This patent application claims priority from Australian Provisional Patent Application No. 2010900194 titled "Low-Fouling Filtration Membranes" and filed 19 January 2010, the entire contents of which are hereby incorporated by reference. FIELD

The present invention relates generally to filtration membranes, and more specifically to reverse osmosis membranes.

BACKGROUND

Osmosis is a natural phenomenon in which a solvent (usually water) passes through a semipermeable membrane from a side of the membrane having a relatively low solute concentration to a side of the membrane having a higher solute concentration. Water flows through the membrane until chemical potential equilibrium of the water is established. At equilibrium, the pressure difference between the two sides of the membrane is equal to the osmotic pressure of the solution. To reverse the flow of water, a pressure difference greater than the osmotic pressure difference is applied and, as a result, separation of water from the solution occurs as pure water flows from the high concentration side to the low concentration side. This is known as "reverse osmosis" or "RO".

Reverse osmosis is commonly used for the desalination of seawater and brackish water. Other applications for reverse osmosis include wastewater treatment, aquifer storage and recovery, production of ultrapure water, water softening, and food and pharmaceutical processing. Water purification processes that use reverse osmosis have several advantages over traditional separation processes such as distillation, extraction, ion exchange, and adsorption. For example, reverse osmosis is a pressure driven process and, as a result, is more cost effective. Furthermore, no potentially expensive solvents or adsorbents are needed. Also, reverse osmosis is a process that is inherently simple to design and operate compared to many traditional separation processes. . Reverse osmosis separations are carried out by causing water to flow through a non-porous active layer (a "reverse osmosis membrane"). Reverse osmosis membranes consist of a dense barrier layer which only allows water to pass through whilst preventing the passage of solutes (such as salt ions). Traditionally, reverse osmosis membranes are made from cellulose acetate ("CA"). These CA membranes are sometimes referred to as Loeb-Sourirajan asymmetric membranes and are produced via a phase inversion production process. More recently, thin film composite ("TFC") membranes were developed for reverse osmosis (RO) by Cadotte et a/. ¹ and have become more widely adopted, principally because they can tolerate wider pH ranges, higher temperatures, harsher chemical environments and they have improved water flux and solute separation characteristics. Indeed, the reverse osmosis membrane of choice worldwide nowadays is the polyamide (PA) thin film composite membrane. The PA composite membrane is made by forming a thin PA film on the finely porous surface of a polysulfone (PS) supporting membrane. Despite the widespread adoption of reverse osmosis technology several major challenges remain.

Membrane fouling continues to be problem. Membrane fouling is an 'irreversible' deposition, accumulation and adsorption of solutes or particles onto surface of the membranes. One of the more problematic foulants is biofouling, caused by the attachment of microorganisms to the membrane surface and the subsequent growth of colonies on the surface. The microorganisms and their secretions of extracellular polymeric substances (EPS) form a biofilm that is stabilised by weak physico-chemical interactions including electrostatic interactions, hydrogen-bonding and van der Waals interactions. This biofilm increases the fluid friction resistance and the overall hydraulic resistance of the membrane. This leads to a reduction in water flux, requiring higher pressures to maintain flow, demanding increased energy usage. In the past few years several variations of TFC membranes have developed in an attempt to reduce/prevent biofouling- While these improvements have led to reductions in biofouling, truly biofouling-resistant membranes are yet to be realised.

A number of approaches to treating biofouling of membranes have been made, but all of them face complications. One such approach is to kill the organisms but this does not help if the biomass is left behind. As such, removal of biomass rather than killing the microorganisms is likely to be a more effective way of dealing with biofouling. However, different microorganisms stick to membranes using different interactions, which consequently means a range of treatment methods are required. These can include surfactants to help remove fouling attached using van der Waals interactions, ionic compounds to break up weak electrostatic interactions, and enzymes to cleave the macromolecules that make up the EPS. All of these biofouling amelioration methods ave costs in terms of money and time and there is a need for methods to prevent or reduce biofouling from occurring in the first place.

Prevention of biofouling of polymeric surfaces has been achieved using non-ionic, hydrophilic polymeric materials to modify the surface of the membrane. For example, poly(ethylene glycol) (PEG) or oligo(ethylene glycol) (OEG) have been shown to improve resistance to nonspecific protein absorption. ^2"4 Unfortunately, PEG polymers are not stable and are easily oxidised in the presence of oxygen or transition metal ions, both of which are present in reverse osmosis filtrations. In addition, the protein resistance is lost above 35°C, which suggests that an alternative material must be sought. Similarly phosphorylcholine (PC)-based polymers ^5"7 have been shown to decrease biopolymer adsorption, but these polymers are also fragile; the phosphoester groups being readily hydrolysable.

There is a need for alternative anti-fouling materials for reverse osmosis membrane surfaces. SUMMARY

The present invention arises from research into sulfobetaine containing polymers and derivatives thereof for use on membrane surfaces to reduce or eliminate biofouling. This research has led to the discovery of reaction conditions suitable for the grafting of sulfobetaine, its derivatives and other zwittenon containing polymers onto membrane surfaces, for example reverse osmosis membrane surfaces, under conditions suitable for the commercial production of anti-fouling filters.

In a first aspect, the present invention provides a composite filtration membrane comprising:

- a porous support;

- a polyamide layer on the porous support; - optionally, a polyhydroxy layer on the polyamide layer; and

- an anti-fouling coating grafted on to the polyamide layer or the polyhydroxy layer, the anti- fouling coating comprising a zwitterionic polymer.

In a second aspect, the present invention provides a process for producing a filtration membrane having an anti-fouling coating, the process comprising: - providing a filtration membrane having amine or hydroxy functionality on a surface thereof; treating the filtration membrane with an agent having atom transfer radical polymerisation (ATRP) initiator functionality, said treatment carried out under conditions to provide an ATRP initiator functionalised filtration membrane; contacting the ATRP initiator functionalised filtration membrane with a ligand stabilised metal ion catalyst capable of participating in a one electron redox reaction with the ATRP initiator, a reducing agent, and a radically polymerisable zwitterionic monomer under ARGET (activators continuously regenerated by electron transfer) ATRP conditions to form a filtration membrane having a zwitterionic polymer coating grafted onto the membrane.

In a third aspect, the present invention provides a process for functionalising a filtration membrane, the process comprising: providing a filtration membrane having amine or hydroxy functionality on a surface thereof; treating the filtration membrane with an agent having atom transfer radical polymerisation ATRP initiator functionality, said treatment carried out under conditions to provide an ATRP initiator functionalised filtration membrane; and

- contacting the ATRP initiator functionalised filtration membrane with a ligand stabilised metal ion catalyst capable of participating in a one electron redox reaction with the ATRP initiator, a reducing agent, and a radically polymerisable zwitterionic monomer under ARGET (activators continuously regenerated by electron transfer) ATRP conditions to form a filtration membrane having a zwitterionic polymer coating grafted onto the membrane.

In a fourth aspect, the present invention provides a filtration membrane having a coating formed by the process of either the second or the third aspect of the invention.

In some embodiments, the step of treating the filtration membrane with an agent having ATRP initiator functionality is carried out in a hydrocarbon solvent. In some specific embodiments, the hydrocarbon solvent is hexane. In some embodiments, the step of treating the filtration membrane with an agent having ATRP initiator functionality is carried out in an ether solvent. In some specific embodiments, the ether solvent is diethyl ether.

In some embodiments, the filtration membrane is treated with water to produce a hydrated membrane which is then subjected to treatment with the agent having ATRP initiator functionality.

In some embodiments, the ATRP initiator functionalised filtration membrane formed has an atomic fraction of bromine on the surface (as determined by XPS) of greater than about 0.7%. In some embodiments, the ATRP initiator functionalised filtration membrane formed has an atomic fraction of bromine on the surface of greater than about 1%. In some embodiments, the ATRP initiator functionalised filtration membrane formed has an atomic fraction of bromine on the surface of greater than about 1.5%. In some embodiments, the ATRP initiator functionalised filtration membrane formed has an atomic fraction of bromine on the surface of greater than about 2%. In some embodiments, the zwitterionic polymer is selected from the group consisting of sulfobetaine polymers, phosphobetaine polymers, carboxybetaine polymers, and derivatives of any of these polymers. In some specific embodiments, the zwitterionic polymer is a sulfobetaine polymer or derivative thereof.

In some embodiments, the zwitterionic polymer is formed by polymerisation of a sulfobetaine, phosphobetaine or carboxybetaine containing monomer in the presence of the polyamide layer using an activator regenerated by electron transfer ATRP (ARGET ATRP) polymerisation procedure. In some embodiments, the transition metal catalyst is added in an amount of about O.OlmM to about 0.9mM. In some embodiments, the transition metal catalyst is added in an amount of about O.lmM to about 0.8mM. In some embodiments, the transition metal catalyst is added in an amount of about O.lmM to about 0.7mM. In some embodiments, the transition metal catalyst is added in an amount of about 0. ImM to about 0.6mM. In some embodiments, the transition metal catalyst is added in an amount of about O.lmM to about 0.5mM. In some embodiments, the transition metal catalyst is added in an amount of about O.lmM to about 0.4mM. In some embodiments, the transition metal catalyst is added in an amount of about 0.2mM to about O.SmM. In some embodiments, the transition metal catalyst is added in an amount of about 0.3mM to about 0.4mM. In some embodiments, the transition metal catalyst is added in an amount of about 0.33mM. In some embodiments, the transition metal catalyst is added in an amount of about O.OlmM to about 0.08mM. In some embodiments, the transition metal catalyst is added in an amount of about O.OlmM to about 0.07mM. In some embodiments, the transition metal catalyst is added in an amount of about O.OlmM to about 0.06mM. In some embodiments, the transition metal catalyst is added in an amount of about O.OlmM to about 0.05mM. In some embodiments, the transition metal catalyst is added in an amount of about 0.02mM to about 0.05mM. In some embodiments, the transition metal catalyst is added in an amount of about O.OSmM to about 0.09mM. In some

embodiments, the transition metal catalyst is added in an amount of about 0.04mM.

In some embodiments, the filtration membrane is a reverse osmosis membrane. In some embodiments, the reverse osmosis membrane is a polyamide thin film composite membrane. In some other embodiments, the reverse osmosis membrane is a polyamide thin film composite membrane comprising a polyhydroxy layer (such as a polyvinyl alcohol layer) on the polyamide layer. In some other embodiments, the membrane is a cellulose acetate containing membrane.

BRIEF DESCRIPTION OF THE ACCOMPANYING DRAWINGS/FIGURES

Figure 1 is a schematic showing a comparison between cross-flow and dead-end filtration

apparatus.

Figure 2 is a plot of functionalisation time (hours) vs atomic fraction of bromine (%).

Figure 3 is a plot showing the analysis of a sulfobetaine coated membrane by TGA.

Figure 4 shows an FTIR spectrum of a sulfobetaine coated membrane.

Figure 5 shows SEM images of biofouled reverse osmosis membranes. Figure 6 shows FTIR spectra of RO membranes coated with PolySBMA before (top) and after aquaria-based static test (middle). Figure 7 is a schematic reaction scheme for covalently grafting the low-biofouling polymeric coating from the surface of commercially available RO polyamide membranes.

Figure 8 is a series of photographs showing the contact angle measurements of the membrane surfaces measured using a sessile drop method: (a) uncoated polyamide; (b) BibBr initiator modified; (c) p3SBMA coated.

Figure 9 is a photograph of stirred cell Sterlitech HP4750 set up.

Figure 10 is a plot showing the flux over time for modified and unmodified commercial polyamide

RO membranes (GE AD) in stirred cell experiments.

Figure 11 is a plot showing the flux over time for modified and unmodified commercial polyamide

RO membranes (Nitto Denko SWC6) in stirred cell experiments.

Figure 12 is a plot showing the flux over time for modified and unmodified commercial polyamide

RO membranes (Dow Water SW30HR) in stirred cell experiments.

Figure 13 is a photograph of the six Sterlitech CF042 units used for cross flow experiments.

Figure 14 is a plot showing the flux over time for modified and unmodified commercial polyamide

RO membranes (GE AD) ^' in cross-flow tests.

Figure 15 is a plot showing the flux over time for modified and unmodified commercial polyamide membranes (Nitto Denko SWC6) in cross-flow tests.

Figure 16 is a plot showing the flux over time for modified and unmodified commercial polyamide

RO membranes (Dow Water SW30HR) in cross-flow tests. DETAILED DESCRIPTION

A first aspect of the present invention provides a composite polyamide filtration membrane. The membrane comprises a porous support, a polyamide layer on the porous support, optionally a polyhydroxy layer on the polyamide layer, and an anti-fouling coating grafted on to the polyamide layer or the polyhydroxy layer (if present). The anti-fouling coating comprises a zwitterionic polymer. As used herein, the term "anti-fouling" and like terms when used in relation to a coating means that the coating is capable of reducing biological fouling of a surface relative to a surface that does not have, the coating. Thus, anti-fouling does not necessarily mean that there is no accumulation of fouling organisms and/or associated biofilm forming materials on the surface of the membrane. Biological fouling results from an accumulation of fouling organisms (micro- or macroorganisms) and/or associated biofilm forming materials on a surface. Any of the tests provided herein or known by the skilled person can be used to determine whether or not there is a reduction in biological fouling. For example, direct measurement of microbial growth on the membrane surface can be used to determine whether or not there is a reduction in biological fouling. The anti-fouling coating may be a continuous layer or a discontinuous layer. The coating may cover the whole of the surface to which it is applied or it may cover only part of the surface to which it is applied.

A second aspect of the present invention provides a process for producing a filtration membrane having an anti-fouling coating. The process comprises providing a filtration membrane having amine or hydroxy functionality on a surface thereof; - treating the filtration membrane with an agent having atom transfer radical polymerisation

(ATRP) initiator functionality, said treatment carried out under conditions to provide an ATRP initiator functionalised filtration membrane; contacting the ATRP initiator functionalised filtration membrane with a ligand stabilised metal ion catalyst capable of participating in a one electron redox reaction with the ATRP initiator, a reducing agent, and a radically polymerisable zwitterionic monomer under

ARGET (activators continuously regenerated by electron transfer) ATRP conditions to form a filtration membrane having a zwitterionic polymer coating grafted onto the membrane.

The filtration membrane may be a reverse osmosis membrane. The filtration membrane having amine functionality may be a polyamide thin film composite membrane. A number of reverse osmosis polyamide membranes are available commercially and any of these may be modified using the conditions described herein. For example, GE Sepa™ AD membranes (available from GE Osmonics Labstore) or Hydranautics SWC6 membranes (available from Nitto Denko) may be used. These membranes typically have a first or top layer which is an ultra-thin barrier or discriminating layer typically comprising a crosslinked polyamide of about 200 nanometres thickness. A second or middle layer typically comprises an engineering plastic, such as polysulfone, and it typically has a thickness of about 30 - 60 microns. This second layer provides a smooth surface for the top layer, and it enables the top layer to withstand relatively high operating pressures. A third or bottom layer is typically nonwoven polyester, e.g., a polyethylene terephthalate (PET) web or fabric, with a thickness typically of about 120 microns.

The filtration membrane having hydroxy functionality may be a polyamide thin film composite membrane having a polyhydroxy layer on the polyamide layer. Some polyamide thin film composite membranes comprise a hydrophilic outer coating, which is typically a polyhydroxy layer. For example, the polyamide thin film composite membrane having a polyhydroxy layer on the polyamide layer may be a Filmtec™ SW30HR membrane (available from Dow Water) which has a polyvinyl alcohol layer on the polyamide layer.

Alternatively, the filtration membrane having hydroxy functionality may be a cellulose acetate membrane. These membranes have a layer of cellulose acetate on a polyester fabric. For example GE Sepa™ CF CA membranes (available from GE Osmonics Labstore) can be used.

Reverse osmosis membranes are usually employed in either flat panel or spiral wound configurations. The flat panel configuration is simply the membrane, or more typically a plurality of membranes separated from one another by a porous spacer sheet, stacked upon one another and disposed as a panel between a feed solution and a permeate discharge. The spiral wound configuration is simply a membrane/spacer stack coiled about a central feed tube. Both configurations are well known in the art.

The filtration membrane is treated with an agent having ATRP initiator functionality under conditions to provide an ATRP initiator functionalised filtration membrane. An "ATRP initiator" is a molecule with one or more transferable halogen or pseudohalogens that can initiate radical chain growth. A variety of initiators, typically alkyl halides, have been used successfully in ATRP. As known in the art, many different types of halogenated compounds are potential initiators.

The agent having ATRP initiator functionality that is used in the methods of the present invention is a compound containing functionalities both for surface attachment and ATRP initiation. Attachment of the ATRP initiator to hydroxy and or amine functionalised surfaces can be achieved using acid halides, such as acid chloride, acid bromide or acid iodide. ATRP can be initiated by tertiary halide groups.

In some embodiments, the agent having surface attachment capabilities and ATRP initiator functionality is a derivative of 2-haloisobutyric acid. For example, 2-bromoisobutyryl bromide, 2-chloroisobutyryl bromide or 2-bromoisobutyryl tosylate could be used. In some embodiments, the agent having ATRP initiator functionality is 2-bromoisobutyryl bromide.

The ATRP initiator functionalised filtration membrane can be prepared using known methods. In some embodiments, the unfunctionalised membrane is immersed in a suitable solvent and the agent having ATRP initiator functionality added.

The solvent may be any solvent that does not react with the agent having ATRP initiator functionality. Non-protic solvents are particularly suitable for this purpose. However, care needs to be taken as some solvents may cause delamination or otherwise damage the membrane and, therefore, may not be suitable. Solvents that may damage the membrane include toluene, ethyl acetate, acetone, dimethylsulfoxide, chloroform, dichloromethane, tetiahydrofuran, diethyl ether, and acetonitrile. In the case of a polyamide membrane, the solvent may be a hydrocarbon. For example, we have found that hexane has no apparent effect on the membrane. In the case of cellulose acetate membranes, we have found that either diethyl ether or hexane have no apparent effect on the membrane.

The unfunctionalised membrane may be dry or it may be hydrated by immersion in water, and then blotted dry, prior to immersion in the solvent. We have found that hydration of the membrane may be beneficial prior to attaching the agent having atom transfer radical polymerisation (ATRP) initiator functionality.

The reaction may then be carried out at a temperature and time sufficient to allow reaction with the hydroxy or amine groups on the surface of the membrane. The temperature of the reaction will depend on the solvent used and the reaction time. Generally, the reaction can be carried out at a temperature of between about 0°C and about 100°C. In some embodiments, the reaction is carried out at a temperature between about 15°C and about 40°C. In practice, it may not be possible to carry out the reaction at a temperature higher than about room temperature (i.e. about 30°C) because higher temperatures may damage the membrane. The reaction time will vary depending on a number of factors including the functionality on the surface of the membrane (e.g. amide functionality or hydroxy functionality), the nature of the agent having ATRP initiator functionality, the concentration of the agent having ATRP initiator functionality, the temperature, and the solvent. In some embodiments, the reaction is carried out for a time of between about 5 minutes and about 10 hours. In some specific embodiments, the reaction is carried out for about 3 hours. In these embodiments, the membrane may be one having predominantly hydroxy functionality on the surface. In some other embodiments, the reaction is carried out for a time of between about 5 minutes and about 30 minutes. In some specific embodiments, the reaction is carried out for about 10 minutes. In these embodiments, the membrane may be one having predominantly amide functionality on the surface. The reaction is carried out under an inert atmosphere. Thus, the reaction mixture may be purged with an inert gas (such as nitrogen or argon) prior to addition of the agent having ATRP initiator functionality to the mixture and or the reaction may be maintained under an inert gas atmosphere.

The reaction mixture can be prevented from being acidified by reaction by-products by adding a sterically bulky organic base such as triethylamine or pyridine.

After the reaction is complete, the ATRP initiator functionalised filtration membrane may be removed from the reaction mixture and washed with suitable solvents. If necessary, the functionalised membrane can be stored in a suitable solvent, such as methanol/water.

If required, the density of the grafted polymers formed on the surface can be controlled by controlling the density of the ATRP initiator functionality on the surface using known methods. For example, if only a portion of the surface functional groups are desired to have attached polymers, a blocking agent without initiation functionality may be reacted with the hydroxy and or amine functionalised surfaces. This will prevent foiming an initiation site. Initiation sites may be patterned only on specific areas of the surface if desired.

The ATRP initiator functionalised filtration membrane is then contacted with a ligand stabiUsed metal ion catalyst capable of participating in a one electron redox reaction with the ATRP initiator, a reducing agent, and a radically polymerisable zwitterionic monomer under ARGET ATRP conditions to form a filtration membrane having a zwitterionic polymer coating grafted onto the membrane.

The antifouling coating that is grafted onto the membrane is a zwitterionic polymer. A "zwitterionic polymer" is a polymer composed from zwitterionic monomers and, possibly, other non-ionic monomer(s). Zwitterionic polymers are electrically neutral (i.e. carry no total net charge) but they carry formal positive and negative charges on different atoms in the polymer chains or segments. Zwitterionic polymers may be homopolymers, copolymers, and terpolymers.

Zwitterionic polymers that are recognised biofouling resistant polymers may be used. Examples include sulfobetaine polymers, phosphobetaine polymers, carboxybetaine polymers, and derivatives thereof. Sulfobetaine polymers and derivatives thereof may be particularly suitable because they tend to exhibit strong biocompatibility and consequently may extend the range of applications for which the membranes may be used (for example, biomedicine). Whilst we have found sulfobetaine polymers to be particularly suitable, it is possible that other zwitterionic polymers such as phosphobetaine and carboxybetaine based polymers could also be used. Sulfobetaine polymers have previously been incorporated into polymer substrates. For example, a random copolymer has been synthesised via conventional radical polymerisation of a sulfobetaine monomer, N,N'-dimethyl-N methacryloyloxyethyl-N-(3-sulfopropyl) ammonium (DMMSA), with butyl methacrylate. This copolymer was then blended with the polyethersulfone in order to fabricate a low biofouling ultrafiltration membrane. ⁸ Another method that has been used to incorporate sulfobetaine in ultrafiltration membranes is via the copolymerisation of 2-dimethylamino ethyl methacrylate

(DMAEMA) with acrylonitrile and the subsequent conversion of the DMAEMA groups to sulfobetaine. This copolymer was then blended with polyacrylonitrile to make the final membrane. ⁹ In both cases, a reduction in protein adsorption was observed. In each of these cases, the sulfobetaine polymer was blended with a copolymer to form a polymer blend. Polymer blends of this type may not be ideal for use on filtration membranes as the sulfobetaine groups tend to be blended throughout the polymer and, therefore, not all of the groups are exposed on the surface. Furthermore, blends of this type may have 'pockets' of sulfobetaine which can act as channels in the filtration membrane and this can detrimentally affect the performance of the membrane. Our approach has been to graft the zwitterionic polymer from the surface of the membrane, thus providing a polymer morphology that is different to that provided by polymer blends. Chiang et al. ¹⁰ have reported grafting of sulfobetaine brushes from poly(vinilidene fluoride) PVDF ultrafiltration membrane surface using ATRP. The surface of PVDF was activated first by ozone treatment, followed by preparation of a poly(bromoisobutyryloxy)ethyl acrylate) PBIEA grafted PVDF membrane. The PBIEA-g-PVDF membrane was then used as a macroinitiator for surface copolymerisation of sulfobetaine methacrylate (SBMA) monomer via surface initiated ATRP.

Using the process we have developed, the filtration membrane does not have to undergo any pre- treatment steps. For example, the surface of the membrane does not need to be activated first by treatment with ozone.

The anti-fouling coating that is formed using the processes of the present invention is covalently bound to the amine or hydroxy groups on the surface of the membrane. This irreversible binding means that the anti-fouling properties of the membranes are maintained over time. This is in contrast to some prior art membranes in which the anti-fouling coating dissolves or dissociates from the membrane surface over time. The irreversible binding of the coating to the polyamide or polyhydroxy layer may also overcome any potential toxicity issues relating to the use of certain anti-fouling compounds.

The process for producing a filtration membrane having an anti-fouling coating comprises providing a filtration membrane having amine or hydroxy functionality on a surface thereof, treating the filtration membrane with an agent having atom transfer radical polymerisation (ATRP) initiator functionality under conditions to provide an ATRP initiator functionalised filtration membrane, and contacting the ATRP initiator functionalised filtration membrane with a ligand stabilised metal ion catalyst capable of participating in a one electron redox reaction with the ATRP initiator, a reducing agent, and a radically polymerisable zwitterionic monomer under ARGET (activators continuously regenerated by electron transfer) ATRP conditions to form a filtration membrane having a zwitterionic polymer coating grafted onto the membrane.

As used herein, the terms "grafted", "grafted from", grafted onto", "graft polymers" and like term ^' s mean that a monomer was polymerised or copolymerised with other comonomers to form a homogeneous macromolecule. The macromonomers and comonomers are linked via covalent bonds by a living free radical mechanism instead of simple blending without reaction. The polymerisation process comprises polymerising free radically polymerisable monomers in a polymerisation medium comprising the radically polymerisable monomers, a ligand stabilised metal ion catalyst, the ATRP initiator, and the reducing agent. The polymerisation process is an ATRP controlled radical polymerisation process. As used herein, the term "ATRP" refers to a controlled radical polymerisation described by Matyjaszewski,' ¹ the disclosure of which is hereby incorporated by reference. More specifically, the controlled radical polymerisation process is an activator regenerated by electron transfer atom transfer radical polymerisation (ARGET ATRP) process.

ATRP is a 'living' method of polymerisation that provides controlled chain length and low polydispersity along with well-defined grafts and a well defined final surface for vinyl monomers. ATRP uses an alkyl halide initiator and a ligand stabilised low oxidation state metal ion catalyst to which the halide is exchanged during chain extension reactions in a dynamic equilibrium between a low amount of propagating radicals and large number of dormant species. As a reduced metal complex is integral to this reaction it is susceptible to oxidation and strictly inert conditions must be provided for storage of the complex and for the reaction, a less than desirable requirement for industrial processes. Another factor holding ATRP back from greater use in commercial applications was the stoichiometric amount of catalyst required, in relation to the initiator, for the reaction and subsequent difficulty in removing used catalyst. We have found that ARGET ATRP is a more commercially applicable process. ARGET ATRP uses up to 1000 times less catalyst than ATRP. Furthermore, ARGET ATRP can tolerate a limited amount of air because a reducing agent is included to scavenge the oxygen and other radical inhibitors and regenerate any oxidised catalyst to the appropriate oxidation state. This allows the use of much lower concentrations of catalyst, the activity of which consequently must be increased by use of appropriately activating tetradentate ligands.

The polymerisation medium may include a solvent. Suitable solvents include, for example, pyridine, water, methanol, ethanol, propanol, butanol, pentanol, hexanol, heptanol, cyclohexanol, methylcellosolve, ethylcellosolve, isopropylcellosolve, and butylcellosolve The solvent may be purged with an inert gas such as nitrogen or argon, to remove dissolved oxygen prior to introduction of the reducing agent into the reaction mixture.

In the present case, the zwitterionic polymer is formed by polymerisation of a sulfobetaine,

phosphobetaine or carboxybetaine containing monomer in the presence of the membrane surface containing hydroxy or amine groups.

Free radically polymerisable monomers for producing sulfobetaine polymers may be selected from the group consisting of: sulfobetaine methacrylates (SBMA), sulfobetaine acrylates, sulfobetaine

acrylamides, sulfobetaine vinyl compounds, sulfobetaine epoxides, and other sulfobetaine compounds with hydroxy, isocyanates, amino, or carboxylic groups.

Free radically polymerisable monomers for producing carboxybetaine polymers may be selected from the group consisting of: carboxybetaine methacrylates, such as 2-carboxy-N,N-dimethyl-N-(2'- methaciyloyloxyethyl)ethananiinium inner salt; carboxybetaine acrylates; carboxybetaine acrylamides; carboxybetaine vinyl compounds; carboxybetaine epoxides; and other carboxybetaine compounds with hydroxy, isocyanates, amino, or carboxylic groups.

The biologically compatible sulfobetaine groups (in the case of sulfobetaine polymers) may be incorporated into the entire backbone, a single block, multiple blocks, branches or in more than one part of the polymer.

In the processes described above, the sulfobetaine groups are part of the monomers used to produce the polymer. In an alternative method, the biologically compatible sulfobetaine groups could also be attached to the terminus after polymerisation by known chemical modification techniques.

Additionally, further non-polymerisation reactions may be performed on the attached polymer, such as, for example, cross-linking, reactions to modify the hydrophobicity of the surface, adding reactive sites, etc.

The ligand stabilised metal ion catalyst is in the form of a complex comprising a transition metal that is initially in an oxidatively stable higher oxidation state, and a ligand. Suitable transition metals include transition metals from Ti of atomic number 22 to Zn of atomic number 30. The metals Fe, Co, Ni, Cu are particularly suitable. In some embodiments, the transition metal is Cull. Cuprous chloride and cuprous bromide may be used. Advantageously, the processes described herein utilise very small amounts of transition metal (e.g. Cu) and, as a result, the concentration of transition metals in the coated membranes is minimal. This is in contrast to some prior art coating processes that utilise higher amounts of transition metal which tends to be difficult to remove from the coated membrane. In some embodiments, the transition metal catalyst is added in an amount of about O.OlmM to about 0.9mM. In some embodiments, the transition metal catalyst is added in an amount of about 0.1 mM to about 0.8mM. In some embodiments, the transition metal catalyst is added in an amount of about 0.1 mM to about 0.7mM. In some embodiments, the transition metal catalyst is added in an amount of about 0.1 mM to about 0.6mM. In some embodiments, the transition metal catalyst is added in an amount of about 0.1 mM to about 0.5mM. In some embodiments, the transition metal catalyst is added in an amount of about 0.1 mM to about 0.4mM. In some embodiments, the transition metal catalyst is added in an amount of about 0.2mM to about O.SmM. In some embodiments, the transition metal catalyst is added in an amount of about 0.3mM to about 0.4mM. In some embodiments, the transition metal catalyst is added in an amount of about 0.33mM. In some embodiments, the transition metal catalyst is added in an amount of about O.OlmM to about 0.08mM. In some embodiments, the transition metal catalyst is added in an amount of about O.OlmM to about 0.07mM. In some embodiments, the transition metal catalyst is added in an amount of about O.OlmM to about 0.06mM. In some embodiments, the transition metal catalyst is added in an amount of about O.OlmM to about 0.05mM. In some embodiments, the transition metal catalyst is added in an amount of about 0.02mM to about O.OSmM. In some embodiments, the transition metal catalyst is added in an amount of about 0.05mM to about 0.09mM. In some embodiments, the transition metal catalyst is added in an amount of about 0.04mM.

Suitable ligands include appropriately activating ligands for ARGET ATRP, particularly tetradentate ligands such as tris(2-pyridylmethyl)amine (TPMA) and tris[2^dimethylamino)ethyl]amine (MeeTREN). In some embodiments, excess ligand is used. Excess ligand means the quantity of ligand present in the polymerisation medium exceeds the amount of ligand required to complex with the transition metal. The amount of excess ligand may be as much as ten times the required amount of ligand. In some embodiments, the molar ratio of ligand to transition metal is about 3 : 1 to about 4: 1. In some specific embodiments, the molar ratio of ligand to transition metal is about 3.8:1. The transition metal in the stable higher oxidation state is reduced to the activator state in situ by the reducing agent to initiate the polymerisation process. As such, the reducing agent may be any reducing agent capable of reducing the transition metal catalyst from the higher oxidation state to the lower oxidation state. The reducing agent may be selected for a particular polymerisation process such that at the polymerisation temperature the reducing agent reduces a sufficient quantity of transition metal catalyst in the higher oxidation state to transition metal catalyst in the lower oxidation state to substantially maintain the polymerisation rate. For example, at the polymerisation temperature the reducing agent reduces the additional amount of transition metal catalyst in the higher oxidation state to substantially maintain the ratio of transition metal catalyst in the higher oxidation state to transition metal catalyst in the lower oxidation state. In some embodiments, the reducing agent is selected from one or more of the group consisting of:

ascorbic acid, S0 ₂ , sulfites, bisulfites, thiosulfites, mercaptans, hydroxylamines, hydrazine (N ₂ H ₄ ), phenylhydrazine (PhNHNH ₂ ), hydrazones, hydroquinone, tin(II) 2-ethylhexanoate (Sn(EH) ₂ ), tintriflates, food preservatives, flavonoids, beta carotene, vitamin A, α-tocopherols, vitamin E, propyl gallate, octyl gallate, BHA, BHT, propionic acids, sorbates, reducing sugars, sugars comprising an aldehyde group, glucose, lactose, fructose, dextrose, potassium tartrate, nitrites, nitrites, dextrin, aldehydes, glycine, and transition metal salts. Ascorbic acid may be particularly suitable.

The amount of reducing agent will be determined by the total concentration of any oxidants in the polymerisation medium, the amount of termination reactions in the polymerisation, and the desired rate of the redox reaction. Excess reducing agent may be added to remove low concentrations of oxygen from the system.

As described previously, the anti-fouling coating that is formed using the processes described herein may form a continuous layer or a discontinuous layer. The density of the grafted polymers formed on the surface will be determined, at least in part, by: the density of the ATRP initiator functionality on the surface; the efficiency of the initiation reaction; etc. In some embodiments, the density of the zwitterionic polymers on the surface of the membrane is about 0.5 x 10 ^"3 mg/cm ² to about 2 x 10 ^"3 mg cm ² . In some embodiments, the density of the zwitterionic polymers on the surface of the membrane is about 0.5 x 10 ^"3 mg/cm ² to about 1.5 x 10 ^"3 mg/cm ² . In some embodiments, the density of the zwitterionic polymers on the surface of the membrane is about 0.5 x 10 ³ mg/cm ² to about 1.5 x 10 ^"3 mg/cm ² . In some embodiments, the density of the zwitterionic polymers on the surface of the membrane is 0.94 x 10 ^"3 mg/cm ² .

Advantageously, the anti-fouling coating that is formed using the processes described herein forms with a substantially uniform coverage over the surface of the membrane. In other words, the density of zwitterionic polymer groups is relatively uniform across the coated surface of the membrane. This is in contrast to prior art membranes that utilise sulfobetaine polymer blends which tend to form localised regions having relatively high density ("pockets") of sulfobetaine polymer. These regions ultimately create sulfobetaine channels through the membrane which reduces the efficiency of the reverse osmosis separation.

The anti-fouling coating can be characterised using a number of methods. Thermogravimetric analysis (TGA), ATR-FTBR. and water contact angle (WCA) measurements may be used to compare the amount of polymer deposited, the surface morphology and the hydrophilicity of the surfaces, respectively.

The biofouling resistance of reverse osmosis membranes produced using the processes described herein can be measured using a number of methods, including measuring the flux and/or salt rejection. There are also other more specific in situ methods, such as measuring the change in mass of the membrane, or analysing the effect of the fouling layer on acoustic or electromagnetic waves. One particularly valuable way of measuring biofouling performance is the direct measurement of microbial growth on the membrane surface. This method provides an unequivocal assessment of the attachment of biofouling to the surface.

To make direct measurements of biofouling on membranes it is best to mimic as closely as possible the environment that the membrane will be under when used in filtration. This can be achieved to some extent with stirred cell, or dead end filtration apparatus in which the direction of pressure and flow are perpendicular to the membrane surface. Such apparatus is more suitable for testing ultrafiltration membranes in which testing for the adhesion of biopolymers and other colloidal particles are more relevant. More closely matching the hydrodynamics for reverse osmosis filtration used in industry is the cross-flow apparatus in which the bulk flow of the water is across the surface of the membrane, while the pressure is perpendicular to the membrane. This cross-flow of water is used to prevent concentration polarisation at the membrane surface and provide shear forces across the surface of the membrane to remove fouling. Assessments of biofouling of reverse osmosis membranes using static bacterial attachment assays have been used in the literature. Using this method it is possible to closely mimic the chemistry and biology of seawater, the disadvantage being that the hydrodynamic conditions are not taken into account. In the experimentation reported herein, initial evaluation of performance of reverse osmosis membranes in reducing biofouling has been conducted using scanning electron microscopy (SEM) of the surface of membrane coupons. The microbial abundance of sulfobetaine-coated membranes was compared to control membranes after immersion in a seawater aquarium capable of supporting living marine organisms. The biofouling resistance of modified membranes (i) immersed in seawater aquarium tanks and (ii) exposed to hydrodynamic cross-flow testing was also measured. For example, based on aquaria experiments, we found that the reverse osmosis membranes that are modified using the processes described herein show a greater resistance to seawater microbial biofouling compared to unmodified membranes.

Presented herein is a commercially realistic approach for producing an anti-fouling coating covalently bound to polyamide layer of TFC membranes or cellulose acetate membranes via ARGET-ATRP polymerisation that provides for a significant reduction of biofilm formation on RO membranes.

The present invention is hereinafter further described by way of the following, non-limiting example(s) and accompanying figure(s).

EXAMPLE(S)

Chemicals Reagents were all purchased from Sigma- Aldrich and used as received. Solvents were purchased from Merck and used as received except dichloromethane and ethanol, which were purchased from Ajax Chemical Company. The ethanol was used as received and the dichloromethane was purified by distillation. All water used in the experimental section was deionised unless otherwise stated. Commercial reverse osmosis membranes used in this experimentation were FilmTec SW30HR, GE Sepa AD, Hydranautics SWC6 and CA.

Synthesis of Tris[(2-pyridyl)methyl]amine (TPMA)

2-Picolyl chloride hydrochloride (9.75 g, 59.5 mmol) was added to water (25 ml) and stirred under nitrogen at 0°C. Aqueous sodium hydroxide (11.25 ml, 5.3 M) was added, producing a bright red emulsion, to which 2-(aminomethyl)pyridine (3.2 g, 29.75 mmol) was subsequently added and the reaction mixture allowed to warm to room temperature. Further aqueous sodium hydroxide (11.25 ml, 5.3 M) was added in small aliquots, maintaining pH below 9.5. After 2 days the mixture was washed with aqueous sodium hydroxide (15% w/v, 25 ml) and the organic phase separated and then dried on magnesium sulfate before filtration and evaporation of the solvent. The product was extracted 3 times with boiling diethyl ether (100 ml) before reduction and purification of the crude yellow product by recrystallization from diethyl ether revealing a white powder (4.127 g, 14.2 mmol) in 48% yield.

Characterisation Methods X-ray Photoelectron Spectroscopy (XPS)

The instrument used for the analysis was a Kratos Axis Nova equipped with a 165 mm hemispherical analyser and delay-line detector. The samples were irradiated with X-rays from a monochromated Al Ka source (hv = 1486.6 eV) operating at 150 W. Survey spectra were acquired over the range 0—1400 eV at 1 eV/step and a pass energy of 160 eV. High-resolution spectra of selected photoelectron peaks were acquired at 0.1 eV/step and a pass energy of 20 eV. A charge neutraliser was used during spectral acquisitions. The analysis area was approximately 300 μπι x 700 μπι. The atomic fractions were calculated from the areas under the principal photoelectron peaks for each element detected, weighted by the appropriate relative sensitivity factors.

Thermogravimetric Analysis (TGA) Membranes were analysed under continuous heating conditions between room temperature (25 °C) and 700 ^* . The heating rate was kept constant at 10 °C min ^"1 with a nitrogen gas flow rate of 50 ml min ^"1 .The instrument used for this work was a TA Instrument Hi-Res Modulated TGA 2950

Fourier Transform Infrared (FTIR)

In order to study the chemical structure of biofilms developed on membrane surface Fourier Transform Infrared (FTIR) spectroscopy was used. A Thermo-Nicolet Nexus 870 FT-IR spectrometer (Thermo

Electron Corporation) fitted with the diamond Attenuated Total Reflectance (ATR) attachment, was used to generate FT-IR spectra and data was manipulated using OMNIC° software. The resolution was 4cm ^"1 and 128 scans and a mirror velocity of 0.6329 cm min ¹ were used for all tests.

Cell Fixative Solution The cell fixative solution was prepared by dissolving paraformaldehyde (4 g, 133 mmol) in Phosphate Buffered Saline (PBS) (60 ml) at 60°C. Sucrose (4 g, 11.7 mmol) was then dissolved and the solution allowed to cool to room temperature before the addition of glutaraldehyde solution (25% soln., 2ml) and then the final volume adjusted to 100 ml using PBS.

Aquarium Testing Reverse osmosis membranes coated with sulfobetaine polymer were analysed for biofouling prevention using aquaria-based static tests in which the 5cm X 5cm coupons were immersed in an aquarium (395 x 295 x 300 mm) fitted with a saltwater recirculation system attached to a bio-filter and protein skimmer. The constant flow rate was provided by a Techniflo pump (150W) with a point flow rate of 360 L/hr and mean temperature, salinity and pH of 16 °C, 36,000 ppm and 8.2 respectively. Controls of non-coated membranes were included in the aquarium for all tests. Control membranes were, as near as was practical, subjected to the same environment (eg solvent and temperature, but not reagents) as membranes being coated during functionalisation and polymerisation.

After 4 weeks the membranes were removed from the aquarium and the samples prepared for SEM to image the biofouling. Firstly the coupons were cut into samples of ~ 6-7 mm square and were rinsed in Phosphate Buffered Saline (PBS) before soaking in the cell fixative solution for 24 hours. After fixing, the samples were rinsed in PBS before dehydration by immersion for 15 minutes each in a succession of ethanol / water solutions (ethanol concentrations were 50% v/v, 70% v/v, 85% v/v and 95% v/v), followed by immersion for 15 minutes in absolute ethanol. The samples were then dried overnight in the ume hood, before preparation for SEM imaging.

Sputter Coater

Sputter coating of samples for scanning electron microscopy was conducted using a Quorum Emitech K575X sputter coater with a film thickness monitor to deposit a 5 nanometre layer of platinum onto the membrane surface. Scanning Electron Microscope (SEM)

Imaging of prepared and coated samples was conducted using an FEI Phenom SEM. Samples were prepared as described in the experimental section and then attached to an SEM stub using double-sided carbon tape. Then samples were sputter coated with platinum as described above. For analysis of biofouling three samples were removed from each coupon and for each sample three images were taken at different locations for each of three regions on the sample. This produced 9 images for each sample and 27 images at ΙΟ,ΟΟΟχ magnification for each coupon that were essentially randomly chosen (an effort was made to avoid membrane defects and large foreign objects, but to otherwise achieve large separation). Microbial abundance was determined for each of these 27 images for each coupon.

Preparation of ATRP initiator fnnctionalised filtration membranes For attachment to polymeric surfaces we chose bromoisobutyryl bromide (BibBr) as it provides both the tertiary bromide ATRP initiator group as well as an acid bromide group to react and bind with hydroxy and amino groups on the membrane surface. The surface of the polyamide layer contains secondary amino groups from the amide bonds as well as amino and carboxylic acid chain ends. The surface of the polyhydroxy layer or the membrane having hydroxy functionality contains predominately hydroxy groups.

Example 1 - Dry membrane A dry reverse osmosis membrane coupon (5 cm X 5cm) was immersed in hexane (25 ml). Next, the reaction vessel was purged with nitrogen before the addition of 2-bromoisobutyryl bromide (0.19 ml, 1.5 mmol). This mixture was stirred vigorously for 3 hours at room temperature. At the end of the reaction time the reaction solution was poured off and the membrane was washed with hexane, twice with methanol and then 3 times with methanol / water (1 : 1 v/v). The membrane was then stored in methanol / water (1 :1 v/v) ready for the polymerisation reaction.

Example 2 - Dry membrane and triethylamine

A dry reverse osmosis membrane coupon (5 cm X 5cm) was immersed in hexane (25 ml). Next, the reaction vessel was purged with nitrogen before the addition of triethyamine (0.23 ml, 1.65 mmol) and 2- bromoisobutyryl bromide (0.19 ml, 1.5 mmol). This mixture was stirred vigorously for 3 hours at room temperature and produced a white suspension. At the end of the reaction time the reaction solution was poured off and the membrane was washed with hexane, twice with methanol and then 3 times with methanol / water (1 :1 v/v). The membrane was then stored in methanol / water (1:1 v/v) ready for the polymerisation reaction.

Example 3 - Dry membrane and pyridine A dry reverse osmosis membrane coupon (5 cm X 5cm) was immersed in hexane (25 ml). Next, the reaction vessel was purged with nitrogen before the addition of pyridine (0.13 ml, 1.65 mmol) and 2- bromoisobutyryl bromide (0.19 ml, 1.5 mmol). This mixture was stirred vigorously for 3 hours at room temperature and produced a white suspension. At the end of the reaction time the reaction solution was poured off and the membrane was washed with hexane, twice with methanol and then 3 times with methanol / water (1 :1 v/v). The membrane was then stored in methanol / water (1 : 1 v/v) ready for the polymerisation reaction.

Example 4 - Dry membrane and DMAP

A dry reverse osmosis membrane coupon (5 cm X 5cm) was immersed in a solution of 4- (dimethylamino)pyridine (0.012g, 0.1 mmol) in hexane (25 ml). Next, the reaction vessel was purged with nitrogen before the addition of 2-bromoisobutyryl bromide (0.19 ml, 1.5 mmol). This mixture was stirred vigorously for 3 hours at room temperature and produced a white suspension that faded over time. At the end of the reaction time the reaction solution was poured off and the membrane was washed with hexane, twice with methanol and then 3 times with methanol / water (1:1 v/v). The membrane was then stored in methanol / water (1 :1 v/v) ready for the polymerisation reaction.

Example 5 - Dry membrane, triethylamine and DMAP A dry reverse osmosis membrane coupon (5 cm X 5cm) was immersed in a solution of 4-

(dimethylamino)pyridine (0.012g, 0.1 mmol) in hexane (25 ml). Next, the reaction vessel was purged with nitrogen before the addition of triethyamine (0.23 ml, 1.65 mmol) and 2-bromoisobutyryl bromide (0.19 ml, 1.5 mmol). This mixture was stirred vigorously for 3 hours at room temperature and produced a white suspension. At the end of the reaction time the reaction solution was poured off and the membrane was washed with hexane, twice with methanol and then 3 times with methanol / water (1 : 1 v/v). The membrane was then stored in methanol / water (1 :1 v/v) ready for the polymerisation reaction.

Example 6 - Hydrated membrane

A dry reverse osmosis membrane coupon (5 cm X 5cm) was soaked in water over night before being pressed between two filter papers to dry the surface. The coupon was then soaked twice for 10 minutes in hexane and then was immersed in fresh hexane (25 ml). Next, the reaction vessel was purged with nitrogen before the addition of 2-bromoisobutyryl bromide (0.19 ml, 1.5 mmol). This mixture was stirred vigorously for 3 hours at room temperature. At the end of the reaction time the reaction solution was poured off and the membrane was washed with hexane, twice with methanol and then 3 times with methanol / water (1:1 v/v). The membrane was then stored in methanol / water (1 : 1 v/v) ready for the polymerisation reaction.

Example 7 - Hydrated membrane and triethylamine

A dry reverse osmosis membrane coupon (5 cm X 5cm) was soaked in water over night before being pressed between two filter papers to dry the surface. The coupon was then soaked twice for 10 minutes in hexane and then was immersed in fresh hexane (25 ml). Next, the reaction vessel was purged with nitrogen before the addition of triethyamine (0.23 ml, 1.65 mmol) and 2-bromoisobutyryl bromide (0.19 ml, 1.5 mmol). This mixture was stirred vigorously for 3 hours at room temperature and produced a white suspension. At the end of the reaction time the reaction solution was poured off and the membrane was washed with hexane, twice with methanol and then 3 times with methanol / water (1:1 v/v). The membrane was then stored in methanol / water (1:1 v/v) ready for the polymerisation reaction.

Example 8 - Hydrated membrane and pyridine A dry reverse osmosis membrane coupon (5 cm X 5cm) was soaked in water over night before being pressed between two filter papers to dry the surface. The coupon was then soaked twice for 10 minutes in hexane and then was immersed in fresh hexane (25 ml). Next, the reaction vessel was purged with nitrogen before the addition of pyridine (0.13 ml, 1.65 mmol) and 2-bromoisobutyryl bromide (0.19 ml, 1.5 mmol). This mixture was stirred vigorously for 3 hours at room temperature and produced a white suspension. At the end of the reaction time the reaction solution was poured off and the membrane was washed with hexane, twice with methanol and then 3 times with methanol / water (1:1 v/v). The membrane was then stored in methanol / water (1 : 1 v/v) ready for the polymerisation reaction.

Example 9 - Hydrated membrane and DMAP A dry reverse osmosis membrane coupon (5 cm X 5cm) was soaked in water over night before being pressed between two filter papers to dry the surface. The coupon was then soaked twice for 10 minutes in hexane and then was immersed in (0.012g, 0.1 mmol) in hexane (25 ml). Next, the reaction vessel was purged with nitrogen before the addition of 2-bromoisobutyryl bromide (0.19 ml, 1.5 mmol). This mixture was stirred vigorously for 3 hours at room temperature and produced a white suspension that faded over time. At the end of the reaction time the reaction solution was poured off and the membrane was washed with hexane, twice with methanol and then 3 times with methanol / water (1 :1 v/v). The membrane was then stored in methanol / water (1:1 v/v) ready for the polymerisation reaction.

Example 10 - Hydrated membrane, triethylamine and DMAP A reverse osmosis membrane coupon (5 cm X 5cm) was soaked in water over night before being pressed between two filter papers to dry the surface. The coupon was then soaked twice for 10 minutes in hexane and then immersed in a solution of 4-(dimethylamino)pyridine (0.012g, 0.1 mmol) in hexane (25 ml). Next, the reaction vessel was purged with nitrogen before the addition of triethyamine (0.23 ml, 1.65 mmol) and 2-bromoisobutyryl bromide (0.19 ml, 1.5 mmol). This mixture was stirred vigorously for 3 hours at room temperature and quickly produced a white suspension. At the end of the reaction time the reaction solution was poured off and the membrane was washed with hexane, twice with methanol and then 3 times with methanol / water (1:1 v/v). The membrane was then stored in methanol / water (1:1 v/v) ready for the polymerisation reaction.

The amount of tertiary bromide attached to the membrane surface after the functionalisation reaction was determined by X-ray Photoelectron Spectroscopy (XPS). Table 1 shows the relative abundances of oxygen (O), nitrogen (N), carbon (C) and Bromine (Br) (abundances of sulfur and silicon of≤0.3% have been excluded for clarity) for various functionalisation reactions. Firstly, it is clear that there is a significant level of bromine in all samples, indicative of successful functionalisation. Indeed, the maximum atomic fraction of bromine on the surface of 2.2% after 6 hours reaction time for FT6 compares favourably with the results found for the attachment of BibBr onto cellulose (maximum of 1.6% bromine after 24 hours). ¹² This confirms that the secondary amines of the polyamide chains retain sufficient reactivity to allow reasonable levels of initiator coverage. It also shows that modification to the membrane surface to produce a high population of hydroxy groups is not necessary in our case, and may in fact decrease our initiator coverage. Indeed, for this type of functionalisation reaction, conducted on a hydroxy modified nylon surface, the ratio of bromine and nitrogen concentrations was reported as 0.13, ¹³ while the current work achieved ratios ranging from 0.10-0.37 with an average of 0.21.

Table 1

Surface Abundance for Elements from XPS Data

Coupons FT1 , FT3 and FT6 were functionalised under conditions that were identical except for reaction time. As the reaction time increases so does the abundance of bromine, but with a decreasing rate as time goes on (see Figure 2). This trend is matched by samples FTDMAP0.5, FTDMAPl, and FTDMAP3 that also only vary within themselves by reaction time.

It is noted that all of the functionalisation reactions that contain 4-(dimetoylamino)pyridine (DMAP) are slower than comparable reactions without this reagent. The difference for the reaction on Fi nTec at 1 hour is relatively small and within the error of XPS determination of 5%. However, for both FilmTec and GE membranes at 3 hours there is significantly less bromine on the surface for the reactions containing DMAP. Obviously the DMAP is not working as a catalyst, and indeed is in some way hindering the reaction. One reason that its catalytic properties may not be shown is due to the fact that there was no organic base in these reactions and the DMAP is probably being neutralised by the hydrogen bromide (HBr) by-product.

Solvents found to cause delamination or damage the commercial thin film composite membranes that were tested include toluene, ethyl acetate, acetone, dimethylsulfoxide, chloroform, dichloromethane, tetrahydrofuran, diethyl ether and acetonitrile. Hexane was found to be a suitable solvent for the functionalisation reaction of the thin film composite membranes, and diethyl ether was found to be suitable for the functionalisation reaction with cellulose acetate membranes. Preparation of filtration membrane having a zwitterionic polymer coating

Example 11 - 2SBMA

[2-(Methacryloyloxy)emyl]dimemyl-(3-sulfopro^ hydroxide (2-SBMA; Formula (I)) (5 g,

18 mmol) was dissolved in methanol / water (1:1 v/v, 18 ml) containing copper (Π) chloride (0.33 mM) and TPMA (1.25 mM). A iunctionalised membrane was immersed in this solution and the reaction vessel was purged with nitrogen before the addition of a methanol / water (1:1 v/v) solution of L-ascorbic acid (41 mM, 1 ml). The reaction vessel was then sealed and stirred for 24 hours at room temperature. At the reactions conclusion the membrane coupon was washed thoroughly with water before storage in methanol / water (1:10 v/v).

(D

Example 12 - 3SBMA

[3-(Memacryloylammo)propyl]mmethyl(3-sulfopropyl)ammonium hydroxide inner salt (3-SBMA;

Formula (Π)) (5 g, 18 mmol) was dissolved in methanol / water (1:1 v/v, 18 ml) containing copper (Π) chloride (0.33 mM) and TPMA (1.25 mM). A functionalised membrane was immersed in this solution and the reaction vessel was purged with nitrogen before the addition of a methanol / water (1:1 v/v) solution of L-ascorbic acid (41 mM, 1 ml). The reaction vessel was then sealed and stirred for 24 hours at room temperature. At the reactions conclusion the membrane coupon was washed thoroughly with water before storage in methanol / water (1 :10 v/v).

(Π) The presence of polysulfobetaine on the membrane surface of the functionalised membranes is illustrated both by Thermogravimetric Analysis (TGA) and Fourier Transform Infrared - Attenuated Total Reflectance (FTIR-ATR).

Analysis of a polysulfobetaine coated membrane by TGA (see Figure 3) shows two significant peaks in the first derivative of the mass loss curve. The largest occurs at 520°C and is due to the other components of the membrane such as the polyamide and polysulfone layers and accounts for 60% of the mass. A smaller peak occurs at 306°C and this is due to the polysulfobetaine attached to the surface.

The presence of the polysulfobetaine is also confirmed by FTIR-ATR. The typical peaks of

polysulfobetaine (see Figure 4) were observed such as 1642 cm ^" , 1539 cm ',1487cm ^'1 , 1204 cm ^" , and 1038cm ^"1 corresponding to the C=0 stretching, N-H bending, quaternary ammonium, S=0 asymmetric stretching and S=0 symmetric stretching respectively.

For analysis of the biofouling prevention characteristics of polysulfobetaine coated reverse osmosis membranes three samples were removed from each membrane coupon, both for uncoated controls and . polysulfobetaine-coated coupons (see Figure 5 for examples). For each of these samples three images at 10,000 times magnification were taken at different locations for each of three regions on the sample, producing 9 images for each sample and 27 images for each coupon. After the images were acquired, microbial abundance for a coupon was determined by counting the number of microbes on each of the 27 images. For every sample mean and median were calculated from the total microbes counted for each image, then the mean of these measures of central tendency were taken for the three samples from the coupon, producing the average median and the overall mean for the coupon. Standard deviations were also taken for data within a sample and within the coupon. This was used to create a 95% confidence interval for the mean for each coupon. To make comparisons between coupons produced at different times relevant, both the average median and the overall mean for coated coupons are expressed as a percentage of the same values recorded for the related control coupons. These numbers will be described as the coupons relative median and relative mean values. The summarized results of this experimentation are shown in Table 2. This table shows the experimental conditions and microbial abundance results for all of the coupons hydrated before the functionalisation reaction. These coupons are separated into sets of experiments based on the associated control. The data shown in Table 2 indicates that polysulfobetaine- coated membranes produce, on average, microbial abundances of approximately one quarter of those for the related control coupon, which suggests that the polysulfobetaine coating provides a 4-fold improvement in biofouling prevention over currently available commercial polyamide reverse osmosis membranes. Table 2

Biofouling Efficiency Determination using Microbial Abundance

Hydrated membranes used in the functionaliztion reaction

The role of polysulfobetaine as a biofouling prevention agent is also supported by FTIR results. Figure 6 shows the FTIR spectra of RO membranes: polySBMA coated RO membrane (top); polySBMA coated RO membrane after aquaria-based static test (middle); and uncoated-control RO membrane after aquaria- based static test (bottom). The peaks in the spectra for the polySMBA coated RO membrane exposed to the aquarium are consistent with the peaks for the polySBMA coated RO membrane not exposed to the aquarium conditions, indicating that the polySBMA coating prevented biofilm formation on the membrane. In contrast, the uncoated-control RO membrane that was exposed to the aquarium shows different features in the spectrum, such as natural polysaccharides secreted by certain bacterial strain (at adsorption frequency of 1030cm ^"1 , assign to C-O stretching vibration) closely matching that of pure biofilm previously published. ¹ In this case the biofilm is thick enough to prevent penetration of the IR beam through to the sub-surface polyamide layer.

Example 13 - Testing Hydrophilicity - Water Contact Angle

Water contact angle (WCA) measurements were carried out in air with a water droplet (static sessile drop method), and analyzed using ImageJ software. Membranes were attached to a glass slide with double- sided tape and placed on a horizontal platform. A water droplet was placed on the membrane surface and an image captured by camera (see Figure 8). The internal angle of both sides of the water droplet was determined for 6 droplets per sample, and the mean value calculated.

Increased surface hydrophilicity has been associated with ideal polymeric coating materials with reduced bacterial affinity. Polysulfobetaine coated membranes visually exhibited greater wettability than the uncoated (pristine) and the ATRP initiator modified (bromoisobutyryl bromide) membranes. This shows an increase in membrane surface hydrophilicity after graft polymerization of 3SBMA.

Example 14 - Stirred cell experiments

In order to determine the thickness of the polysulfobetaine coating that gives optimum permeation and rejection properties, a stirred cell experiment was used as a screening test. The thickness of the polysulfobetaine coating was optimised by varying the number of initiation sites and/or varying the polymerisation time, followed by performance measurements, using a Sterlitech HP4750 (Figure 9; Cumberland Industrial Center Kent, WA 98032).

MilliQ water at a pressure of 400psi was used for the compaction of the membranes before the salt water tests (2000ppm NaCl) were started. Salt water flux was measured as a function of pressure (up to 400 psi) using a stirred cell, in which the direction of pressure and flow are both perpendicular to the membrane surface. The collection of membrane permeate was carried out in a beaker placed on the electronic balance. The balance was connected to a computer and weight measurements were collected every 5 min using Lab VIEW (National Instruments, USA) software program. The salt concentrations were measured using a conductivity meter (Extech Equipment, Australia).

Figures 10, 11 and 12 show the permeate flux performance of polysulfobetaine coated RO commercially available membranes, using salt water concentration of 2000ppm NaCl. The results of control-uncoated membranes are labelled unmodified in the figures. The results for membranes modified with

polysulfobetaine (3SBMA) are labelled commercial membrane grade-g-3SBMA. Salt rejection results are presented in Table 4.

Table 4

# AD - GE Sepa™ AD polyamide TFC membrane (GE Osmonics Labstore) SWCLF - Hydranautics SWC6 polyamide TFC membrane (Nitto Denko)

SW30-HR - Filmtec™ SW30HR polyamide TFC membrane with PVA coat (Dow Water)

In the cases of the AD and SWC6 membranes the flux of polysulfobetaine coated membranes is, in general, the same or slightly better than uncoated membranes. However, polysulfobetaine coated SW30HR membranes lose relatively little salt water flux when compared to the uncoated membranes. For all coated membranes the salt rejection is lower than respective uncoated-control membranes except for AD-g-3SBMA-i-3min-p-l 8h where salt rejection is better than its uncoated counterpart. It is important to note that the stirred cell is more suitable for testing ultrafiltration membranes and therefore the stirred cell is only used as a screening tool. Based on the stirred cell performance tests polymerisation coating conditions were determined, and scaled up membranes were produced for testing with the cross flow apparatus.

Scale up of filtration membranes having a zwitterionic polymer coating

Commercial reverse osmosis (RO) membranes were modified by reacting surface amine and/or hydroxy groups as described previously.

Example 15 - Synthesis ofATRP initiator modified polyamide membranes The membrane coupon was presoaked in water overnight. The membrane coupon was then blotted between absorbent papers to remove excess water, and then subsequently stirred in hexane for 20 minutes. A 240 mL glass jar containing the membrane coupon was charged with hexane (180 mL) and flushed with nitrogen for approximately 15 minutes. Triethylamine (TEA, 2.9 mL, 2 x 10 ^"2 mol) and 2- bromoisobutyryl bromide (BibBr, 1.2 mL, 9.8 x 10 ^"3 mol) were injected simultaneously, into the reaction jar. The reaction mixture was stirred for 10 minutes at room temperature. The membrane coupon was thoroughly washed using the following procedure: washed with copious amounts of deionized water; rinsed in hexane; washed with deionized water, and sonicated in deionized water for 3 minutes.

Example 16 - Synthesis of ATRP initiator modified RO membranes with surface hydroxy groups

Several commercially available RO membranes contain surface hydroxy groups rather than surface amine groups. For example, cellulose acetate (CA) membranes contain surface hydroxy groups. Additionally, we have found that some commercially available polyamide membranes contain surface hydroxy groups because some manufacturers also include a hydrophilic coating layer on top of the polyamide layer. For example, Filmtec SW30HR membranes have a hydrophilic coating layer, suggested to be polyvinyl alcohol (PVA), on top of the polyamide skin layer. ¹⁴ The surface chemistry needs to be taken into account when functionalising the surface because different reaction conditions are used to attach the ARGET ATRP initiator to membranes with surface hydroxy groups (e.g. SW30HR and Cellulose Acetate membranes) than surface amine groups.

A Filmtec SW30HR membrane coupon was presoaked in water overnight. The membrane coupon was then blotted between absorbent papers to remove excess water from the membrane, and then subsequently stirred in hexane for 20 minutes. A 240 mL glass jar containing the membrane coupon was charged with hexane (180 mL) and flushed with nitrogen for approximately 15 minutes. Triethylamine (TEA, 2.9 mL, 2 x 10 ^"2 mol) and 2-bromoisobutyryl bromide (BibBr, 1.2 mL, 9.8 x 10 ^"3 mol) were injected

simultaneously, into the reaction jar. The reaction mixture was stirred for 5 minutes at 0 °C, then allowed to warm to room temperature, and stirred for a further 3 hours. The membrane coupon was thoroughly washed using the following procedure: washed with copious amounts of deionized water; rinsed in hexane; washed with deionized water; and sonicated in deionized water for 3 minutes.

Example 17 - Growth of polysulfobetaine coating using surface-initiated ARGET ATRP

A 240 mL glass jar containing the ATRP initiator modified membrane coupon was charged with 3-SBMA monomer (10 g, 3.4 x 10 ² mol), CuCl ₂ (0.001 g, 6.84 x 10 ⁶ mol) and TPMA ligand (0.02 g, 6.84 x 10 ⁵ mol) in methanol/water (1:1 v/v; 155 mL). The concentration of copper catalyst was approximately 0.04 mM. The jar was sealed with a screw cap containing septa, and the solution was stirred under nitrogen for 20 minutes before a solution of ascorbic acid reducing agent (0.6 g, 3.4 x 10 ^"3 mol) in methanol/water (1:1 v/v, 5 mL) was injected. Polymerization was conducted at room temperature for 24 hours. After polymerization the membrane coupon was thoroughly washed with deionized water, sonicated in deionized water for 3 minutes, and then stirred in deionized water for >5 hours prior to hydrodynamic cross-flow testing. Example 18 - Cross-flow experiments

Cross-flow filtration data is preferred industrially because, although the pressure is normal to the membrane surface, the fluid flow is tangential to the membrane surface, mimicking conditions in desalination plants. For this reason, a series of cross-flow filtration tests were carried out. Membrane samples having a minimum size of 14 cm x 8 cm were used. The cross flow filtration unit used to test the performance of coated commercial RO membranes was the Sterlitech CF042 (Figure 13).

The device is equipped with six filtration cells, each with an effective filtration area of 9.2cm x 4.6cm (42cm ² ). The membrane performance was tested using seawater from West Beach, South Australia; 36000 ppm NaCl, pH of 8.2 at the cross flow rate of 2L/min and pressure of 2758kPa (400psi). The constant flow rate and pressure were provided by a G-l 3 Hydra-Cell, pump. MilliQ water was used for the compaction of the membranes before the sea water performance tests were started. The collection of membrane permeate was carried out in a beaker placed on an electronic balance. All balances were connected to a computer and weight measurements were collected every 5 min using a Lab VIEW (National Instruments, USA) software program. The salt concentrations were measured using a conductivity meter (Extech Equipment, Australia). Three membranes from the same membrane manufacturer were placed in series with a control uncoated membrane in the middle and two coated membranes on the side. The three membranes for other manufacturers were placed in parallel. After the collection and calculations of flux and salt rejection, sea water was recirculated through the cells for 18 days and the degree of biofouling was assessed using SEM and compared to membranes tested under static conditions.

After the fouling experiment, the cross flow system was cleaned using 1500ppm of sodium hypochlorite (-5% available chlorine) in distilled water for two hours, followed by distilled water flushing for three hours and MilliQ water flushing overnight.

Figures 14, 15 and 16 show the performance of RO coated commercially available membranes. The results of control uncoated membranes are labelled unmodified in figures. The results of membranes modified with sulfobetaine, 3SBMA are labelled commercial membrane grade-g-3SBMA. Salt rejection results are presented in Table 5. Table 5

# AD - GE Sepa™ AD polyamide TFC membrane (GE Osmonics Labstore) SWCLF - Hydranautics SWC6 polyamide TFC membrane (Nitto Denko)

SW30-HR - Filmtec™ SW30HR polyamide TFC membrane with PVA coat (Dow Water)

As can be seen, all of the sulfobetaine modified membranes tested exhibited higher or equivalent flux as the unmodified membranes, thereby indicating that modification of the surface does not detrimentally affect the flux.

Furthermore, the SWCLF-g-3SBMA membrane showed a better salt rejection than the unmodified membrane.

Example 19 - TGA experiment

Scaled up membranes were modified by polysulfobetaine as described and labelled (commercial membrane grade-L). The amount of the polysulfobetaine attached to the surface of the commercially available membranes was detected in the weight loss step between 200°C and 400°C with the peak temperature of ~ 300°C and presented in Table 6.

From the table it is evident that a similar amount of polysulfobetaine remains on the surface of the membranes after 18 days of exposure to the hydrodynamic conditions in the cross flow experiment (labelled commercial membranes grade-L after 18days) compared to their un-exposed counterparts.

In addition, the amount of the bio-film attached to the membrane surface was detected in the step of weight loss between 100°C -200°C with the peak temperature ~173°C. Table 6

For analysis of the biofouling prevention characteristics of 'scaled up' polysulfobetaine coated reverse osmosis membranes, subjected to hydrodynamic conditions, three samples were removed from each membrane coupon, both for uncoated controls and polysulfobetaine-coated coupons. For each of these samples three images at 10,000 times magnification were taken at different locations for each of three regions on the sample, producing 9 images for each sample and 27 images for each coupon. After the SEM images were acquired, microbial abundance for a coupon was determined following the same procedures as outlined for aquaria-based static tests. Microbial fouling for coupons subjected to hydrodynamic conditions were consistent with findings for coupons subjected to aquaria-based static tests, which suggests that the polysulfobetaine coating provides a 4-fold improvement in biofouling prevention over currently available commercial polyamide reverse osmosis membranes.

In summary, the fabrication of polysulfobetaine coated RO desalination membranes resulted in a minimum of at least 4 times biofouling improvement over commercially available membranes, while still maintaining competitive permeation flux and rejection properties.

It will be appreciated by persons skilled in the art that numerous variations and/or modifications may be made to the invention as shown in the specific embodiments without departing from the spirit or scope of the invention as broadly described. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive.

REFERENCES

1. Cadotte, J. E.; Petersen, R. J.; Larson, R. E.; Erickson, E. E. A new thin-film composite seawater reverse osmosis membrane. Desalination 1980, 32, 25-31.

2. Ostuni, E.; Chapman, R. G.; Holmlin, R. E.; Takayama, S.; Whitesides, G. M., A Survey of Structure - Property Relationships of Surfaces that Resist the Adsorption of Protein. Langmuir 2001 , 17 (18), 5605-5620.

3. Shen, M.; Horbett, T. A., The effects of surface chemistry and adsorbed proteins on

monocyte/macrophage adhesion to chemically modified polystyrene surfaces. Journal of Biomedical Materials Research 2001, 57 (3), 336-345. 4. Herrwerth, S.; Eck, W.; Reinhardt, S.; Grunze, M., Factors that Determine the Protein Resistance of Oligoether Self- Assembled Monolayers -Internal Hydrophilicity, Terminal Hydrophilicity, and Lateral Packing Density. Journal of the American Chemical Society 2003, 125 (31), 9359-9366.

5. Lewis, A. L.; Hughes, P. D.; Kirkwood, L. C; Leppard, S. W.; Redman, R. P.; Tolhurst, L. A.; Stratford, P. W., Synthesis and characterisation of phosphorylcholine-based polymers useful for coating blood filtration devices. Biomaterials 2000, 21 (18), 1847-1859.

6. Murphy, E. F.; Lu, J. R.; Lewis, A. L.; Brewer, J.; Russell, J.; Stratford, P., Characterization of Protein Adsorption at the Phosphorylcholine Incorporated Polymer - Water Interface. Macromolecules 2000, 33 (12), 4545-4554.

7. Iwasaki, Y.; Ishihara, K., Phosphorylcholme-containing polymers for biomedical applications. Analytical and Bioanalytical Chemistry 200S, 381 (3), 534-546.

8. Wang, T.; Wang, Y.-Q.; Su, Y.-L.; Jiang, Z.-Y., Antifouling ultrafiltration membrane composed of polyethersulfone and sulfobetaine copolymer. Journal of Membrane Science 2006, 280 (1-2), 343-350.

9. Sun, Q.; Su, Y.; Ma, X.; Wang, Y.; Jiang, Z., Improved antifouling property of zwitterionic ultrafiltration membrane composed of acrylonitrile and sulfobetaine copolymer. Journal of Membrane Science 2006, 285 (1-2), 299-305.

10. Chiang, Y.-C.; Chang, Y.; Higuchi, A.; Chen, W.-Y.; Ruaan, R.-C., Sulfobetaine-grafted poly(vinylidene fluoride) ultrafiltration membranes exhibit excellent antifouling property. Journal of Membrane Science 2009, 339 (1 -2), 151-159.

11. Matyjaszewski, et al. Journal of American Chemical Society 1995, // 7, 5614. 12. Carlmark, A.; Malmstrom, E. E., ATRP Grafting from Cellulose Fibers to Create Block- Copolymer Grafts. Biomacromolecules 2003, 4 (6), 1740-1745.

13. Xu, F. J.; Zhao, J. P.; Kang, E. T.; Neon, K. G.; Li, J., Functionalization of Nylon Membranes via Surface-Initiated Atom-Transfer Radical Polymerisation. Langmuir 2007, 23 (16), 8585-8592. 14. Tang, C. Y., won, Y-N and Leckie, J. O., Effect of membrane chemistry and coating layer on physiochemical properties of thin film composite polyamide RO and NF membranes: I. FTIR and XPS characterization of polyamide and coating layer chemistry. Desalination, 2009, 242, 149-167.

Throughout this specification the word "comprise", or variations such as "comprises" or "comprising", will be understood to imply the inclusion of a stated element, integer or step, or group of elements, integers or steps, but not the exclusion of any other element, integer or step, or group of elements, integers or steps.

All publications mentioned in this specification are herein incorporated by reference. Any discussion of documents, acts, materials, devices, articles or the like which has been included in the present specification is solely for the purpose of providing a context for the present invention. It is not to be taken as an admission that any or all of these matters form part of the prior art base or were common general knowledge in the field relevant to the present invention as it existed in Australia or elsewhere before the priority date of each claim of this application.