the body. The interconnecting internal spaces are defined by the internal surface. In this case, the body preferably includes an impermeable portion which defines the perimeter of the internal pathway between the inlet and the outlet. For example, the impermeable portion may be an impermeable tube with the inlet and the outlet at opposite ends of the tube and the body may include at least one porous member within, and preferably filling, the impermeable tube. In this case, the entirety of the surface of the porous member (or the combined surfaces of the porous members), including internal surface areas bordering internal spaces, forms the internal surface of the invention. In such an embodiment, the impermeable tube forms the perimeter of the internal pathway so as to contain the flow of the mixture of the gas and the fluid. The length of the internal pathway lies along the length of the impermeable tube.

As discussed above, both the streaming of the mixture of the gas and the fluid and the progressive application of localised heat, take place over at least part of the length of the internal pathway (and preferably over substantially the whole length of the internal pathway). Preferably the body is elongate - i.e. it has a length greater than its thickness. In this case, the progressive application of localised heat to the internal pathway may comprise progressive application of localised heat to the external surface of the body along at least part of the length of the body. The part of the length of the body that is subjected to progressive application of localised heat (or all of it as the case may be) is co-extensive with the part of (or all of) the internal pathway that is subjected to progressive application of localised heat. This is readily visualised when the body is a tube. The progressive application of localised heat along the internal channel of the tube is effected simply by applying heat in a localised manner to, and progressively along, the external surface of the tube.

Where the body is elongate, localised application of heat to the external surface of the body may be effected by a heater which surrounds a portion of the length of the body so that heat is applied to the external surface of the body all around the portion. There is progressive relative movement between the body and the heater so that successive portions of the length of the body are surrounded by and heated by the heater. For example, the heater may be an oven and successive portions along the length of the body may be progressively moved into, through and out of the oven (or other heater). Alternatively the heater may be annular, surrounding the body, and either the heater or the body (or both) may be moved to achieve the progressive relative movement.

In general, although not exclusively, the mixture of the gas and the fluid will be introduced into the internal pathway though either the inlet or the outlet for the streaming.

As discussed above, the fluid comprises one or more substances which are suitable for forming the coating. The substance (or if there are more than one such substance any one of them) may be of a type which forms the coating while retaining the same chemical composition in the coating. Alternatively, the substance (or if there are more than one such substance any one of them) may be of a type which undergoes a change in chemical composition on heating to form the coating. If there are more than one such substance, they may be of the same type - i.e. all retaining the same chemical composition or all undergoing a change in chemical composition. Alternative, they may be of different type - for example one substance which retains the same chemical composition and another which undergoes a change in chemical composition.

In a preferred embodiment, the substance suitable for forming the coating is a solute. Alternatively, if there are a plurality of substances for forming the coating, one, some or all of them may be a solute or respective different solutes. This applies when the fluid is either a liquid solution, or a dispersion of solid particles in a liquid continuous phase. In the former case, the liquid solution comprises the solute (or solutes) and a solvent. In the latter case, the liquid continuous phase of the dispersion comprises the solute (or solutes) and a solvent. Substances in the form of solutes may undergo thermal decomposition to form decomposition products during the localised heating of the internal pathway. The coating then comprises the decomposition product(s). When the fluid is a dispersion of solid particles and the dispersion comprises a solute, the solid particles in the dispersion may also form part of the coating, together with the decomposition product of the solute.

In one example in which the fluid contains a solute, the solute comprises a metallic cation and the solute forms a decomposition product in the form of a metal oxide on heating. The solute may be a metal salt. The metallic cation may be selected from the group consisting of: titanium, zinc, aluminium, magnesium, calcium, scandium, vanadium, chromium, manganese, iron, cobalt, nickel, copper, gallium, germanium, strontium, yttrium, niobium, molybdenum, ruthenium, rhodium, palladium, silver, cadmium, indium, tin, antimony, tellurium, barium, tantalum, tungsten, rhenium, osmium, iridium, platinum, gold, mercury, thallium, lead, bismuth, zirconium, and lanthanum or actinium group metals. Where it is desired for the metal oxide to be particularly porous, porosity is favoured by selecting the metallic cation from the group consisting of: titanium, zinc, aluminium, magnesium, calcium, scandium, vanadium, chromium, manganese, iron, cobalt, nickel, copper, gallium, germanium, strontium, yttrium, niobium, molybdenum, cadmium, indium, tin, antimony, tellurium, barium, tantalum, tungsten, thallium, lead, bismuth, zirconium, and lanthanum or actinium group metals. Where the solute comprises a metal cation, the solute may contain an anion or ligand selected from the group consisting of: nitrate, acetate, acetyl acetonate, nitrite, chloride, citrate, ammonia, carbonyl, cyclopentadienyl and its derivatives, and anions of organic acids including amino acids. The fluid will also contain a suitable solvent for the cation and anion/ligand.

When the fluid is a dispersion in a liquid continuous phase of solid particles of at least one substance for forming the coating, the dispersion is preferably a colloidal sol. Generally, a colloidal sol is understood to have particles having a size of from about lnm to about Ιμηη. However, the dispersion may have larger particles so long as they are maintained dispersed during the method by movement of the fluid.

When the fluid is a dispersion, the solid particles may form or contribute to the coating without undergoing chemical transformation, although it is possible for the solid particles to undergo chemical transformation during the progressive application of localised heat. As discussed above, a fluid which is a dispersion may have one of more coating forming substances in the form of dispersed solid particles together with one or more coating forming substances in the form of solutes in the liquid continuous phase.

Suitable solid particles for forming the coating include silicon dioxide, aluminium oxide, titanium dioxide, cerium oxide, zirconium oxide, iron oxide, carbon, mixed oxides, or combinations thereof.

Suitable suspending liquids include: water, acetone, isopropanol, methanol, acetic acid, acetonitrile, butanol, carbon tetrachloride, 1,4-dioxane, ethanol, ethylene glycol, glycerol, ethyl acetate, methyl acetate, propylene glycol, 1-propanol, chloroform, dichloromethane, tetrahydrofuran, toluene, hexane, heptane, petroleum ether or other hydrocarbons, mixtures or solutions thereof.

Where it is desired to form a porous coating, the fluid may also contain a molecule in solution which acts to increase the mean pore diameter of the coating. The molecule may be a large molecule which may be decomposed or oxidised during the progressive application of localised heat to leave a relatively large pore. Alternatively, the molecule may be a large molecule which can be washed or evaporated away after the formation of the coating to leave a relatively large pore. The molecule may be a polymer. One type of polymer which is particularly useful for increasing the mean pore diameter is a block copolymer having the structure poly(ethylene glycol)-poly(propylene glycol)-poly(ethylene glycol). Block copolymers of this type are sold under the trade name Pluronic (trade mark) by BASF. Pluronic F127 (trade mark) has been found to be particularly useful for increasing the mean pore diameter. Other polymers that may be used to increase the mean pore diameter are: latex; poly(methyl methacrylate); polystyrene; and cross-linked polymers. One particularly suitable cross-linked polymer is polystyrene-divinylbenzene. Other large combustible molecules may be used.

In one embodiment, the fluid contains particles of carbon dispersed in the liquid continuous phase and also another substance (or more than one other substance) for forming the coating. The other substance (or one of them) may be a solute dissolved in the liquid continuous phase. For example, the solute may undergo thermal decomposition to form a solid oxide. Alternatively, the other substance (or one of them) may be dispersed particles suitable for forming a coating, such as particles of an oxide. On formation of the coating, the carbon particles are part of and distributed within the coating. Subsequently the coating may be heated. The heating may cause the carbon particles to burn off so that the carbon is no longer part of the coating. This would leave relatively large pores in the remaining component(s) of the coating. In one particular example, the coating could initially comprise silicon dioxide and carbon particles in a ratio of 5:1 and the carbon could then be combusted to leave the oxide coating with large pores.

As discussed above, it is preferred that when any location along the internal pathway is being subjected to the localised heating, that location is preferably exposed to a portion of the mixture of the gas and the fluid that has not been substantially depleted of coating substance (or substances) by previous upstream heating of the mixture. In a particularly preferred embodiment, this is achieved by ensuring that the streaming of the mixture along the internal pathway and the progressive application of localised heat along the internal pathway proceed in respective opposite directions. However, this is not essential. It is also possible to ensure that a particular location along the internal pathway is heated while being exposed to a portion of the mixture of the gas and the fluid that has not been substantially depleted of coating substance, when the streaming of the mixture along the internal pathway and the progressive application of localised heat along the internal pathway proceed in the same direction. In this case, the streaming along the internal pathway and the progressive application of heat proceed at respective different rates along the internal pathway, and preferably the streaming proceeds at a faster rate along the internal pathway as compared to the application of localised heating.

Preferably, the mixture is formed by flowing the fluid along a first conduit to a junction and flowing the gas along a second conduit to the junction so that the fluid and the gas meet and form the mixture at the junction. The junction may be, for example, a T-junction or an X-junction. The fluid may be flowed at a rate of between 0.001 to 10 ml/minute, and more preferably at a rate of between 0.02 to 0.2 ml/minute. The gas may be flowed at a rate of 0.001 to 1000 ml/minute (STP), and more preferably at a rate of between 0.1 to 30 ml/minute (STP). Higher gas flow rates may be preferred when the coating takes place under elevated pressure, such as when the internal pathway is long and has a high pressure drop, or when the internal pathway is provided with a back-pressure regulator to increase the pressure in the internal pathway. Increasing the pressure may help to achieve a more uniform coating, especially for long internal pathways. In general, however, it is often desirable to keep the gas flow rate as low as possible because low gas flow rates favour greater deposition of the coating on the internal surface. If the gas flow rate is too low, uneven coating may result.

An optimum gas flow rate often falls in the range of from 1 to 10 times the fluid flow rate.

While the mixture is streamed along the internal pathway, before it is subjected to the localised heat, the mixture generally takes the form of slugs of the fluid separated by portions of the gas, the slugs filling the cross-section of the internal pathway. However, other forms are possible, for example, droplets of the fluid suspended in the gas.

Normally, but not exclusively, after the coating has been formed as discussed above, the body is subjected to a further heating step to stabilise the coating. This further heating step may be carried out in an oven. During the further heating step, the body may be heated to a temperature in the range of 200°C to 1000°C, and preferably 300° to 900°C. This further heating step may be conducted for a duration in the range of 2 hours to 6 hours, preferably from 3 hours to 5 hours. The further heating step may cause the coating to adhere more firmly to the internal surface. The further heating step reduces or prevents loss of coating when fluid is subsequently passed through the internal pathway.

Where the coating is porous, the coating may include or be provided with particles trapped in the porous coating. The particles may be, for example, metal particles. Preferably, the particles are metal particles with a desired catalytic activity. For example, the particles may be metal particles selected from the group consisting of: platinum, palladium, rhodium, osmium, iridium, ruthenium, copper, silver, cobalt, iron, nickel and gold, or mixtures thereof.

In one embodiment, the porous coating is impregnated with the particles after formation of the porous coating. The impregnation may take place before or after the stabilising heating step described above but is preferably carried out after the stabilising heating step.

The impregnation of the porous coating with particles may be carried out by filling the length of the internal pathway that is provided with a coating, with a dispersion of the particles in a liquid and then evaporating the liquid to leave the particles entrapped in the porous coating. The evaporation will generally be facilitated by heating. Alternatively, instead of using a dispersion of the particles in a liquid, the length of the internal pathway that is provided with a coating may be filled with a solution that has a solute which decomposes on heating to form the particles.

In the latter case, the solute may be a metal salt or compound with a cation selected from the group consisting of: platinum, palladium, rhodium, osmium, iridium, ruthenium, copper, silver, cobalt, iron, nickel or gold. The metal salt or compound may comprise an anion or ligand selected from the group consisting of: acetate, nitrate, nitrite, citrate, chloride, carbonyl, ammonia and acetyl acetonate.

When either a dispersion or solution is used in this way to impregnate a coating with particles, a potential problem is that, during the heating, gas bubbles may form within the dispersion or the solution and expel the dispersion or the solution from the internal pathway. The expulsion of the dispersion or the solution is undesirable (although expulsion of gas formed from evaporation is inevitable and acceptable). The following preferred embodiment avoids or minimises the expulsion of the dispersion or the solution. In this preferred embodiment, the previously coated portion of the internal pathway is filled with a column of the dispersion or the solution. The column has first and second ends. The heating is performed while one of the inlet and the outlet is closed and the other one of the inlet and the outlet is open. The first end of the column lies closer than the second end to the open one of the inlet and outlet along the length of the internal pathway, and the second end of the column lies closer than the first end to the closed one of the inlet and outlet along the length of the internal pathway. The heating comprises applying heat progressively to successive regions of the column starting at the first end of the column and moving towards the second end of the column. In this way, expulsion of the dispersion or the solution from the liquid pathway is reduced or prevented. This is because evaporation takes place at the end of the column (the first end) that is closest to the open end of the internal pathway, and evaporation does not take place to any substantial extent within the column so as to form gas bubbles within the column. The gas produced on evaporation escapes harmlessly from the open end of the internal pathway. The application of heat and the evaporation progress towards the second end of the column. This is done sufficiently slowly so that evaporation substantially only takes place at the first end of the column.

The inlet or outlet may be closed in any convenient manner, such as being blocked in a reversible way by laboratory film, or by a valve.

The rate at which the application of heat moves progressively along the column of dispersion or solution may preferably be from ΙΟμιτι to 10cm per second, and more preferably from 0.1mm to 10mm per second, while being sufficiently slow to prevent or reduce expulsion of dispersion or solution.

During the process of impregnating the porous coating with the particles, the heat may be applied by a heat source that extends around the body and there is relative movement between the body and the heat source to cause the progressive application of heat.

This method of applying heat progressively along a column of dispersion or solution to cause impregnation of a porous coating with particles is not essential. Instead, the expulsion of the dispersion or solution may be avoided if the body is uniformly heated at a low temperature just sufficient to cause evaporation. Slower preparation times may result. It is also possible to incorporate particles, such as metal particles, during formation of a porous coating. In this case, the particles may be considered as part of the coating. For example, the fluid may comprise both a first substance which forms a porous component of the coating and a second substance providing the particles. The second substance may simply be, for example, metal particles dispersed in the fluid which become entrapped in the porous component as the coating is formed. Alternatively, the second substance could be a dissolved metal salt or compound which undergoes thermal decomposition during the progressive application of localised heat to form the metal particles. For example, the metal salt or compound may comprise a cation selected from the group consisting of: platinum, palladium, rhodium, osmium, iridium, ruthenium, copper, silver, cobalt, iron, nickel or gold. The metal salt or compound may comprise an anion or ligand selected from the group consisting of: acetate, nitrate, nitrite, citrate, chloride, carbonyl, ammonia and acetyl acetonate. The substance which forms the porous component could be, for example, a dissolved metal salt which decomposes to form a porous metal oxide, or a dispersion of metal oxide particles.

The provision of particles entrapped within the coating is optional. For example, the method may be used to provide a porous coating within the internal pathway so that the body can be used for chromatography.

The coating does not need to be porous. For example, the coating may be a metallic coating, which may have catalytic properties. In one such example, the fluid comprises a substance in the form of a solute comprising a metallic cation and an anion (or another ligand). The metallic cation is selected from the group consisting of: platinum, palladium, rhodium, osmium, iridium, ruthenium, copper, silver, cobalt, iron, nickel or gold. During the progressive application of localised heat, the solute decomposes to form the metallic coating.

Where the fluid comprises a substance in the form of a solute which undergoes thermal decomposition, the concentration of the solute can vary greatly. Generally, the greater the concentration, the thicker the coating that is formed. Suitable concentrations may be in the range from 0.1 wt/wt% to 80 wt/wt%. Where the fluid comprises a substance in the form of a solute which undergoes thermal decomposition, preferably the fluid comprises a solvent selected from the group consisting of: water, methanol, ethanol, toluene, xylene, isopropanol, hexane, tetrahydrofuran, dimethylformamide, acetonitrile, dimethyl sulfoxide, and mixtures thereof. However, other suitable solvents may be used. In many cases, solvents with relatively low boiling points are preferred because they evaporate more quickly during the progressive application of localised heat and this may allow a shorter overall preparation time. The solvent should be chosen such that the solute is sufficiently soluble in the solvent. In addition, the solvent should have a boiling point that allows evaporation at a sufficient rate at a temperature that is sufficiently low to avoid damage to the body. In addition, the solvent should preferably not damage the coating or dissolve the coating to any significant extent.

Preferably, in cases where only a part of the internal surface is covered by the coating, the coating covers at least 50% of the internal surface, and preferably at least 60%, or at least 70%, or at least 80%, or at least 90% of the internal surface. More preferably, the part of the internal surface that is covered by the coating is a continuous part of the length of the internal surface. Of course, the coating may cover the internal surface completely.

Preferably, the body is formed from a material selected from the group consisting of: silica, steel, titanium, copper, aluminium, and plastics. However, other suitable materials may be used. The material should be stable at the temperatures used for heating. The body may be formed form a plurality of materials.

The method may be used in many applications. For example, the method may be used to provide a tube or cartridge or other body with a coating so as to form a chromatography column. In this case, it may be sufficient for the coating to consist of a porous solid without any additional components. Alternatively, the method may be used to coat an internal surface with a catalytic coating so that chemical reactions catalysed by the coating may be performed in an internal pathway, or within the internal spaces of a porous body. This may be particularly useful for coating internal channels of micro-reactors or mili-reactors.

The following is a more detailed description of embodiments of the invention, by way of example, with reference to the appended drawings, in which: Figure 1 is a schematic representation of an apparatus for performing the method;

Figure 2 is graph showing the effect of different temperatures for the stabilising heating step;

Figures 3a and 3B are cross-sectional scanning electron microscope images showing a porous coating on an internal surface of a capillary tube; and

Figures 4a and 4b are c,ross-sectional schematic representations of bodies with more than one channel.

Example 1

Referring to Figure 1, the apparatus used to perform the method of Example 1 comprised a syringe pump 10, a gas mass-flow controller 12, a T-junction 14, a stepper motor 16, a vertical furnace 18 and an extraction hood 20. The syringe pump 10 was connected to the T- junction 14 by a first conduit 22 and the gas mass-flow controller 12 was connected to the T-junction by a second conduit 24.

Figure 1 also shows a tube 26 (the body in this Example) that was provided with a coating. The tube 26 was formed from 316L stainless steel and had a length of 5m. The tube 26 had an external diameter of 1.55mm and an internal diameter of 1.27mm. The current method was used to provide a coating on the internal surface of the tube 26 adjacent the internal channel of the tube 26.

As a first step, a porous coating of zinc oxide was formed on the internal surface of the tube 26. Firstly, the syringe pump 10, the gas mass-flow controller 12, the first and second conduits 22, 24, the T-junction 14 and the tube 26 were placed in a desiccator to remove any traces of water. (This prevents precipitation of the zinc solution as zinc hydroxide.) The syringe pump 10 was then filled with a solution of zinc (II) nitrate hexahydrate (28.6g) in 50ml of ammonia solution (28 wt% in water). The gas mass-flow controller 12 was connected to a source of dry compressed air. An inlet 28 of the tube 26 was connected to the T-junction 14. The syringe pump 10 was then operated to pump the zinc solution to the T-junction 14 at a flow rate of 500μΙ min ^"1 . The gas mass-flow controller 12 was operated to feed air to the T-junction 14 at a flow rate of 4ml min ^_1 (STP). At the T-junction 14, the zinc solution and the air mixed to form a mixture. The mixture was streamed continuously into and through the tube 26 to a tube outlet 28. The mixture is believed to have progressed through the tube 26 in the form of slugs of the solution filling the cross-section of the tube 26, the slugs being separated from one another by air.

While the mixture of the air and the zinc solution was streamed through the tube 26, the tube 26 was fed into the vertical furnace 18 by the stepper motor 16 at a rate of 5mm s-1, starting with the tube outlet 30 and progressing towards the tube inlet 28. The vertical furnace 18 was maintained at a temperature of 350°C. Hence, successive portions of the length of the tube 26 passed into, passed through for a predetermined residence period, and passed out of the vertical furnace 18.

The feeding of the tube 26 through the vertical furnace 18 was continued until no further feeding was possible (leaving only a minimum length of the tube 26 located adjacent the tube inlet 28 that was not heated in the vertical furnace 18). Streaming of the mixture of the air and the zinc solution dispersed through the tube 26 was continued for the whole of this process.

The heating in the vertical furnace 18 caused the zinc solution to undergo thermal decomposition to form zinc oxide. Specifically, zinc ammonium hydroxide dissolved in the solution decomposes to form zinc oxide. The zinc oxide formed a coating on the internal surface of the tube 26.

Gas and solution passing out of the tube outlet 30 entered the extraction hood 20.

The zinc oxide coating produced in this way was found not to be stable and up to 70% could be removed by washing.

As a second step, to stabilise the zinc oxide coating, the tube 26 was heated in an oven for 4 hours. To optimize this step, four different experiments were performed at respective different temperatures with four different tubes 26 each provided with a zinc oxide coating as described above. Temperatures of from 550°C to 950° were tested. In each case, after heating for 4 hours at the test temperature, the tube 26 was flushed with 20 ml of isopropanol or acetone at a flow rate of 100ml min ^"1 to remove loosely bound zinc oxide. The results are shown in Figure 2 which shows the mass of the remaining zinc oxide coating after washing on the y-axis, and the temperature used for the heating step on the x-axis. It was found that heating at a greater temperature has a greater stabilising effect on the zinc oxide coating. A temperature of 800°C was chosen for routine experiments as this temperature was effective at stabilising the coating while it could be readily achieved using inexpensive equipment and without undue energy expenditure.

After the formation and stabilization of the zinc oxide coating in the first and second steps as described above, the zinc oxide coating, which was porous, was then impregnated with catalytic palladium nanoparticles in a third step.

To impregnate the porous zinc oxide coating with palladium nanoparticles, the tube 26 was first filled with a solution of palladium (II) acetate dissolved in acetone to form a column of the palladium solution in the internal pathway. The concentration of the palladium (II) acetate was calculated to give a final palladium metal loading of 5wt% with respect to the total weight of the zinc oxide coating. The tube outlet 30 was closed and the tube inlet 28 was left open. The tube 26, together with the column of palladium (II) acetate solution, was then advanced into an oven heated at 300°C at a rate of 5mm s ^"1 starting with the open tube inlet 28. During this process, the acetone solvent evaporated and the palladium (II) acetate decomposed to form palladium metal nanoparticles entrapped in the porous zinc oxide coating. The rate of advancement of the tube 26 into the oven was sufficiently slow so that evaporation of the acetone solvent took place only at the end of the column of palladium solution closest to the open tube inlet 28. In this way, evaporation in the body of the column, which is undesirable because it could cause ejection of the palladium solution, was avoided.

The coating is shown in Figures 3a and 3b. The coating is shown at 32 and the wall of the tube is shown at 34. The coating 32 has an even thickness and is about 18μιτι thick.

The tube may be used to perform conversion of nitrobenzene into aniline, 2-methyl-3- butyn-2-ol into 2-methyl-3-buten-2-ol with the palladium nanoparticles catalysing the reaction or cinnamaldehyde into cinnamyl alcohol with platinum nanoparticles catalysing the reaction.

Example 2 In Example 2, a commercial Si0 ₂ sol was used (Ludox 30wt%) supplied by Sigma-Aldrich as a coating precursor. The sol was diluted 30-fold to obtain a 1 wt% Si0 ₂ sol which was used for the coating method. The method was performed using the apparatus described in Example 1 and shown in Figure 1.

The tube 26 that was coated in Example 2 was identical to the tube 26 coated in Example 1. The tube 26 was washed with petroleum spirit and acetone and dried prior to the coating.

The 1 wt% Si0 ₂ sol was pumped from the syringe pump 10 to the T-junction 14 at a rate of 150 μΙ min ^"1 . Air was passed to the T-junction 14 by the gas mass-controller 12 at a flow rate of 4 ml min ¹ (STP). At the T-junction 14, the sol and the gas mixed to form a mixture of air and sol. This mixture was streamed into and through the tube 26. It is believed that the mixture progressed through the tube 26 in the form of slugs of sol, filling the cross-section of the tube, separated from one another by pockets of air.

During the streaming of the mixture of the air and the sol through the tube 26, successive portions of the length of the tube were moved into, through and out of the vertical furnace 18 at a constant displacement velocity of 3 mm s ^"1 . The furnace temperature was 180°C. As the tube passed though the vertical furnace 18, the liquid continuous phase of the sol evaporated and the silicon dioxide particles of the sol formed a coating on the internal surface of the tube 26.

After the silicon dioxide coating had been formed on the internal surface of the tube 26, the tube was subjected to an annealing step at 350°C for 4 hours in an oven to stabilise the coating. The mass of the coating obtained was around 200 mg.

After the silicon dioxide coating had been stabilised, the coating was impregnated with platinum nanoparticles. The platinum nanoparticles were first obtained by dissolving hexachloroplatinic acid (lg) in 50 ml ethylene glycol and heating the solution to 160°C under reflux and with stirring for 4h to obtain a stable dispersion of platinum nanoparticles. This dispersion of platinum nanoparticles was diluted with sufficient ethylene glycol calculated to give a 12 wt% platinum loading (with respect to the weight of the silicon dioxide coating) assuming full incorporation of the platinum nanoparticles into the silicon dioxide coating. The diluted dispersion of platinum nanoparticles was introduced into the tube 26 until the tube 26 was filled with the platinum dispersion which formed a column within the tube 26. The tube outlet 30 was closed and the tube inlet 28 was left open. The tube was then introduced into an oven starting with the open tube inlet 28 at a rate of 5 mm s ¹ . The oven temperature was 350°C. The rate of introduction was sufficiently slow so that evaporation of the ethylene glycol dispersant occurred only at the end of the column of the platinum dispersion (and not within the body of the column). In this way, the dispersant evaporated in a controlled manner, without formation of air bubbles in the body of the column and ejection of the dispersion, and the platinum nanoparticles were left incorporated into the porous silicon dioxide coating.

Example 2 was repeated several times to study the effect of varying various parameters. It was found that an increase in gas flow rate reduces the coating yield. Increasing the heating temperature used to stabilise the coating had no effect on the coating. Instead of using air as the gas used to form the mixture of gas and fluid, helium or nitrogen may be used but there is no effect on the coating.

If a tube 26 with a different internal diameter is used, a coating of similar properties may be obtained by increasing or decreasing either the flow rate of the silicon dioxide sol, or the concentration of the silicon dioxide sol, in either case so as to keep the flow rate or the concentration proportional to the internal volume of the tube, while keeping the gas flow rate the same.

Example 3

Using the same apparatus described in Example 1 with reference to Figure 1, and a similar methodology, an internal surface of a tube 26 was provided with a coating of magnesium oxide. Instead of the zinc solution used in Example 1, an aqueous solution of 1 wt% Mg(N0 ₃ ) ₂ was used as a coating precursor. The solution was displaced into the stainless tube 26 at a flow rate of 100 μΙ min ^"1 and the air flow rate was 4 ml min ^"1 . The tube 26 was displaced into the vertical furnace 18 at a velocity of 2 mm s ^"1 and the furnace temperature was 350°C. The heating caused the magnesium nitrate to decompose to magnesium oxide. The magnesium oxide coating obtained had a mass of 140 mg. After the coating was formed, it was heated at 500°C for 4 hours to stabilise the coating. Examples 4a and 4b

Examples 4a and 4b are examples of the use of the method to form a coating within a body 36, 38 which has a plurality of channels extending therethrough.

In Example 4a, the body 36 consisted of a quartz tube 40 and two stainless steel tubes 42, 44 inserted in the quartz tube 40. The quartz tube 40 had a length of 50 cm, an outside diameter (OD) of 6 mm and an inside diameter (ID) of 4 mm. The two stainless steel tubes 42, 44 had outer diameters of 3 mm and 1.5 mm respectively, and each had a wall thickness of 0.2 mm. The two stainless steel tubes 42, 44 were inserted coaxially into the quartz tube 40, as shown in Figure 4a.

The quartz tube 40 was connected to gas and fluid flows via a T-junction as in Example 1.

1 wt% Si0 ₂ sol was pumped from the syringe pump 10 to the T-junction 14 at a rate of 150 μΙ min ^"1 . Air was passed to the T-junction 14 by the gas mass-controller 12 at a flow rate of 4 ml min ^"1 (STP). At the T-junction 14, the sol and the gas mixed to form a mixture of the air and the sol. This mixture was streamed into and through the quartz tube 40 and the stainless tubes 42, 44 inserted into it.

During the streaming of the mixture of the air and the sol, successive portions of the length of the quartz tube 40 (with the stainless steel tubes 42, 44 therein) were moved into, through and out of the vertical furnace 18 at a constant displacement velocity of 3 mm s ^"1 . The furnace temperature was 180°C. As the body passed though the vertical furnace 18, the liquid continuous phase of the sol evaporated and the silicon dioxide particles of the sol formed a coating on the internal surface of the quartz tube 40 and on the inner and outer surfaces of the steel tubes 42, 44. About 80 mg of the coating was obtained inside the body 36.

In Example 4b, the body 38 was made of a 6mm OD quartz tube 46 with 3 stainless steel tubes 48, each having an OD of 1.6 mm and an ID of 1.2 mm, inserted into the quartz tube 46, in a generally triangular configuration as shown in Figure 4b. The method was conducted as described above for Example 4a. About 70 mg of the coating was obtained including about 30 mg inside the stainless tubes 48.