Technical Field

The present application relates to filters, particularly water filters, methods used to deposit composite surface coatings for filters and the use of them to remove contaminants from fluids.

Background

Many environmental contaminants exist that can make their way into water systems and pose a threat to humans, wildlife and the environment in general. Contaminants include heavy metals,

pesticides, residues and metabolites from pharmaceuticals, waste products from industrial processes, bacteria and many others.

Removal of contaminants from water systems presents a

significant challenge, especially if additional process restrictions are placed on the removal system. For example in many remote areas of the world contaminated water sources present a serious threat to human and animal life but a system to remove contaminants must be low cost, easy to install, operate and repair and be able to remove contaminants from large volumes of water before requiring

replacement or recharging. For commercial reasons, such systems should also be simple and cost effective to manufacture. For these reasons, many water treatment solutions are based on filtration systems, often including an active filter element.

Many known active filter systems include species that absorb or neutralise contaminants. For example activated carbon and zero valent iron have both been shown to be effective to remove

contaminants from water systems.

One method for making an active filter element is described in US 2010/0307978 Al . In this document a foam substrate is coated with an adherent composition, such as a polyurethane , an acrylic, a silicone or another inherently tacky material, which dries to a water resistant coating. A reactive absorbent powder, such as powdered zero-valent iron, carbon, copper, zinc or zeolite, is then applied to the coated surface of the substrate which is subsequently shaken or moved to spread the powder over the coated surfaces of the foam substrate. This coated substrate is then proposed as a water filtration substrate.

Summary

The present proposals relate generally to a filter element comprising a substrate with a composite layer on a surface of the substrate, the composite layer comprising nanoparticles (NPs) comprising a first metal or a first metal oxide and a layer of a second metal, a second metal oxide or a combination thereof between the NPs on the substrate surface.

In other proposals the present invention relates to a filter element comprising a substrate with a composite metal layer on a surface of the substrates, the composite layer comprising

nanoparticles (NPs) comprising a first metal and a layer of a second metal between the NPs on the substrate surface.

In preferred embodiments the NPs comprise a first metal.

In other embodiments the NPs comprise a first metal oxide.

Of course, the NPs may comprise a combination of a first metal and a first metal oxide.

In preferred embodiments the layer between the NPs on the substrate surface comprises a second metal.

In other embodiments the layer between the NPs on the

substrate surface comprises a second metal oxide.

In other embodiments the layer between the NPs on the

substrate surface comprises a second metal and a second metal oxide.

Further proposals relate to a method of forming a composite layer on a substrate surface by providing a suspension of

nanoparticles in a liquid, the nanoparticles (NPs) comprising a first metal or a first metal oxide and the liquid comprising second metal ions, and electrochemically depositing a composite layer of NPs and second metal on a substrate surface.

In other proposals the present invention relates to a method of forming a composite metal layer on a substrate surface by providing a suspension of nanoparticles in a liquid, the

nanoparticles (NPs) comprising a first metal and the liquid

comprising second metal ions, and electrochemically depositing a composite layer of NPs and second metal on a substrate surface.

Other aspect of these proposals relate to filters including the filter elements of these proposals and methods of treating a fluid system, e.g. a water system, to remove a contaminant using such filters and/or filter elements.

The present proposals also include a fluid (e.g. water) treatment system and method comprising removal of a contaminant from a fluid (e.g. water) system using the filter and/or filter element of the present proposals, and regenerating the filter element to release the contaminant and restore contaminant-removal potential of the filter element.

Brief Description of the Figures

Fig 1 shows a SEM image of a carbon foam surface treated according to the method described in Example 1. Scale bar 200nm.

Fig 2 shows a SEM image of a carbon foam surface treated according to the method described in Example 2. Scale bar 200nm.

Fig 3 shows a SEM image of a carbon foam surface treated according to the method described in Example 3. Scale bar 200nm.

Fig 4 shows a carbon foam surface at various stages of treatment according to the method described in Example 4. Scale bars (L-R top row then L-R bottom row) 200μιτι, ΙΟμιτι, Ιμιτι, 200nm, lOOnm.

Fig. 5 shows a STEM image of a carbon foam surface treated according to the method described in Example 4. Fig. 6 shows a bright-field STEM image of a cross-section through a carbon foam surface treated according to the method described in Example 4 also showing a cross section through the nanoparticles deposited on the surface.

Fig. 7a shows a SEM image of a carbon foam surface treated according to the method described in Example 6. Scale bar Ιμιτι.

Fig. 7b shows a SEM image of a carbon foam surface treated according to the method described in Example 7. Scale bar 50μιτι.

Fig. 8a shows a SEM image of a carbon foam surface treated according to the method described in Example 8. Scale bars (L/R)

Fig. 8b shows a SEM image of a carbon foam surface treated according to the method described in Example 9. Scale bars (L/R)

Fig. 8c shows a SEM image of a carbon foam surface treated according to the method described in Example 10. Scale bars (L/R)

Fig. 8d shows a SEM image of a carbon foam surface treated according to the method described in Example 11. Scale bars (L/R)

Fig. 9a shows a SEM image of a carbon foam surface treated according to the method described in Example 12. Scale bar Ιμιη.

Fig. 9b shows a SEM image of a carbon foam surface treated according to the method described in Example 13. Scale bar 4μιη.

Fig. 10 shows the experimental apparatus used to perform the electrodeposition described in Example 14. Fig. 11 shows SEM images of the nano-structure on a top surface of the treated RVC substrate (top and bottom left) and a section approximately 3.8mm from the top surface (top and bottom right) . Scale bars 300nm (top row) , 500nm (bottom row) .

Fig. 12 shows an XPS spectrum of Fe 2p3/2 peak with fitted curves for before (top) and after (bottom) vacuum annealing.

Fig. 13 shows percentage of organic removed normalised by surface area of reactive material (iron nano-composite 12m ² g ^_1 , granulated activated carbon 65m ² g ^_1 , organo-clay 825m ² g ^_1 ) .

Description

The filter element of the present proposals or substrate surface coated according to the present methods may provide improved contaminant removal activity compared to known filter elements .

This may be manifested in increased reaction speed (e.g. fluid has to pass over a smaller area of the filter surface for the

contaminants to be removed), increased reaction potential (e.g. a given area of the filter surface can remove a larger amount of contaminant before the reactive composite layer on the surface of the filter element is depleted to the extent that the filter element no longer effectively removes contaminants from the fluid system) and/or increased lifespan of the filter element (e.g. a larger volume of fluid can be treated by a single filter element before the reactive composite layer on the surface of the filter element is depleted to the extent that the filter element no longer effectively removes contaminants from the fluid system) compared to known filter elements .

The filters and coated surfaces of these proposals comprise a substrate with a composite layer on a surface of the substrate, the composite layer comprising nanoparticles (NPs) comprising a first metal or a first metal oxide and a layer of a second metal, second metal oxide or a combination thereof between the NPs on the

substrate surface.

In some embodiments, the filters and coated surfaces of these proposals comprise a substrate with a composite metal layer on a surface of the substrate, the composite layer comprising nanoparticles (NPs) comprising a first metal and a layer of a second metal between the NPs on the substrate surface.

In preferred filter elements the substrate component can be made from any material that forms a porous substrate body.

Preferably the substrate has a high surface area. Exact surface area may depend on the application of the final filter element but a preferred surface area is between about 4 and 40 m ² /g. The

substrate is preferably a material having a solid foam structure, preferably an open-cell foam, and preferred substrates are formed from an electrically conductive material because this allows the methods of the present proposals to be used to manufacture the filter element. Examples of materials for the substrate include porous metallic bodies such as porous copper, porous carbon

materials such as porous graphite or carbon foams, and porous, electrically conductive resins. Preferably the substrate is formed from porous carbon materials and most preferably from reticulated vitreous carbon foam.

As an alternative possibility which also forms part of these proposals, the substrate may be a particulate substrate on which the composite coating of the present proposals may be formed. Such particulate substrates may include silica or silicate particles (such as sand) , alumina particles, graphite or carbon particles, metallic particles, and glass, plastic or resin particles or beads. The particulate substrate preferably has a particle diameter of between about 1 μιτι and 5 mm, more preferably between about 0.5 mm and 2mm. A preferred particulate substrate may be particulate carbon such as granular activated carbon (GAC) .

The use of an open cell foam structure as the substrate has an advantage that as fluid flow through an initial path of the filter element is restricted over time, the fluid can divert to travel via a different pathway through the filter element rather than blocking the filter itself and requiring higher head pressure to force the fluid through restricted pathways. This is particularly

advantageous if the composite material deposited on the surface of the substrate increases in volume on reaction with contaminants in a fluid system causing the pores in a porous substrate body to become blocked or flow through them restricted over time.

For filtration applications, a discrete porous substrate body is preferred, although a particulate substrate can also be useful if packed into a filter housing so that fluid to be filtered flows through the packed coated particles and contacts the coated

surfaces .

The composite layer on the surface of the substrate comprises both NPs comprising a first metal or a first metal oxide (preferably a first metal) and a layer of a second metal, second metal oxide or combination thereof (preferably a second metal) between the NPs on the surface of the substrate. Sometimes the second metal layer, second metal oxide layer or combination of second metal and second metal oxide layer also covers the NPs, i.e. it forms a layer covering both the surface of the substrate between the NPs and the NPs themselves (in such situations, the NPs may be referred to as being "overgrown" by the metal layer) .

When the NPs are metallic, they are preferably formed from at least one metal but may comprise more than one metal, e.g.

bimetallic NPs (i.e. the "first metal" combined with another metal) or trimetallic NPs (i.e. the "first metal" combined with two additional metals) . It is understood that the metallic NPs may have a thin outer coating of oxidised material, e.g. if they have been exposed to atmospheric oxygen or other oxidising environment.

The first metal may be any metal that has the ability to react with contaminants but is preferably a transition metal and

preferably selected from: Fe, Ni, Ag, Au, Cu, Zn, Pt and Co, or Ce or Al . Preferably the first metal is Fe . Fe is particularly preferred because Fe NPs are cheap to make and the size of the NPs can be easily controlled to give a narrow size distribution in a target size range. Fe is also shown to be effective for the treatment of fluids, especially water, to remove contaminants.

However, other metals can also provide advantages, for example Ag exhibits good antibacterial properties and is effective to remove Hg from fluid systems and Au also demonstrates effective fluid, especially water, treatment properties to remove contaminants. Where the NPs are formed from a mixture of metals (i.e. the first metal and one or more others), they preferably include one or more of the elements: Fe, Ni, Ag, Au, Cu, Zn and Co; combined in any proportions. Preferred mixtures include Fe in combination with one or more of Ni, Cu and Co. Preferably NPs formed from a mixture of metals are an alloy of Fe and Ni, more preferably in an 80:20 w/w Fe:Ni ratio.

The first metal oxide may be any metal oxide but is preferably a transition metal oxide, for example Ti0 ₂ , ZnO, Fe ₂ 0 ₃ , Fe ₃ 0 ₄ or CrO, preferably Ti0 ₂ . The advantage of using metal oxide nanoparticles is that the composite layer will then have the advantageous

properties related to the metal oxide. For example, Ti0 ₂ is a semiconductor, and in addition when Ti0 ₂ nanoparticles are used to form the composite layer the material property of Ti0 ₂ allows the formation of reactive oxygen species (ROS) with UV illumination.

The NPs preferably have a particle size of about 1 - <1000nm, preferably 2-100nm and more preferably 5-25nm. Above about lOOOnm reactivity of the particles diminishes undesirably.

The second metal (forming the layer between the NPs) may be any metal but is typically one with the ability to react with contaminants and is preferably a transition metal and more

preferably selected from: Fe, Ni, Ag, Au, Cu, Zn, Pt and Co, or Ce or Al . Preferably the second metal is Fe .

In a similar way to the metallic NPs, the layer between the NPs on the substrate surface may comprise one or more metals in addition to the second metal. For example, the layer comprising the second metal may further comprise one or more metals selected from Fe, Ni, Ag, Au, Cu, Zn and Co.

When the nanoparticles are metallic, preferably both the first metal and the second metal are the same and more preferably are both Fe .

Most preferably, the NPs are formed from and Fe/Ni mixture and the layer between the NPs is formed from Fe . Although preferences are described above, the exact choice of first metal or first metal oxide, second metal, second metal oxide or combination thereof and the composition of the NPs and metal layer between them can be tailored to provide the most effective and efficient removal of desired, target contaminants.

Typically there is only a single layer of NPs on the surface as part of the composite layer with the second metal, metal oxide or combination thereof covering the surface between the NPs. The result of this is that the thickness of the composite layer is about the diameter of the NPs (or slightly thicker if the NPs are

overgrown with a layer of the second metal, metal oxide or

combination thereof) . Therefore the thickness of the composite layer is preferably less than 1 μιτι, more preferably about 10-500 nm, more preferably about 50-300nm. If the composite layer is too thick, little or no additional active surface area or catalytic activity is gained but the layer takes longer to form and uses more material so is economically less desirable.

The use of the composite coating on the substrate surface, i.e. a coating comprising NPs comprising a first metal or a first metal oxide (preferably a first metal) and a layer of a second metal, a second metal oxide or a combination thereof (preferably a second metal) between the NPs on the surface of the substrate, provides a filter element having a high capacity for removal of contaminants from a fluid system (e.g. a water system) . Typically this capacity for contaminant removal is higher than for known filter elements. This may be due to the very high reactive surface area achieved by the composite coating. This reactive surface area is typically higher than simple metal-plated surfaces or surfaces on which metallic NPs have been deposited. Simple metal-plating of a substrate can achieve a reactive surface but typically this has a similar surface area to the underlying substrate. Deposition of metallic NPs on a substrate surface can provide a higher active surface area than the underlying substrate, due to the NPs standing up from the substrate surface, but between the NPs the substrate surface is usually exposed unreactive substrate. The composite coating of the present proposals provide a higher reactive surface area than either of these two known situations because the NPs stand up from the substrate surface so exposing a high active surface area and in addition, an active surface layer is provided between the NPs so the exposed active surface is increased still further.

The present proposals also relate to a method of forming a composite layer on a substrate surface. These methods involve providing a suspension of NPs in a liquid, the NPs comprising a first metal or a first metal oxide and the liquid comprising second metal ions, and electrochemically depositing a composite layer of NPs and second metal on a substrate surface.

In embodiments, these methods involve providing a suspension of NPs in a liquid, the NPs comprising a first metal and the liquid comprising second metal ions, and electrochemically depositing a composite layer of NPs and second metal on a substrate surface.

The electrochemical deposition of the composite layer on a substrate surface typically involves providing an electrically conductive substrate as the cathode in an electrochemical cell with the liquid comprising second metal ions as the electrolyte. In preferred cases, the electrochemical deposition involves immersing an electrically conductive substrate in the suspension of NPs in the liquid comprising the second metal ions within an electrically conductive vessel. An electrolytic cell is then set up with the substrate acting as the cathode and the electrically conductive vessel acting as the anode. Application of an electric current moves the second metal ions out of the liquid to deposit on the substrate. The small size of the NPs suspended in the liquid means that they are also manipulated by the electric field in the vessel and deposit on the surface of the substrate with the ions of the second metal from the liquid to form the composite coating. Due to the use of nano-sized particles, an electrostatically charged region forms around the surface of the NPs allowing them to be manipulated by the electrical field (set up in the electrochemical cell) .

This electrochemical arrangement is preferred to one in which a sacrificial anode is used (i.e. where the metal for deposition on the substrate is provided by the anode itself which is consumed during the electrochemical deposition) because it provides a more uniform coating on the substrate surface and is a more robust and scalable process because the deposition is not significantly influenced by pH or concentration gradients in the electrochemical cell (which can be influential and problematic in sacrificial anode arrangements and can lead to uneven deposition on the substrate) .

In preferred methods, the electrochemical deposition current has a current density of less than 3 Acm ^-2 , preferably less than 1 Acm ^"2 , more preferably less than 0.6 Acm ^-2 , more preferably less thar 0.4 Acm ^-2 , more preferably less than 0.2 Acm ^-2 , and most preferably less than 0.1 Acm ^-2 , for example, a current density of 0.09 Acm ^-2 has been found useful in providing a suitable composite layer.

The current density of the electrochemical deposition current has been found to affect the formation of the composite layer; a greater current density causes a larger flux of metal ions towards the cathode, but this also causes more rapid evolution of hydrogen at the cathode as H ⁺ ions are generated. The production of hydrogei bubbles at the cathode may prevent or impede the second metal ions from binding to the substrate. The use of a deposition current with a current density of less than 1 Acm ^-2 can improve the adhesion of the NPs and/or second metal ions to the substrate and can help to prevent uneven deposition of the composite layer so that a stable and regular surface may be obtained. Of course, if the current density is too low the rate of deposition may be too low to be commercially useful, for example a current density of less than 0.001 Acm ^-2 may be too low to be commercially useful.

In preferred methods the ratio of the concentration of NPs to the concentration of the metal salt in the liquid is less than 80:2 and greater than 20:80, more preferably less than 75:25 and greater than 25:75, more preferably 50:50 (for example 5 gL ^-1 NPs: 5 gL ^-1 metal salt) . It has been found that a ratio of the concentration o NPs to the concentration of the metal greater than 80:20, or less than 20:80 may have a detrimental effect on the composite layer formed. Therefore the ratio of the concentration of NPs to the concentration of the metal salt in the liquid may help to provide a uniform composite layer. A preferred concentration of NPs in the liquid is about 1 - <10 gL ^-1 , preferably about 1-8 gL ^-1 , more preferably about 2-5 gL ^-1 .

In preferred methods the liquid electrolyte is an aqueous solution containing the second metal ions. However other solvents are also possible, such as organic solvents e.g. glycerol, or high purity volatile organic compounds. The exact preferred solvent is typically dependent on the dissolution characteristics of the metal ions in question, although the preferred water solvent is widely applicable .

The electrolyte liquid may also contain other optional components such as surfactants (e.g. carboxymethyl cellulose (CMC), guar gum, polyacrylic acid, sodium dodecyl sulphate (SDS), block copolymers such as Triton X-100 (RTM) , ammonium laureth sulphate, sodium laureth sulphate, benzalkonium chloride, fatty alcohols such as stearyl alcohol or cetyl alcohol, polyoxyethylene glycol alkyl ethers, poloxamers etc.) .

In preferred methods, the polarity of the electrolytic cell is reversed periodically, preferably in a cycle of alternative forward and reverse polarity. This can help to prevent uneven deposition of the composite layer resulting in a more even thickness and improved coverage of composite layer across the surface of the substrate. It is preferred that the period of time for which the polarity is reversed is less than or equal to (preferably less than) the time for which the polarity is in the forward direction. Preferably the ratio of the time for which the polarity is in the forward direction to the time for which the polarity is reversed is between 1:0.1 - <1, preferably about 1:0.2 - 0.5 and most preferably about 1:1/3. Below a ratio of about 1:0.1, i.e. with a shorter duration of reversed polarity, the effectiveness of the polarity reversal in preventing uneven deposition starts to be undesirably reduced. In preferred methods, the current flows in the forward direction for between 0.5 s and 5 min, preferably between 1 and 120 s, or between 1 and 60 s, more preferably between about 15 and 45 s, most

preferably 30 s, with the length of the period of reverse current determined according to the ratios above. If the current flows in the forward direction for longer than 5 min without a period of reverse current, there is a risk that the quality and uniformity of the coating may be degraded. If the current flows in the forward direction for less than 0.5 s before a period of reverse polarity, the rate of deposition is low and the overall coating takes an undesirably long time to form. Most preferably, the

electrodeposition cycle includes a repeated cycle of current flowin in the forward direction for about 30 s followed by current with th reverse polarity for about 10 s.

In some methods the liquid is agitated intermittently during the electrochemical deposition process. This agitation can help to keep the NPs in suspension in the liquid which results in a more uniform coverage of NPs on the surface of the substrate. This agitation is preferably ultrasonic agitation, e.g. using an ultrasonic bath or probe to agitate the liquid, although other agitation methods may also be effective to keep the NPs in

suspension in the liquid. Typically the electric current through the electrochemical cell is switched off during agitation of the liquid to prevent any interference of the agitation with the deposition of a uniform coating; typically adhesion of the NPs to the surface of the substrate is reduced if the agitation is performed at the same time as the electrodeposition process. It is preferred that the period of time for which the liquid is agitated is less than or equal to (preferably less than) the time for which the current flows in the forward direction. This preference is primarily for efficiency reasons; longer periods of agitation are possible and do not have a detrimental effect on the surface coating. However, longer periods of agitation interrupt the electrodeposition process so the whole coating process takes an undesirably long time if the duration of agitation is too long compared to the duration of the electrodeposition. Preferably the ratio of the time for which the current is in the forward direction to the agitation time is between about 1:0.01 - <1, preferably 1:0. - <1, more preferably about 1:0.2 - 0.5 and most preferably about 1:1/3. As noted above, in preferred methods, the current flows in the forward direction for between 0.5 and 5 min, preferably between 1 and 120 s, or between 1 and 60 s, more preferably between about 1 and 45 s, most preferably 30 s, with the length of the period of agitation of the liquid determined according to these ratios.

Preferably, the electrodeposition cycle includes a repeated cycle of current flowing in the forward direction, followed by a period of current flowing with the reverse polarity, followed by a period of agitation of the liquid (with no current flowing) , preferably with the time periods determined by the ranges and ratios described above. This combination of polarity reversal and

intermittent agitation provides good coverage of a highly uniform layer of the composite coating. In most preferred methods, the electrodeposition cycle includes a repeated cycle of current flowing in the forward direction for 30 s, followed by a period of current flowing with the reverse polarity for 10 s, followed by a period of agitation of the liquid (with no current flowing) for 10 s.

In an alternative method the electrodeposition cycle includes a repeated cycle of a first period of current flowing in the forward direction alternating with a period of agitation (with no current flowing) followed by a second period of current flowing in the reverse direction alternating with a period of agitation (with no current flowing) . A preferred example of this method may use a repeating cycle of a first period of 40s comprising 5s agitation alternating with 5s forward current, alternating with a second period of 20s comprising 5s agitation alternating with 5s reverse current .

Both method steps of agitation of the liquid to maintain the NPs in suspension and the occasional reversal of the polarity of the electrodeposition current tend to improve the surface area and quality of the deposited composite coating. In some situations, if the deposition occurs without agitation or reversal of the

electrical polarity, the composite coating tends to lack uniformity and can flake away from the substrate surface.

Straying outside the preferred electrodeposition parameters may also lead to the NPs being "overgrown", i.e. covered with a layer of metal from the electro-deposition of the second metal ions from the liquid. This overgrowth tends to reduce the overall available surface area for reaction and also covers the NPs which means that the any surface of the NPs that is overgrown is not available as a reactive surface. This can negate or reduce the effectiveness of any careful selection of the composition of the NPs to target particular contaminants so impairing performance of the coating .

In the present methods, the first metal or first metal oxide, the second metal and NP sizes are as described in relation to the composite layers above. In methods where the first metal (forming metallic NPs) is ferromagnetic, the deposition of NPs on the substrate surface can be enhanced by placing a magnet inside the substrate which attracts the ferromagnetic NPs towards the surface of the substrate. Therefore, some methods also include the step of providing a magnet inside the substrate during the electrochemical deposition. In preferred methods, the magnet is an electromagnet because these typically provide a stronger attractive magnetic force than permanent magnets .

The performance of the composite layer (e.g. composite metal layer) on the surface of the substrate can also be enhanced by annealing the layer in a vacuum or inert atmosphere following deposition. This annealing step can improve the crystallinity of a deposited composite coating and can improve the electron transfer between the coating and the substrate surface. It also has the effect of driving impurities in the originally deposited composite coating to the grain boundaries in the metallic layer where they are concentrated which means that the metal layer between the grain boundaries is improved in purity.

The annealing preferably involves heating the composite layer in a vacuum (e.g. <lxlO ^~3 mbar and preferably <lxl0 ^~4 mbar) or inert atmosphere (such as He, Ne, Ar or dry N ₂ ) at a temperature up to the lower melting point of the NPs and second metal layer of the composite layer. Preferably the annealing temperature is between about 200°C and 50°C below the lower melting point of the NPs and second metal layer. In preferred methods the annealing temperature between about 100 °C and about 1000 °C, more preferably between about 200°C and about 800°C, most preferably between about 300°C and about 600°C, such as about 500°C. The annealing time is not particularly critical but is typically at least about 2 hours, preferably at least about 6 hours, more preferably at least about 12 hours, most preferably between about 12 and 24 hours. Less than 2 hours may not be sufficient time to achieve the benefits mentioned above, e.g. improved crystallinity of the metal layer; and annealing times of more than 24 hours may not achieve significantly better results but do incur additional production costs.

The performance of the composite layer on the surface of the substrate can also be enhanced by oxidation of the second metal layer. The degree of oxidation of the second metal layer can be varied depending on the target contaminant for filtration.

In some methods oxidation of the second metal layer produces a composite layer with a second metal oxide layer. In other methods oxidation of the second metal layer produces a composite layer with a second layer comprising a second metal and a second metal oxide.

The second metal oxide may be any oxide of the second metals described above, for example a transition metal oxide.

A further aspect of the present proposals is the use of a filter element of these proposals or a coated substrate formed by the present methods, in a method of treating a fluid to remove contaminants. These proposals also include a filter comprising one or more filter elements as described herein or coated substrates formed by a method as described herein.

The filter typically includes one or more filter elements as described herein or coated substrates formed by the present methods, and a filter housing which has a fluid inlet, a fluid outlet and a fluid flow pathway between the two, the fluid flow pathway passing through a filtering region which contains the one or more filter elements as described herein or coated substrates formed by the present methods. Where the substrate of the filter element is a discrete porous substrate body, the fluid flow pathway typically passes through that body in the filtering region. Where the substrate is a coated particulate substrate, the fluid flow pathway typically passes through a region of packed coated particles in the filtering region.

The filters of the present proposals may also contain

additional filter components (such as screen filters) upstream of the filtering region to remove particulate impurities from the incoming fluid prior to treatment with the filter elements and coated substrates described herein.

In all of the filters and filtering methods described herein, the fluid is preferably a liquid and most preferably is water.

Typically the method of treating a fluid to remove

contaminants involves flowing a fluid over the composite coating on the surface of the substrate. As mentioned above, the fluid may be pre-filtered before passing over the coated surface to remove particulate contaminants .

In the most preferred methods and uses described herein, the fluid is water and the filter elements and substrates coated according to the methods described can remove contaminants from the water. The present proposals provide effective methods and products for the removal of contaminants from aqueous systems. The

contaminant may, for example be present in, groundwater flow, drinking water, water treatment plants, industrial plant waters both pre- and post-treatment, or mine tailings. For example, a possible application includes the removal of Uranium from deep mine waters, by the adsorption and chemical reduction of U ⁶⁺ U ⁴⁺ on a substrate coated with a composite coating in which both first and second metals are Fe .

The contaminants that can be removed depend on the particular coating, e.g. on the exact composition of the NPs and the metal layer between them, and the coatings can be tailored to be most effective at removal of particular targeted contaminants. However, some typical contaminants that may be removed from water systems include heavy metals (such as Cd, Hg, Cr, Cu, U, As, Sn, and Pb) , pesticides, residues and metabolites from pharmaceuticals (such as hormones e.g. estrogen), compounds that interfere with the endocrine system in animals, e.g. any endocrine disruptor compounds, waste products from industrial processes, bacteria (such as Pseudomonas aeruginosa or Escherichia coli) , and many others.

The present proposals also include the regeneration or restoration of filters, filter elements and coated surfaces formed according to the present methods. When the composite coatings of these proposals have been used to remove contaminants from a fluid system, the composite coating can become occluded by the products of the reaction between the coating and the contaminants or can show a reduced activity for removing contaminants. Many filter systems simply have to be discarded at this stage because it is not possible (or is commercially not viable) to remove the surface contamination to restore or revive the contaminant-removing activity. The electrodeposition methods used to form the present composite coatings mean that it is straightforward to remove the coating by reverse electroplating. This involves placing the contaminated surface or filter element in the same apparatus as used for the present coating methods and applying an electric current with the opposite polarity to that used for the plating step, i.e. wherein the contaminated, coated substrate is the anode in the

electrochemical cell and an electrolyte liquid is provided.

Typically the electrolyte liquid is water.

This reverse electroplating removes the composite surface from the substrate along with any contaminants deposited on the composite surface. An advantage of this regeneration is for systems where the composite surface has been used to filter out valuable components from a fluid system. In that case, the reverse electroplating can release the valuable components along with the composite surface components into the electrolyte form which they can be chemically isolated or simply dried into a solid deposit which may in itself have commercial value. For example, this could be used in the mining industry as a method for concentrating minerals or a

desirable extract from a fluid stream onto the composite surface and then releasing the desired extracted product in purified form from the composite surface by reverse electroplating.

The substrate can of course be re-coated (using the present methods) and re-used. The ability to perform this reverse electroplating method to remove the coating and associated contaminants is due to the same physical properties that are required for the electroplating methods. So the use of an electrically conductive substrate makes this reverse electroplating step possible.

Therefore, the present methods may also include a step of removing the composite surface and any associated contaminants by reverse electroplating and isolating the contaminants from the electrolyte solution.

In respect of numerical ranges disclosed in the present description it will of course be understood that in the normal way the technical criterion for the upper limit is different from the technical criterion for the lower limit, i.e. the upper and lower limits are intrinsically distinct proposals.

EXAMPLES

The following examples are provided to illustrate the present proposals and do not limit their scope.

Example 1

A 10 x 10 x 30 mm carbon foam block (commercially available from ERG Materials and Aerospace corporation (Oakland, CA, USA) was suspended in FeCl ₃ , 2 g/L salt solution in a 316 stainless steel beaker. 0.2 g of bimetallic nanoparticles (80% Fe, 20% Ni) were added to the electrolyte solution (a concentration of 2 g/L Fe NPs in the electrolyte solution) . A current of 0.3 A was applied for 0.5 minutes, with ultrasound sonication in place continuously to keep the nano particles in suspension. The substrate was removed from solution, washed in high purity methanol and stored under ultra-high vacuum. The surface of the carbon foam block is shown in fig 1.

Example 2

A carbon foam block was treated as in Example 1. However during the electrochemical treatment, ultrasound sonication and electric current were pulsed alternately with 5 s of electro- deposition followed by 5 s of sonication for a total treatment time of 60 s. The surface of the carbon foam block is shown in fig 2.

In fig 2 the improved NP morphology and uniformity can be seen compared to fig 1 where the electro-deposition current was not stopped during the ultrasound sonication. However, the NPs (1) are still largely "overgrown", i.e. covered with a layer of metal from the electro-deposition from the metal ions in the liquid. This overgrowth tends to reduce the overall available surface area for reaction .

Example 3

A carbon foam block was treated as in Example 1. However during the electrochemical treatment the forward and reverse polarity current periods were interspersed with periods of

ultrasound sonication. The overall sequence of steps was:

5 seconds ultrasound sonciation

5 seconds electrodeposition (forward polarity current)

5 seconds ultrasound sonication

5 seconds electrodeposition (forward polarity current)

5 seconds ultrasound sonication

5 seconds reverse plating

5 seconds ultrasound sonication

5 seconds electrodeposition (forward polarity current)

5 seconds ultrasound sonication

5 seconds electrodeposition (forward polarity current)

5 seconds ultrasound sonication

5 seconds reverse plating

This sequence of steps was repeated for a total of 5 minutes.

The forward polarity current was 0.3 A and the reverse polarity current was 0.3 A. No current flowed during the ultrasound sonication steps. The surface of the carbon foam block is shown in fig 3.

Further improvements to the surface coating are seen in fig 3 which shows more discrete NPs (2) with fewer NPs that have been "overgrown" (1) . Example 4

A carbon foam block was treated as in Example 3. However, following the electro-deposition treatment, the treated carbon foam was vacuum annealed at 500 °C for 12-24 hrs at a vacuum pressure of <1 x 10 ^~4 mbar . The surface of the treated carbon foam block is shown (at various magnification levels) in fig 4 and shows no significant change in surface structure or topography compared to the surface prior to vacuum annealing. However, measurements of surface chemistry using X-ray photoelectron spectroscopy (XPS) demonstrate a surface refinement, wherein the proportion of Fe(II) in the surface oxide film is significantly increased as shown in Table 1. Previous work on vacuum annealed iron nanoparticles has demonstrated that this refinement in surface chemistry improves the reactivity of the iron-based material with aqueous contaminants. The XPS system was a Thermo Fisher Scientific Escascope equipped with a dual anode X-ray source (Α1 _Κα 1486.6 eV and Mg _Ka 1253.6 eV) . Samples were analysed under high vacuum (<5xl0 ^~8 mbar) with Α1 _Κα radiation at 400W(15 kV, 23 mA) .

Table 1. Comparison between pre-annealed and post-annealed samples showing significant changes in surface Fe(II) content of the oxide. The loss of metallic iron from the XPS signal indicates oxide thickening/restructuring on the foam surface

Fig. 5 shows a STEM image of a carbon foam surface treated according to the method described in Example 4 showing good surface coverage with a coating including well defined, discrete NPs (2) .

Fig. 6 shows a bright-field STEM image of a cross-section through a carbon foam surface (3) treated according to the method described in Example 4 also showing a cross section through the nanoparticles deposited on the surface (3) . The discrete NPs (1) can be clearly seen deposited on the surface (3) .

Example 5

The treated carbon foam block formed in Example 4, was suspended in 200 ml of various aqueous solutions containing the following contaminants: Cr lOppm; Cu lOppm; and As lOppm, for 300 minutes at an ambient room temperature.

After a treatment time of 30 mins, liquid samples were prepared for analysis by inductively coupled plasma atomic emission spectroscopy

(ICP-AES) by diluting 10 times in 1% nitric acid (analytical quality concentrated HN0 ₃ in Milli-Q water) . Blanks and standards for analysis were also prepared in 1% nitric acid. ICP-AES analysis

(Jobin Yvon Ultima ICP-AES (sequential spectrometer) fitted with a cyclone spray chamber and a Burgener Teflon Mira Mist Nebulizer) showed contaminant levels below the detection limit of the machine

(thought to be about 10 ppb) showing that the contaminants were very effectively removed.

Example 6

A 10 x 10 x 30 mm carbon foam block (commercially available from ERG Materials and Aerospace corporation (Oakland, CA, USA) was suspended in FeCl ₃ ,5 g/L salt solution in a 316 stainless steel beaker. 5g/L of Fe nanoparticles were added to the electrolyte solution. The carbon foam block was treated electrochemically .

During electrochemical treatment the polarity of the current was reversed and the forward and reverse polarity current periods were interspersed with periods of ultrasound sonication. The overall sequence of steps was:

5 seconds ultrasound sonciation

5 seconds electrodeposition (forward polarity current)

5 seconds ultrasound sonication

5 seconds electrodeposition (forward polarity current)

5 seconds ultrasound sonication

5 seconds reverse plating This sequence of steps was repeated for a total of 3 times.

The forward polarity current was 0.1 A (current density of 0.03 Acrrf ² ) and the reverse polarity current was 0.1 A (current density of 0.03 Acrrf ² ) . The substrate was removed from solution, washed in high purity methanol and stored under ultra-high vacuum. The surface of the carbon foam block is shown in Fig. 7a.

Example 7

A carbon foam block was treated as in Example 6. However, the forward polarity current and the reverse polarity current used during electrochemical treatment were 1 A (current density of 0.3 Acrrf ² ) . The surface of the carbon foam block is shown in Fig. 7b.

Figures 7a and 7b show the effect of current used during

electrochemical treatment. The surface of Figure 7b was formed using a high current of 1 A, cracks in the surface can be seen.

Figures 7a and 7b show that the current used must be high enough to draw the ions to the surface of the substrate; however H ⁺ ions may also be drawn to the cathode and result in the generation of hydrogen gas. This evolution of gas can cause cracks in the surface and hinder even deposition of the composite layer.

Example 8

A 10 x 10 x 30 mm carbon foam block (commercially available from ERG Materials and Aerospace corporation (Oakland, CA, USA) was suspended in AgN, 5g/L salt solution in a 316 stainless steel beaker. 5g/L of Ag nanoparticles were added to the electrolyte solution. The carbon foam block was treated electrochemically .

During electrochemical treatment the polarity of the current was reversed and the forward and reverse polarity current periods were interspersed with periods of ultrasound sonication. The overall sequence of steps was:

5 seconds ultrasound sonciation

5 seconds electrodeposition (forward polarity current)

5 seconds ultrasound sonication

5 seconds electrodeposition (forward polarity current) 5 seconds ultrasound sonication

5 seconds reverse plating

This sequence of steps was repeated a total of 3 times.

The forward polarity current was 0.3 A and the reverse polarity current was 0.3 A. No current flowed during the ultrasound sonication steps. The substrate was removed from solution, washed in high purity methanol and stored under ultra-high vacuum. The surface of the carbon foam block is shown in Fig. 8a.

Example 9

A carbon foam block was treated as in Example 8. However, the salt solution used was PtCl and the nanoparticles added to the electrolyte solution were Pt NPs . The surface of the carbon foam block is shown in Fig. 8b.

Example 10

A carbon foam block was treated as in Example 8. However, the nanoparticles added to the electrolyte solution were Ti0 ₂ NPs. The surface of the carbon foam block is shown in Fig. 8c.

Example 11

A carbon foam block was treated as in Example 8. However, the salt solution used was FeCl ₃ and Pt and Fe nanoparticles were added to the electrolyte solution. The surface of the carbon foam block is shown in Fig. 8d.

Fig. 8d shows that for this mixed phase PtNP/INP system a uniform structure was also observed.

Example 12

A carbon foam block was treated as in Example 3.

However, a ratio of 0.8 Fe salt: 0.2 Fe NPs (8g/L : 2g/L) was used. The surface of the carbon foam block is shown in Fig. 9a.

Example 13 A carbon foam block was treated as in Example 3.

However, a ratio of 0.2 Fe salt: 0.8 Fe NPs (2g/L : 8g/L) was used. The surface of the carbon foam block is shown in Fig. 9b.

The surfaces of Figures 9a and 9b were formed under the same conditions as the surface of figure 3 except for the ratio of Fe salt to Fe NPs used. The surface of Fig. 9a is much smoother, and therefore has a much smaller surface area, than that of the surface of Fig. 3 due to the low ratio of the concentration of NPs to concentration of metal salt used in the liquid. It is thought that the low concentration of NPs provide only a small number of

nucleation sites which results in the smoother surface of Fig. 9a. The surface of Fig. 9b has conglomerates of fluff like structures on its surface. These structures are the result of uneven deposition caused by the high ratio of concentration of NPs to concentration of metal salt in the liquid in the method of Example 13.

Example 14

Filter substrate disks, 9mm thick and 47mm in diameter, of Reticulated Vitreous Carbon (RVC) foam (ERG Aerospace, U.S.A.) with a porosity of 45 pores per inch (PPI), were washed ultrasonically in acetone and MilliQ water as a preparatory cleaning step.

The RVC foam disks were maintained in contact with a stainless steel plate forming the cathode. A stainless steel beaker acted as the anode and container for an electrolyte solution, which was a combination of 0.4g FeCl ₃ (99.9%, Sigma) and lg Iron Nanoparticles (INP) s (Nanolron Ltd., Czech Republic) in 200ml deionised water (pH4) . The beaker was held within a sonic bath to assist suspension of the INPs and prevent them aggregating. The Experimental

apparatus is shown schematically in Fig. 10.

Before the INPs were added, the electrolyte was purged of oxygen using argon to reduce the oxidation of the INPs. The RVC disk was sonicated in the electrolyte to allow the INPs to penetrate the foam before beginning the deposition. Electrodeposition was performed using a 9 minute treatment consisting of 40s of 5s sonication and 5s forward current pulses, alternating with 20s of 5s sonication and 5s reverse current pulses. The RVC substrate was then turned over and the cycle repeated. The samples were rinsed with acetone to remove any excess INPs and they placed immediately in a vacuum desiccator to dry.

Control samples were formed using electrolyte solutions void of nanoparticles (0.0086mol FeCl ₃ ) under the same treatment

conditions. Resulting nano-structures were analysed using Scanning Electron Microscopy.

The formation of the nano-coating was found to be influenced by the electrodeposition conditions. Current density was identified as an important variable; increased potential difference between electrodes causes an increase in ion/INP flux due to the larger electrostatic attraction. Another important variable is the length of electrodeposition; longer deposition times, although producing a thicker layer, forfeit the structural integrity of the coating and it becomes easily detached from the substrate. The presence of both iron salt and INPs in the electrolyte is also important; the iron ions cement the INPs to the surface, whilst the INPs cause

heterogeneous iron crystal growth.

Sonication of the electrolyte improves nano-particle

suspension and reduces aggregation and sedimentation without introducing a surfactant. However, sonicating whilst deposition is taking place can cause ions and INPs to be drawn away from the vicinity of the cathode, lowering deposition. This is unfavourable as it is thought the electrostatic attraction of the cathode is relatively short range for INPs.

The process is therefore improved by the use of alternating pulses of sonication and electrodeposition. Finally, by introducing pulses that reverse the polarity of the electrodes, there is a reduction in uneven accumulation of the coating and the nano- structures produced are smaller by drawing away some of the iron for short periods. This electrodeposition process creates a nano-structured layer on the surface of the carbon throughout the porous structure (Fig. 11) . Fig. 11 shows SEM images of the nano-structure on a top surface of the treated RVC substrate (top and bottom left) and a section approximately 3.8mm from the top surface (top and bottom right) . The coating consists primarily of quasi spherical features, approximately 20nm in size, created by the deposition of INPs cemented to the surface by the dissociated iron from FeCl ₃ . These features are interspersed by faceted metallic crystals, resulting from the INPs acting as nucleation sites for heterogeneous crystal growth .

Without INPs in the solution the iron coating appears

amorphous at this scale because the smooth glassy carbon limits heterogeneous crystal growth, resulting in a smooth and uniform film coating of iron.

Example 15

A sample of the coated RVC foam formed in Example 14 was vacuum annealed at 600°C for 24hrs in a vacuum. XPS analysis was performed both before and after the vacuum annealing to investigate the change in physicochemistry .

The vacuum annealing process improves the structure of both the metallic NPs that are deposited and any oxide layer on the outside of the NPs. Vacuum annealing of iron NPs causes the metallic core to recrystallise into a more ordered structure, grain sizes to increase and impurities to migrate to the surface and grain boundaries. The surface oxide is also significantly improved with an improvement in crystallinity to a thinned uniform and conductive magnetite (Fe ₃ 0 ₄ ) layer. The improved layer allows for more

efficient electron transfer through the oxide layer to the surface sorbed contaminants. The vacuum annealed iron NPs therefore demonstrate improved reactivity whilst reducing the aqueous

corrosion of the metallic iron.

The improvements in the oxide layer can be seen to occur when the nano-composite is vacuum annealed (Fig 12) . Fig. 12 shows an XPS spectrum of Fe 2p3/2 peak with fitted curves for before (top) and after (bottom) vacuum annealing. The ratio of Fe(II) :Fe(III) increases (from 0.160 to 0.373) in accordance with the thermal decomposition of Fe(III) . This physiochemical improvement implies that vacuum annealing the nano-composite will increase the

reactivity with contaminants as it does with iron NPs.

Example 16

Organic contaminant sorption tests were performed by ChemTest Ltd. to compare the remedial abilities of the sorbents Granulated Activated Carbon (GAC) , Organo-clay (OC) and the vacuum annealed nano-composite product of Example 15 (i.e. the product of Example 14 after being subjected to the vacuum annealing described in Example 15) . A groundwater sample containing benzene, toluene, ethyl- benzene, and m-, p-, o-xylene (the BTEX group) was abstracted from a contaminated site in Portsmouth, UK.

Kinetic experiments were conducted using continuously stirred 1000ml glass sealed reactors at a suspension density of 0.4gl ^_1 .

Samples were taken at time points (t) = 5, 10, 30, 90, 360, 1440 minutes. Supernatants were separated by centrifugation at 10,000 rpm. Residual solution chemistries were determined by headspace GC- MS for volatile organic compounds.

Data from ChemTest Ltd., normalised by the surface area of the reactive material, shows that the iron nanocomposite (INC)

significantly outperforms the comparator materials (Fig. 13) . Fig. 13 shows percentage of organic removed normalised by surface area of reactive material (INC 12m ² g ^_1 , GAC 65m ² g ^_1 , OC 825m ² g ^_1 ) .

The mechanism involved for BTEX removal onto the INC is thought to be adsorption followed by degradation caused by the organic-iron coordination weakening of the benzene ring. This is compared to OC and GAC which are typically thought to remove organics via adsorption alone.