Water Treatment by Molecularly Imprinted Materials

Final Report Summary - WATERMIM (Water Treatment by Molecularly Imprinted Materials)

Executive Summary:
The objective of the WATERMIM project (http://lpre.cperi.certh.gr/watermim/) was to bring the imprinting technology one step closer to the development of smart materials, thus, enabling the selective separation and sensing of harmful compounds from water sources at very low concentrations (i.e. below 0.1ppb). In the first phase of the project, the work focused on the computational modeling and combinatorial screening for the optimization of molecularly imprinted polymers (MIPs) composition towards specific target molecules. Taking into consideration the results from the computational simulations and the combinatorial screening, MIP nanoparticles (NPs) were synthesized using various polymerization techniques in order to produce NPs that have well-defined physical and chemical properties and exhibit high selectivity and specificity towards the selected template molecules, in large scale and in a cost-effective manner.
After a critical assessment, the work focused on the most promising polymerization techniques that, based on previous work, resulted to materials that exhibited totally controllable physical properties and thus, could successfully be incorporated into the composite membranes. More specifically, precipitation, miniemulsion, inverse miniemulsion polymerization were employed for the synthesis of MIPs using pesticides, pharmaceutically active compounds and endocrine disruptors (e.g. atrazine, clopyralid, 2,4-dichlorophenoxyacetic acid (2,4-D), propranolol, penicillin G, pentoxifyllin beta-estradiol and bisphenol A), using either bifunctional or trifunctional cross-linkers (CERTH, ULUND, USTUTT, MIPTech). Also, molecularly imprinted nanoparticles with novel structure, which were synthesized using exposed clickable chemical groups on particle surface (ULUND), were applied for developing composite nanoparticle-based membranes. Additionally, novel MIP nanoparticles have been developed for various triazines by iniferter grafting technique using modified dendrimers as core (CU). Moreover, the work has focused on the development of both iniferter and RAFT based grafting techniques on mesoporous silica for final transfer to nanoparticulate format (UDO).
Composite membranes were produced by the incorporation of MIP nanoparticles into (USTUTT) or between polymeric substrates (CERTH, USTUTT) and fibers (ULUND) as well as the grafting of the polymer onto organic (CU) or inorganic substrates (CU, CERTH, LNU, Keranor) for the development of composite membranes to be used in separation (CERTH, USTUTT, CU) and/or sensing applications (CU, ULUND, LNU) depending on the substrate chosen. The work focused on the application of the synthesized composite membranes in continuous separation processes and the realization of a series of benchmark tests for the evaluation of MIM performance. Additionally, combined systems for the separation and catalytic decomposition of the pollutants were studied leading to the conclusion that the optimum process was first to concentrate the pollutants into the sample and then to proceed with the catalytic decomposition. Various tests were performed (vTI) in order to study the decomposition rate of the selected pollutants through various experimental paths. For several pollutants, i.e. clopyralid, atrazine and bisphenol A, protocols for a reductive and oxidative decomposition were elaborated. For all tested pollutants a fast and complete decomposition could be achieved at least by one of the two ways of treatment.
Within the project lifetime huge steps have been made in order to bring the imprinting technology one step closer to the water treatment applications. Due to environmental reasons, these novel materials, are needed for niche applications in water treatment industry, such as sensing and monitoring systems. Apart from water treatment industry, pharmaceutical and biomedical industry could also be supported by innovative diagnostic, analytical and separation tools based on imprinting technology. These tools could further form the basis for new standards which can support commercialization, since new water-related environmental legislation in Europe clearly defines the need for even stricter regulations. To this direction, WATERMIM has set the basis for the establishment of a new range of techniques and methods for the monitoring of the regulated compounds at very low levels.

Project Context and Objectives:
In recent years, the excessive exploitation of natural resources and widespread use of chemicals has led to the accumulation of various pollutants in our environment. This has received much attention due to the adverse effects of pollutants on human health. In particular, several contaminants (i.e. disinfection by-products, pathogens and synthetic or natural organic compounds) largely affect present and future treatment technologies utilized by the drinking-water industry. Some of the existing, pending or potential emerging contaminants are synthetic organic compounds (SOCs), such as pesticides, pharmaceutical active compounds (PhACs) and endocrine disrupting compounds (EDRs). More specifically, the prolonged use of pesticides induces the risk of their retention in crops and soils from which, in turn, due to washing and leaching processes, they pass to surface and ground waters. On the other hand, a series of medicines employed for the treatment of various diseases (i.e. cancer, hormone imbalance, osteoporosis, etc.) end up in the aquatic environment via domestic and industrial wastewater discharges.
For the removal of these substances, several methods, such as ozonation, UV radiation, membrane filtration, and activated carbon adsorption have been developed. However, these methods are normally non-selective and result in a secondary pollution. Catalytic oxidation/reduction of organic pollutants by metal oxides (i.e. titanium oxide) or other active metals (i.e. Pd-Cu) is the newest water treatment technique. However, these catalytically active materials usually do not target specific pollutants but also attack other water compounds, which leads to unwanted side reactions and additional costs for reductants/oxidants. The addition of substrate specificity to the catalysts would help to reduce unwanted side reaction and costs.
It is evident that novel materials with high selectivities for the eclectic removal of toxic chemicals and other contaminants are urgently required. More specifically, innovative methods for the synthesis of advanced functional materials and the development of novel tailor-made membranes/filters with high selectivity and long-term performance stability are demanded. One of the most promising class of new and highly selective functional materials is the molecularly imprinted polymers (MIPs).
Essentially, molecular imprinting is a process where a functional monomer and a cross-linker are co-polymerized, via a free-radical polymerization mechanism, in the presence of a target molecule (i.e. the imprint molecule) that acts as a molecular template. The functional monomer molecules initially form a complex with the imprint molecules. This is followed by the polymerization of the functional monomer with a cross-linker. The resulting highly cross-linked polymeric structure keeps in position the template molecules through covalent or non-covalent molecular interactions between the monomer and the template functional groups. Subsequent removal of the imprint molecules reveals the specific binding sites that are complementary in size and shape to the template molecule. MIPs can be used as selective sorbents for a wide range of biological, pharmaceutical and environmental applications. These developments underline the potential use of MIPs as highly-selective adsorbing materials in water treatment, detection and monitoring of pollutants at very low concentrations (i.e. below 0.1ppb). Molecularly imprinted materials can be also combined with catalysts resulting in novel composite adsorbent/catalyst systems.
The WATERMIM project was focused on the advancement and optimization of the MIP technology in order to produce functional materials with well-defined morphologies with respect to pore structure and selectivity for water treatment applications. The objective was to bring imprinting technology one step closer to the development of smart materials, thus, enabling the selective separation and sensing of harmful compounds from water sources.
To this direction, the target was the elimination of the random distribution and the uneven accessibility of receptor sites in the volume of the imprinted material that is crucial for its performance. Such novel materials will immediately gain practical relevance, especially, due to their increased selectivity and superior stability under long and harsh technical conditions. The simultaneous optimization of the imprinting efficiency, polymer membrane morphology and separation conditions will enable the development of a truly molecular selective water purification process, based on affinity interactions that would have a large application impact on the water treatment industry. All types of SOCs (i.e. triazines, pharmaceutical compounds and endocrine disruptors) were considered target compounds in the WATERMIM project.
More specifically, the WATERMIM project aimed at the development of a) highly selective MIMs with high membrane permeability for water treatment applications and removal of SOCs at very low concentrations (i.e. below 0.1ppb) b) novel sensing devices for monitoring the concentrations of unwanted pollutants and c) novel MIP/catalyst system for the combined separation/degradation of harmful contaminants. More specifically, the workplan has been divided into four R&D activities, dealing with the design of MIPs, the production of molecularly imprinted composite membranes and the assessment of membranes performance in terms of selectivity, sensing, stability and regeneration ability. The ultimate goal was the development of molecularly imprinted materials that can provide innovative, reliable and cost-efficient tools for water treatment applications that can afford a truly selective separation approach, at molecular level, based on affinity interactions.

Project Results:
DESCRIPTION OF THE MAIN S&T RESULTS

OVERALL STRATEGY AND GENERAL DESCRIPTION
The WATERMIM project aimed at the development of highly selective Molecularly Imprinted Materials (MIMs) with high membrane permeability for the selective separation and sensing of Synthetic Organic Compounds (SOCs) from water at very low concentrations (i.e. below 0.1ppb). To this direction, novel composite membranes and sensing devices have been developed for monitoring the concentrations of unwanted pollutants. Also, new MIP/catalyst systems have been experimentally studied for the combined separation/degradation of harmful contaminants.
More specifically, the workplan has been divided into four R&D activities, dealing with the design of MIPs (WP1), production of molecularly imprinted composite membranes (WP 2 & 3) and assessment of membranes performance in terms of selectivity, sensing, stability and regeneration ability (WP4). Each workpackage is further divided into a number of specific R&D activities/Tasks. The main S & T results of the WATERMIM Project are presented in this report for each Task.

WP 1: TEMPLATE SELECTION AND MIP OPTIMIZATION
Objectives: Optimization of molecularly imprinted polymer composition by computational design techniques and combinatorial screening.
Action Done / Main Results

Task 1.1: Template selection, design and synthesis of functional monomers
The first objective of the WATERMIM project was the identification of the target template molecules. To this direction the consortium decided to realize, firstly, a short survey on the regulations regarding the emerging toxic organic compounds that need to be removed from aqueous solutions. Afterwards, a list of candidate template molecules, including pesticides, pharmaceutical active compounds and endocrine disruptors, has been selected depending on their toxicity and their ability to be used in MIP synthesis.
For the selection of the template molecules it was important to keep the "balance" between having a range of targets and specific/concrete goals. Actually, a critical issue was to use templates that can cover a class of target compounds. Additionally, these templates should not only be imprintable, but also should contribute to solving problems in the real world. Traditionally utilized templates in the imprinting technology should also be included in the list, keeping in mind that the system should be focused for water treatment, in order to compare the MIP membranes with other purification technologies.
According to the above statement, beta-Estradiol has been chosen because of the potential to be used as "selective class" to the most of the steroid hormones (i.e. androgens, estrogens, contraceptives and glucocorticoid). Another interesting template for the EDCs is Bisphenol A, since this compound is widely used as additive in the polymer or/and paper industry.
Regarding the pharmaceutically active compounds beta-lactam antibiotics (i.e. Penicillin G), were considered as an interesting class of compounds, since they are the most widely-used group of antibiotics. Another class of compounds is the xanthine derivates such as Caffeine, Theophyllin, Theobromine, Pentoxifyllin, etc. Pentoxifyllin, which can be considered as a prominent water pollutant, and, thus very interesting to be used as a model template in MIP synthesis for the binding of the rest of the xanthane derivatives. Additionally, from beta blockers, that belong to an interesting class of drugs used for various indications, propranolol was selected that is widelly used for the management of hypertension.
From pesticides, atrazine is a widely used herbicide. Its use is controversial due to its effects on nontarget species, such as on amphibians, and because of widespread contamination of waterways and drinking water supplies. There are also thoughts to have implications for human birth defects, low birth weights and menstrual problems. Even if its use is banned in the European Union it is still detected and it is one of the most widely used herbicides in the United States. Moreover, 2,4-Dichlorophenoxyacetic acid (2,4-D) is a common systemic herbicide used in the control of broadleaf weeds. It is the most widely used herbicide in the world, and the third most commonly used in North America. Finally, Clopyralid is notorious for its ability to persist in dead plants and compost, and has accumulated to phytotoxic levels in finished compost. Because of their toxicity, their high detection levels in water resourses and their capability to be incorporated into a protocol of polymer synthesis, these three molecules are also selected to be used as templates for the synthesis of MIPs.
To this direction, the above mentioned template molecules were selected for MIP synthesis (solely or in combination) using either commercial available functional monomers and cross-linkers (i.e. methacrylic acid, itaconic acid, trifluoromethacrylic acid, 4 vinyl pyridine, 1 vinyl imidazole, ethylene glycol dimethacrylate, trimethylopropane trimethacrylate, divinylbenzene, etc) (CPERI, ULUND, LNU/HiK, USTUTT, MIPTech) or chemically synthesized monomers (i.e 1,3 diaryl urea based monomers) (UDO) since it was recently shown that a stoichiometric imprinting approach based on the use of 1,3-diaryl urea based monomers led to water compatible MIPs, capable of extracting the target from water-rich media. This has resulted in identification of host monomers displaying much enhanced affinity for such guests accompanied by chromogenicity and fluorogenicity. The latter suggests that such MIPs could exhibit responsive properties useful in sensing applications (WP4). Additionaly, polymerizable cyclodextrins that were obtained after synthesis were used together with charged monomers in the MIP synthesis to interact with the Penicillin G potassium salt. The objective was to use hydrophobic interactions in combination with electrostatic interactions (USTUTT). Also, for the synthesis of MIP nanoparticles that will be suitable for incorporation into membrane structures, additional non-interfering monomers were selected so as to introduce clickable tags on the surface of MIP nanoparticles, including propargyl acrylate and glycidyl methacrylate (ULUND). The choice of the functional monomers was based on previous results as well as on the optimization made within the project by computational tools or/and combinatorial screening (Tasks 1.2 and 1.3) (LNU/HiK, UDO, CU).
Finally, an extra effort has been placed on the production of imprinted polymers in large scale (MIPTech). To this direction suspension polymerization was initially utilized for the synthesis of MIP microparticles towards triethylmelamine for the selective separation of triazines and triazine betabolites. The MIP microparticles produced by this technique were in the range of 20-90 µm, but since suspension polymerization takes place in an aqueous environment, it was expected that the transfer from the micro- to the nano- or submicro- scale, which is necessary in WP2, would be feasible. Triethylmelamine was preferred as an alternative template because it avoids the use of atrazine or other potentially harmful chlorotriazines as templates during the polymer synthesis. For the further optimization of the results the work focused on the modification of the produced polymers in order to enhance their hydrophylic character. It was found that the MIPs towards triethylmelamine are able to extract atrazine from water even at very low concentrations (i.e. 1 ppb).

Task 1.2: Computational optimization of MIP composition
For the optimum choice of the reaction parameters (e.g. functional monomers, cross-linker, etc.) computational methods have been applied. These computational techniques utilize molecular dynamics simulations to perform virtual, high throughput screening of a library in order to guide the selection of the functional monomers for the optimum synthesis of MIPs. By this way the investigation of the highly complex systems, consisted of thousands of atoms, has been realized at reasonable computational costs.
More specifically, a virtual library consisting of around 20 commonly used functional monomers has been created in order to screen the possible interactions of the functional monomers with the molecular model of each template selected at Task 1.1. These functional monomers were acidic, basic or neutral molecules. Most of the monomers used for MIP preparation are described extensively in the literature as a result of their ease of thermo or photo polymerization, availability and cost. For the library preparation the charges for each atom of each monomer were calculated and the structures of the monomers refined using molecular mechanical methods. The monomers that give the highest binding score with a specific template molecule, lead to a polymer with higher affinity and specificity. Since, at the most polymerization techniques applied by the involved partners the presence of a solvent during the polymerization reaction was necessary, this parameter was also investigated. To this direction, the series of simulations were done separately for each solvent. Also, the simulations repeated using apart from ethylene glycol dimethacrylate, trimethylopropane trimenthacrylate as cross-linker.
Once the best monomers and solvent had been selected, refining experiments were carried out using molecular modelling to analyse the organization of the monomers around the template molecule. The monomers, which gave the highest binding scores during the screening step, were placed around the template, in order to simulate the pre-arrangement of the monomers (with the template) in the mixture prior to polymerization. At the end of the program, the number and the position of the functional monomers were examined to ascertain information regarding the optimum functional monomer template ratio and consequently the optimum MIP composition (CU).
Additionally, the work focused on the imprinting effect of polymers prepared in different formats (nanoparticles, nanogels, etc) by different polymerization techniques (i.,e., precipitation and miniemulsion polymerization). A strong collaboration between the partners initiated so as to combine the information derived from the theoretical study (HiK/LNU, CU) with the results obtained from the experimental work (USTUTT, UDO and CPERI) and, thus, to synthesize optimized imprinted polymers.
The molecular dynamics simulations and the correlated experimental data revealed insights into the effects of the structure of the functional monomer and the functionality of the template molecule on the imprinting efficiency. Based on simulations the conclusion is that all components need to be present and explicit in correct stoichiometries in order to obtain realistic result, which shows strong correlation to experimental data obtained through polymer evaluation and NMR measurements. Another conclusion is that in order to obtain statistically relevant data, sufficient molecular entities and simulation time are required. These are factors that need to be weighted carefully against each other to ensure relevant results. Another fact that emerged through the studies of different polymer prepolymerization compositions is that each system is unique and needs to be treated as such. (LNU/HiK) The results derived from the theoretical study were distributed to the rest of partners so as to be utilized for the synthesis of MIP nanoparticles in WP2.

Task 1.3: Combinatorial screening of MIP
For the optimum choice of the reaction parameters also combinatorial methods has been utilized. Combinatorial screening (high-throughput analysis) is based on the preparation of a large number of compounds or materials in parallel, if possible in a form that is easy to screen for properties of interest. Small-scale procedures for the rapid synthesis can be developed based on the preparation of small amounts of material (mini-MIPs) at the bottom of small vials or of wells of 96-well plates and their in situ testing by equilibrium batch rebinding experiments. The general procedure is essentially a scaled-down version of the traditional monolith procedure. The small scale protocol can also be partly automated. Using the parallel analysis mode the complete polymer library can be evaluated in one to two weeks.
To this direction, the work has focused on the development and characterisation of MIPs targetting endocrine disruptors (EDCs) (UDO). The best performing combinatorial MIPs based on a 96 well library were upscaled and characterised with respect to binding affinity and crossreactivity for other endocrine disruptors. It was shown that the combinatorial MIPs were capable of strongly retaining the EDCs in fair agreement with their respective affinity for the native estrogen receptor. The MIPs were then assessed as sorbents for EDC extraction from spiked water samples at ppb level. Such selective filters may ultimately be useful in water processing handling pollutants not removable by the established adsorbents such as active carbon. The results showed that the established adsorbents such as active carbon and C18-silica only resulted in low removal efficiency, while optimized MIPs were effective for quantitative or near quantitative removal of the entire group of tested EDCs.
Apart from this work effort has been placed on the combinatorial synthesis of MIPs in the form of nanoparticles using either precipitation or miniemulsion polymerization (CERTH, ULUND and USTUTT) in order to optimise both their composition and their physical properties, such as size, morphology and particle strength/resistance. More specifically, it was studied the effect of important process parameters, such as presence of a second template molecule, the type of the functional monomers (acidic, basic, neutral), the cross-linker (bifunctional or trifunctional) and the solvent (polar, non polar) on the final product using as template molecules pesticides, pharmaceutically active compounds and endocrine disruptors (e.g. atrazine, simazine, bisphenol A, 17beta -estradiol and propranolol).
It was found that all the mentioned parameters can influence the final product and thus it is extremely important to be taken into account for the optimum synthesis of MIP nanoparticle that will be used for the production of MIP composite membranes (i.e. sandwich-type by CERTH and USTUTT and nanofiber-based membranes by ULUND).

WP 2: COMPOSITE MEMBRANES MADE OF MIP NANOPARTICLES
Objectives: Synthesis of molecularly imprinted polymers with well defined morphology (i.e. nanoparticles and microgels) by precipitation, emulsion and miniemulsion polymerization techniques. Production of novel composite membranes utilizing preformed MIP nanoparticles, by embedding nanoparticles either in a plasticizer or between two stable membranes-substrates (sandwich-type). Preparation of nanofiber membranes using MIP nanoparticles as building blocks.
Action Done / Main Results

Task 2.1: Synthesis of molecularly imprinted nanoparticles
For the synthesis of molecularly imprinted nanoparticles different polymerization and grafting techniques were utilized. The target was to produce uniform nanoparticles with totally controllable physical properties that will be successfully incorporated into the composite membranes of WP2.
More specifically, precipitation polymerization was employed for the synthesis of MIPs using pesticides, pharmaceutically active compounds and endocrine disruptors (e.g. atrazine, simazine, 2,4-dichlorophenoxyacetic acid (2,4-D), propranolol, pentoxifilline and bisphenol A). In this case, both bifunctional and trifunctional cross-linkers, such as ethylene glycol dimethacrylate (EGDMA) and trimethylopropane trimethacrylate (TRIM) respectively, were employed for the synthesis of polymers. Trifluoromethacrylic acid (TFMAA), acrylamide (Am), itaconic acid (IA), 4 vinylpyridine (4VP), 1 vinyl imidazole (1 VID) and methacrylic acid (MAA) were used as functional monomers. Acetonitrile, methanol, tetrahydrofurane, toluene and the mixtures of acetonitrile-toluene and acetonitrile-water were used as solvents. In all cases, the polymerization was thermally initiated (at 60°C), using 2,2´-azobisisobutyronitrile (AIBN) as initiator (CERTH, USTUTT).
It was found that the structure of the polymer, the specific surface area and the mean pore diameter at the polymer surface were dramatically dependent on the presence of the template molecule in the reaction mixture, the type of the solvent used as porogen and the type of the cross-linker (e.g. its functionality). Additionally, it was proved that the type of the functional monomer affected not only the non-covalent interactions during the process of imprinting, but also the structure of the polymer. Actually, when the solubility of the polymer in the reaction solvent was low, the phase separation occurred earlier and led to polymer microspheres with smaller size. The best results were obtained when pure acetonitrile was used as solvent. In this case, small isolated particles with a very narrow polydispersity index were produced.
The second step was to study the effect of the process parameters on the imprinting efficiency and the correlation between the obtained polymer structure and the rebinding capacity. Thus, several batch-wise guest binding experiments were carried both in organic and aqueous solutions to determine the rebinding capacity of the synthesized imprinted polymers towards the template molecule. In case of atrazine, when Am, MAA were used as functional monomers and EGDMA as cross-linker the binding capacity (selectivity) of the imprinted polymers to the atrazine molecules was higher than that of the non-imprinted polymers even when the binding experiments were carried out in aqueous solutions. The results were also positive regarding the binding efficiency when simazine was used as template molecule and MAA and TRIM as functional monomer and crosslinker, respectively. In case of Bisphenol A, basic functional monomers were chosen, such as 4VP and 1VID, while both EGDMA and TRIM were used as cross-linkers. The best results were obtained when the synthesis was realized using 4VP and EGDMA and acetonitrile instead of methanol. For the imprinting of Pentoxifylline again EGDMA was used as cross-linker and acetonitrile as solvent, while both MAA and 4VP were used as functional monomers. The binding experiments showed that in both cases the binding of Pentoxifylline to MIPs and NIPs increases when the initial concentration of the analyte increases and that differencies of the binding are observable only at the highest concentartion (i.e. 100ppm).
Also, molecularly imprinted nanoparticles with a well-defined core-shell structure were synthesized (ULUND). The core of the nanoparticles contain molecular binding sites for two target compounds, propranolol and 2,4-dichlorophenoxyacetic acid (2,4-D). After formation of the imprinted core, new monomers were added into the precipitation polymerization system to grow a loosely cross-linked shell that contains either acetylene or epoxy group. The epoxy was converted into azide in one step. Finally, MIP nanoparticles bearing surface exposed clickable chemical groups were obtained. By introducing clickable acetylene and azide groups on particle surface, the selective MIP nanoparticles are made easy to be immobilized on a variety of solid support, e.g. for developing composite nanoparticle-based membranes.
In the case of miniemulsion polymerization, DVB, EGDMA and TRIM were used as cross-linkers and 4VP, MAA and IA as functional monomers. Additionally, chloroform, toluene or tetrachloroethylene were added as porogens into the oil phase (CERTH, USTUTT). The effect of the type and the content of the cross-linker, the functional monomer and the initiator on the particle size and the polymerization yield was extensively studied, realizing both photo- and thermal-polymerization. Apart from sodium dodecyl sulfate (SDS) both cationic surfactant and non-ionic surfactants were used for the stabilization of the oil phase. It was proved that this method provides polymeric nanoparticles in relatively high yield and with controllable physical properties, such as size (around 100-150nm), morphology and particle strength/resistance, especially when the proccess parameters are carefully chosen (see details in appendixes). Also, this technique was used for scale up studies for the synthesis of atrazine imprinted polymeric nanoparticles, since miniemulsion polymerization is considered to be an economical and scalable method (MIPTech). These nanoparticles can be used to prepare sandwich-type nanoparticles-based membranes, nanoparticles-embedded polymer membranes and polymer nanofibers with encapsulated nanoparticles in other parts of the project.
In particular, by this polymerization technique atrazine, pentoxifylline and Bisphenol A were used as template molecules. When batch-wise guest binding experiments of the produced P(MAA/EGDMA) nanoparticles towards atrazine were realized in aqueous solutions of the template molecule it was shown that the binding capacity of MIPs and NIPs is higher (as compare to that in organic medium) partially due to the hydrophobic character of atrazine. Also, it was shown that the presence of tetrachloroethylene during the polymer synthesis increases the specific surface area of the produced polymeric nanoparticles and thus improves their binding capacity. It was also proved that the binding capacity of MIPs is higher than that of NIPs, thus proving the selectivity of the produced polymers, even in aqueous medium.
For the imprinting of Bisphenol A (BPA), in the case of miniemulsion polymerization, EGDMA was used as cross-linker and the cross-linker content were 80% for both polymer systems (p(MAA-co-EGDMA and p(4-VP-co-EGDMA)). In both cases MIPs adsorbed higher amounts of BPA than NIPs, especially in case of using MAA as functional monomer, thus proving the imprinting effect. For the further optimization of the system miniemulsion polymerization was carried out with different molar ratios of template molecule to functional monomer. More specifically, syntheses with various templatemonomer ratios were performed. It was found that when the amount of BPA as template is increased during the synthesis the amount of adsorbed BPA at the binding experiment is increased, too. To determine the best template-monomer ratio for the MIPs against BPA modelling studies were also performed (CU, HiK/LNU) with BPA and MAA. The results indicated the optimum ratio at which the stronger interactions are foreseen. These results were very close to the previous experimental results and thus selected for future research. It was also proved that the presence of MAA is not necessary for the increase of the binding capacity but for the optimization of the polymer structure which is strongly related with MIPs binding efficiency.
In order to test the selectivity of the MIPs, competitive adsorption experiments were performed using three analogue compounds (Bisphenol A bis(2,3-dihydroxypropyl) ether (BPA-Ether), para-Bisphenol F (p-BPF) and ortho-Bisphenol F (o-BPF)). It was shown that MIPs towards BPA are able to adsorb molecules with a very similar structure to the template molecule, and thus can be used as group selective adsorption material.
Also, novel MIP nanoparticles have been developed for various triazines by iniferter grafting technique using modified dendrimers as core (CU). Dendrimers represent a novel class of structurally controlled macromolecules deriving from a branches-upon-branches well defined structural motif. They are highly branched macromolecules that radiate from a central core and are synthesized stepwise, repetitive reaction sequence that guarantees complete shells for each generation, leading to polymers that are monodisperse. This method gives rise to nanoparticles which permit nearly complete control over the critical molecular design parameters, such as size, shape, surface/interior chemistry, flexibility, and topology.
Moreover, the work has focused on the development of both iniferter and RAFT based grafting techniques on mesoporous silica for final transfer to nanoparticulate format. One technique is based on the iniferter mediated grafting from silica to full monomer conversion. This technique was applied to the grafting of stoichiometrically imprinted polymers for BLAs using two different urea-based monomers (UDO). The polymers were subsequently packed in HPLC columns and tested for their ability to retain the template PenG in two different mobile phases. It was shown that the MIP prepared using the more potent monomer displayed as a rule higher retentivity and imprinting factors than the MIP prepared using the weaker binder. This also resulted in pronounced imprinting effects when assessed using a water rich mobile phase.
Finally, regarding bupivacaine imprinting a good correlation between MD simulations and experimental studies (BET, radioligand binding) was observed, indicating that the developed method for MD simulations is accurate and highly relevant in the development and production of MIPs (HiK/LNU). The MD simulations showed that both monomers, functional and cross-linker (MAA and EGDMA) interacts with the template. The most stable interaction being between the acidic proton of MAA and the carbonyl oxygen of bupivacaine. An interesting observation from simulations is the competitive situation emerging regarding hydrogen bonding between the carbonyl oxygen of either MAA or EDMA and the amide proton of bupivacaine. It could also been concluded that the hydrogen bonding interactions between EGDMA and the template are only mildly affected by the different molar fractions of MAA studied, with only a small local optimum regarding hydrogen bonding between cross-linker and template of minor significance.

Task 2.2: Preparation of sandwich-type nanoparticles-based membranes
One promising approach for the production of MIP membranes is based on the confinement of MIP nanoparticles, as a filter cake, between two microfiltration membranes (sandwich-type membrane). To this direction the work focused on the preparation of sandwich-type nanoparticles-based membranes using polyamide membranes (N66, Pall) with a diameter of 45 mm and a pore size of 0.1 pm as support and cover of the composite membrane. One type of imprinted nanoparticles were prepared by miniemulsion polymerization using atrazine as template molecule, MAA as functional monomer and EGDMA as crosslinker. For coating the support, 50 mL suspension of 0.05 g nanoparticles in ultra pure water was filled in an ultrafiltration cell (Amicon 8050, Millipore) with the installed support membrane. The suspension was pressed through the support membrane by nitrogen at a constant pressure of 0.1-0.2 bar, causing deposition of the particles onto the surface of the membrane. After drying the coated membrane, the nanoparticle layer was covered by a second membrane of the same material. In order to check the efficiency of the system 20ml of 1ppm triazine solution in water passed under constant pressure through the composite membrane and the concentration of triazine at the ultrafiltrate was measured with HPLC. It was found that MIP-membrane bound higher quantity of the template molecule (i.e. atrazine) than NIP-membrane. Additionally, when the binding experiment was repeated using nanoparticles that were synthesized in the presence of tetrachloroethylene, the comparative results towards other triazines (i.e. simazine and propazine) showed non-selective binding, thus proving the selectivity of the produced membranes (CERTH).
Adittionally, a second sandwich-type membrane test system was constructed and built up. The pressure-stirring cell with a total volume of 75 mL is made of stainless steel. The diameter of the cell is 47 mm and the cell resists a maximum pressure of 100 bar. The cell is connected to a nitrogen gas tank and the pressure can be controlled by a valve and a pressure indication. The system has the possibility to be upgraded for continuous flow operation. To this system alternative support- and cover membranes that have the purpose to give a solid structure for the selective MIP layer between the membrane sheets were suggested. The support- and cover membranes have to be unalterable in shape and size. Thus, these alternative membranes were tested under applied pressure in different solvents, solvent-water mixtures and pure water. All of them are either hydrophilic or have a hydrophilic top layer (USTUTT). After the choice of the optimum support- and cover membranes (Nylon 66), adsorption experiments were realized using also membranes with no particles and with non-imprinted particles so as to evaluate the amount of non-specific adsorption on the support- and cover membrane and the non-imprinted particles. The results show that the MIP-membrane adsorbs nearly the double amount of Bisphenol A compared to the NIP-membrane, while compared to the sandwich-type membrane with no particles the MIP-membrane adsorbs nearly four times more Bisphenol A. To test the selectivity performance of the MIP-membrane competitive adsorption experiments were performed using three analogue compounds. MIP-membrane adsorbs BPA most of all, followed by the o-BPF, the p-BPF and the BPA-Ether. The difference in the adsorbed amounts of the different molecules can be explained mainly by the size of the molecules, the polarity and the hydrophobic behaviour of the compounds.

Task 2.3: Preparation of nanoparticles-embedded polymer membranes
A very promising approach for the production of MIP membranes is based on the confinement of MIP nanoparticles as specific adsorber embedded in a polymer matrix. Embedding MIP particles into a suitable porous polymer matrix, in analogy to the existing commercially established membrane-SPE discs, will be an alternative approach to be pursued. In comparison with the membrane-SPE discs, these membranes are prepared by wet phase inversion process, which is widely used in membrane industry for manufacturing membranes, but in the presence of nanoparticles instead of microparticles. By this technique, a homogeneous polymer solution is cast as a thin film and then, immersed into a nonsolvent coagulant bath. The diffusional exchange of solvent and nonsolvent across the interface between casting solution and nonsolvent can make the casting solution phase-separated to form a membrane with a symmetric or asymmetric structure.
The first step was the selection of suitable commercially available polymers and the appropriate solvent for dissolving these polymers. The dissolving tests were performed, initially without nanoparticles and using various solvents in order to optimize the preparation conditions. To investigate the influence of polymer content of the casting solution on the properties of the membranes (e.g. thickness, flux, etc) the polymer content of the casting solution was varied from 10 wt% to around 20 wt%. After the characterization of the membranes without any nanoparticles, membranes with non-imprinted (NIP) and imprinted particles (MIP) were prepared to evaluate the effect of the polymeric particles on the membranes structure. As polymer particles p(MAA-co-EGDMA) NIPs and MIPs were used. The casting solution for the membrane preparation was prepared with around 10 wt% polymer content and more than 30 wt% nanoparticles related to the polymer matrix. It was proved that the nanoparticles were very well distributed in the entire membrane and no particle agglomerates were visible. It was also apparent that the particles are attached at the wall of the big finger pores and enclosed in the small sponge-like pores of the membrane walls. After characterization of the membrane morphology, thickness and flux, adsorption experiments were carried out in a continuous flow mode.
The results of the adsorption tests showed that the membrane prepared with MIP particles bound up to seven times more template molecule (i.e. BPA) as compared to that with no particles or to that containing NIP particles. This shows clearly the improvement of the adsoprtion properties of the final membrane by embedding polymer MIP nanoparticles in the polymer matrix.
Also, the work focused on the preparation of catalysts for subsequent fixation on membranes, the testing of different catalyst preparation methods and the deposition of catalyst on inorganic material similar to the membranes (vTI). Additionally, work focused on the immobilisation of MIPs inside a hydrogel coating based on PVA (poly vinyl alcohol) in order to introduce the MIP functionalities to the inorganic membrane (KeraNor). PVAs are synthetic polymers, which can form hydrogels and provide a highly elastic non-biodegradable working material with good diffusion properties. Therefore tests have been realised to check the MIP adsorption capacity in the membrane coating.

Task 2.4: Synthesis of nanofiber-based membranes
The work focused on the production of nanofiber-based membranes containing molecularly imprinted polymer (MIP) nanoparticles (ULUND). The nanofiber membranes were prepared using electrospinning technique. Electrospinning involves the use of a high-voltage electrostatic field to charge the surface of a droplet of polymer solution, thereby inducing the ejection of a liquid jet through a spinneret. In a typical electrospinning process, an electrical potential is applied between a droplet of a polymer solution held at the end of a capillary tube and a grounded target. When the electric field that is applied overcomes the surface tension of the droplet, a charged jet of polymer solution is generated. On the way to the collector, the jet is subjected to forces that allow it to stretch immensely. Simultaneously, the jet is partially or fully solidify through solvent evaporation or cooling, and an electrically charged fiber remains, which can be directed or accelerated by electrical forces and then collected in sheets or in other useful formats. A characteristic feature of the electrospinning process is the extremely rapid formation of the nanofiber structure, which occurs on a millisecond scale. Other notable features of electrospinning are a huge material elongation rate in the order of 1000 s-1 and a reduction of the cross-sectional area in the order of 105 to 106, which have been shown to affect the orientation of the structural elements in the fiber.
Electrospinning offers the possibility to construct affinity nanofibers based on pre-made MIP nanoparticles. MIP nanoparticles have been encapsulated into polymer nanofibers using electrospinning, where the nanoparticles are simply dispersed in the polymer solution. The conditions regarding the physical size and surface polarities of MIP nanoparticles, the electrospinning parameters, the choice of different supporting polymers affected the quality and productivity of the obtained composite nanofibers. Thus, these important parameters have been optimized in order to achieve uniform nanofiber materials containing encapsulated MIP nanoparticles. The nanofibers containing encapsulated MIP nanoparticles showed selective binding for the original templates (estradiol, theophylline, propranolol) used in the imprinting reaction, suggesting that the encapsulation-by-electrospining can be a general approach to introduce MIP nanoparticles into functional membranes.
It was proved that electrospinning is a simple and efficient method that allows composite nanofiber membranes to be fabricated and selective molecular separation for target water pollutants. Both hydrophobic and hydrophilic nanofiber membranes can be produced, and pre-made molecularly imprinted nanoparticles can be easily encapsulated to tune the binding selectivity of the composite nanofiber membranes. Given that small and uniform MIP nanoparticles can be synthesized for a wide range of target pollutants, the electrospinning method can be adopted to fabricate an array of nanofiber-based membranes, providing possibilities to separate multiple water pollutants through simple filtration procedures.

WP 3: MOLECULARLY IMPRINTED POLYMER AND HYBRID MEMBRANES
Objectives: Production of composite filters both on organic and inorganic support via novel grafting techniques and preparation of MIP membranes through the simultaneous method.
Action Done / Main Results

Task 3.1: Synthesis of molecularly imprinted polymer membranes
An effort has been placed on the synthesis of MIP membranes prepared by in-situ approach. The work focused on the testing of various parameters for the synthesis of the membranes. More specifically, different solvents (i.e. dichloromethane, DMF, chloroform, toluene, acetone and ethyl acetate), pore forming components (i.e. PEG with MW up to 20,000, polymethyl methacrylate with MW around 15,000, polyvinyl alcohol with MW 10,000 and polyvinyl acetate with MW up to 500,000), plasticizers (polyurethane, polybutadiene and polyisoprene of relatively low molecular weight, ~ 3,000-5,000) and initiators (i.e. azobisisobutyronitrile, azobisdimethylvaleronitrile and azobiscyclohexane carbonitrile) were used. In all cases, methacrylic acid and tri(ethyleneglycol) dimethacrylate were used as monomers. The optimum system was proved to be that of using PEG, polybutadiene and DMF as pore forming component, plasticizer and solvent respectively. In that case, the thickness of the imprinted membranes (towards atrazine) varied from 100 to 150?m and their weight from 9 to 10mg. In order to study the substrate uptake of atrazine, which was used as template molecule, heterogeneous batch experiments were carried out both in organic solvent (i.e. acetonitrile) and water. The rebinding experiments showed differences between MIPs and NIPs proving the imprinting efficiency even in aqueous environment.
Also, a first attempt has been made for the synthesis of poly(acrylonitrile-co-acrylic acid) [P(AN-co-AA)] membranes by phase inversion method. For the casting process, DMSO and water were selected as cast solvent and coagulation medium of the copolymer, respectively, while theophylline was chosen as template molecule. The resulting membrane was washed with 0.1 wt. % acetic acid (AcOH) aqueous solution for template extraction and then was rinsed repeatedly with a large excess of water for the removal of acetic acid. Finally, the membranes obtained were dried by lyophilisation. The membrane structure consists of a dense top layer that seems to cover another porous support sublayer with finger like structure especially when the molar ratio of acrylic acid to acrylonitrile was increased. In order to study the substrate uptake of THO, both heterogeneous batch experiments and SPE are in progress. (CPERI)
Also, the work focused on studies with QCM-surfaces as models and results are pending, as well as chemometric studies regarding correlations of binding in various media with MIP morphology. Grafting methodologies are being established, with thiol-based chemistries and electropolymerisation methods now having been established in HiK/LNU laboratory.
Several different methods for fusing MIPs and QCM-chips are under evaluation. This work includes approaches for preparing molecularly imprinted surfaces and thin films have been explored showing promising results. A range of MIPs and reference polymers have been designed and evaluated using MD simulations and spectroscopic (NMR) techniques. Based on this, a series of bulk polymers have been produced and evaluated using radioligand and fluorescence methods. The promising systems have been used in the preperation of MIP-QCM surfaces and initial evaluation indicates functionality and template selectivity. The incorporation of electropolymerisation methods has shown great possibilities for preparing molecularly imprinted (QCM) surfaces. These MIP-QCM-chips provides valuable information regarding surface grafting and prodution and fusion of surfaces and thin films with molecular recognition properties. Furthermore, these studies show potential as model systems for the selective extraction and, especially, quantitative analysis of specific molecular species in solution through utilization of molecularly imprinted or fused surfaces.(LNU/HiK)

Task 3.2: Synthesis of multi layer MIP-polymeric / ceramic membranes
MIPs films were produced on ceramic substrates provided by KeraNor (ceramic disks with 25 or 44mm diameter and 1.3mm or 0.8mm thickness). The composite MIP filter was prepared using an azoinitiator, physically adsorbed to the surface. This technique is based on the confinement of the initiated radical to the support surface so as to achieve a higher density of the grafted polymer chains. In the absence of chain transfer this could lead to chain growth occurring mainly from the surface of the support with minimal polymerization occurring in solution.
The percentage of polymer loading was determined gravimetrically. In ceramic samples with TiO2 the polymer loading was varied from 3 (NIP) to 3.5 (MIP) % and without TiO2 from 3.4 (NIP) to 5.3 (MIP) % based on the weight of the ceramic substrate. It was proved that even after many washing steps the polymer grafted on the ceramic substrate was very stable. In order to test the binding efficiency of the synthesized membranes continuous filtration experiments were performed. The equipment used for dead-end filtration consisted of the filtration cell (where the membrane was placed), the reservoir (for the increase of the volume sample) and the nitrogen gas tank connected by a valve for controlling the applied pressure. The adsorption experiments were performed with desmetryn solutions in water of various initial concentrations (500, 100, 10 and 1 ppb). It was proved that in all cases MIPs adsorbed more desmetryn than NIPs, thus, proving their selectivity.
Also, alternative methods have been developed for grafting MIPs over aluminium oxide membranes provided by Keranor. A layer by layer grafting approach has been developed for controlled deposition of cross-linked polymers by living radical polymerisation. Composite membranes (pore size 4 µm, 25mm diameter) were modified with, in order to confine the iniferter reactive groups solely at the surface, then placed in solution with monomers and cross-linker (EGDMA) along with the template to generate MIPs onto the surface of membranes only. The polymerisation was initiated by UV irradiation. Formation of the cross-linked MIP polymers was studied in terms of time course of the reaction, type of monomers incorporation and influence of various porogens (acetonitrile, DMF). Gravimetrically and through changes in contact angle, grafting of MIPs over membranes was confirmed. Removal of atrazine was done using methanol and acetic acid mixture. As a result, membranes covered with a thin layer of imprinted polymer selective for atrazine were obtained with different degrees of modification and different binding properties in aqueous medium. Equilibrium binding assays followed by HPLC detection were used to determine atrazine in aqueous samples. MIPs showed better binding than corresponding NIPs in pure aqueous medium. This experiment was done to validate the results regarding the selection of the monomers obtained by computational modelling (CU).
In order to increase the specific surface area, the porous, ceramic substrate has also been coated externally and internally with different oxide coatings as titania and gama-alumina. As these coatings have significantly larger specific surface area, the resulting porous material including the coating, will achieve a much larger specific surface area available for grafting of MIPs (KeraNor).

Task 3.3: Synthesis of hybrid MIP membranes
For the synthesis of hybrid MIP membranes attempts have been made to develop novel grafting techniques on polypropylene filtration membranes and aluminium oxide composite membranes which were utilised for grafting MIPs and will be used for ultra-filtration and separation purposes.
Different approaches are compared for the synthesis of thin-layer molecularly imprinted polymeric hybrid membranes (MIMs) based on various functionalised materials, for sensing and rapid pre-concentration of analytes (e.g. atrazine) from complex matrices. The unique features of these MIMs is the interplay of selective binding and transmembrane transport of molecules, making them potentially superior to state-of-the-art synthetic separation membranes already applied in various industries.
The first approach used, involved functionalising hydrophilised polypropylene (PP) membranes using aminolysis, followed by immobilisation of dithiocarbamate iniferter. UV photoinitiation generated thin MIP films of controlled thickness and functionality over PPs. Then atrazine-MIMs for atrazine, using methacrylic acid as functional monomer onto the surface of iniferter-functionalised PP membranes were synthesised. Membranes were characterised through SEM, contact and angle and IR in ATR mode. Degree of grafting was calculated gravimetrically. Imprinted membranes showed highly selective binding of atrazine from aqueous mixtures of structurally similar herbicides.
In the second approach, a novel method has been recently developed, which involved the oxidative polymerisation of novel functionalised aniline-containing monomers over hydrophobic PP membranes, with incorporation of iniferter (dithiocarbamate ester) groups. The main advantage of the latter approach is the potential to synthesize MIMs by controlled deposition onto any type of polymer support. MIP preparation is rapid and localized, due to photoiniferter being directly attached to the membrane surface, resulting in thinner MIP layers for faster separation and binding. MIP membrane recognition properties were evaluated by adsorption of herbicides from aqueous solution during filtration followed by HPLC quantification. MIP membranes showed 94% binding for 125 ?g/mL atrazine, while non-imprinted (blank) reference membranes bound only 15 % under the same conditions.(CU)

WP 4: MEMBRANE PERFORMANCE ASSESSMENT AND MONITORING
Objectives: Membrane testing. Separation and catalytic decomposition of the pollutants. Advanced monitoring of the target compounds.
Action Done / Main Results

Task 4.1: Stability and regeneration of MIMs
The binding capacity of composite membranes was evaluated repeatedly so as to confirm their ability to be regenerated successfully. The evaluation of the regeneration properties of the composite membranes were carried out by realizing multiple adsorption and desorption experiments. Gravimetrically, it was proved that the weight of the filters was kept almost the same, thus, proving their stability regarding the grafting of the polymer onto the ceramic substrate. It was also shown that the composite filters retain their recognition properties almost independently from the repeated binding and washing steps (CERTH, USTUTT).
Electrospun nanofiber composite was also used as affinity matrix for solid phase extraction (SPE) of propranolol and atrazine from spiked water samples (100 ng/mL). Prior to use, the composite membrane was treated with methanol and water. Samples were loaded in propranolol- and atrazine-spiked water, washed with acetonitrile containing different amount of ammonium acetate buffer and finally eluted with a mixture of methanol/Water/formic acid.
The nanofiber membrane was tested in repeated SPE operations. Between two consecutive tests, the nanofiber membrane was regenerated to ensure that no carryover of the target compounds took place. The propranolol recovery from the MIP nanofiber at the elution step remained at the same level of 80 % to 90 %, indicating the MIP nanofiber composite has very high stability (ULUND).

Task 4.2: Separation and catalytic decomposition of the pollutants
Initially, the work focused on screening for suitable nanoscale metal and metal oxide catalysts and on testing systems for catalytic decomposition of the pollutants to convert them into harmless compounds. Clopyralid was initially selected as model substance. For the degradation of Clopyralid a batch reactor operated at room temperature and under atmospheric pressure was used. The degradation of Clopyralid was realized either with Fenton catalyst or by Hydrodechlorination (HDC). After the screening of suitable catalysts for decomposition of the pollutants, the next step was the fixation of the catalysts and the MIPs to the membranes resulting in MIP adsorbent/catalyst membrane systems able to remove and to degrade the pollutants.
Mounting a catalytic functionality to membranes was successfully accomplished by two different ways: a) impregnation of a alfa-Al2O3 membrane with a gama-Al2O3 washcoat with noble metal precursors and subsequent activation resulted in active catalysts for the hydrodechlorination reaction and b) embedding preformed powdered catalysts in polymeric membranes. So, different ways of the catalytic degradation were tested, namely the separate catalytic treatment of the pollutants after their adsorption on MIPs and successive desorption from the MIPs as well as the catalytic treatment with a catalytically active membrane.
For the preparation of the composite MIP adsorbant systems, ceramic membranes used as support material were provided by the project partner KeraNor AS. First, the ceramic membranes were tested for their flux properties with water at room temperature. Mounting of the MIPs on the ceramic alumina membrane was achieved by a polyvinyl alcohol (PVA) hydrogel coating. Further on, it was tested if the PVA coating might interfere with the MIP adsorption and desorption ability. Therefore, atrazine adsorption experiments were carried out with the parental alfa-Al2O3 membrane, with a PVA coated alfa-Al2O3 membrane and the MIP/PVA coated alfa-Al2O3 membrane. Atrazine adsorption of the MIP/PVA coated alfa-Al2O3 membrane was further compared to the free MIPs as provided by CERTH.
These experiments were carried out using an atrazine concentration of 15 mg / l (3 mg atrazine in 200 ml deionized water). The results showed that the alfa-Al2O3 membrane does not show any atrazine adsorption and the PVA coating only a negligible adsorption. The MIP/PVA coated alfa-Al2O3 membrane adsorbs in 60 min only slightly less atrazine than the free MIPs, i.e. 16 mgatrazine / gMIP compared to 17.6 mg atrazine / gMIP, which means that 91 % of the MIP adsorption capacity could be preserved in the PVA coating. The atrazine adsorption capacity raises linearly with increasing MIP loading. The adsorption/desorption of atrazine from the MIP/PVA coated alfa-Al2O3 membrane was tested in continuous mode and it was proved that even after several adsorption/desorption cycles the system preserves its stability.
In parallel, work was done to mount a catalytic functionality to the ceramic membrane. Those catalytically active membranes were tested for the catalytic degradation of chlorinated pollutants by means of the hydrodechlorination (HDC) reaction. This reaction is applicable for the treatment of chlorinated pollutants, i.e. the treatment of clopyralid, atrazine, 2,4 dichlorophenoxyacetic acid (2,4 D) as selected target compounds in the WATERMIM project. The HDC reaction is catalyzed by noble metal catalysts, namely palladium or platinum, which need to be mounted to the ceramic membrane. Therefore, an impregnation procedure has been elaborated. The catalytically active membranes were used for HDC of clopyralid. The results revealed that it is possible to prepare catalytically active membrane systems and to use them for the catalytic degradation of pollutants.
In a second approach a preformed catalyst supported on powdered gama-Al2O3 was embedded together with preformed spherical MIPs imprinted against atrazine in a polymeric membrane. MIPs imprinted against atrazine were provided by MIPTech/Biotage and the polymeric membranes were prepared by USTUTT. It was shown that the particle size of the MIPs and the size of the catalyst agglomerates are not suitable for the preparation of a thin flat sheet membrane. It can be assumed that due to particles size and agglomerate size the mechanical stability of the membrane is weakened. If particles are embedded in flat sheet membranes prepared by phase inversion method the particle size should be smaller than one µm. To check whether the catalyst preserved its catalytic activity or was damaged by the membrane preparation procedure, the composite MIP adsorbant/catalyst system was tested for the HDC of clopyralid. Under these conditions only 45 % clopyralid degradation was obtained within 15 h. Such low activity is not tolerable. Obviously the catalyst was damaged by the membrane preparation procedure, most likely by adsorbed monomers/polymers. However, a low residual catalytic activity was still noticed.
The test with combined MIP/catalyst/membrane systems revealed that those systems although in principle working may not be beneficially used for the removal of pollutants. It is by far more feasible to separate the adsorption/desorption step from the catalytic decomposition of the pollutants. For several pollutants, i.e. clopyralid, atrazine and bisphenol A, protocols for a reductive and oxidative decomposition were elaborated. For all tested pollutants a fast and complete decomposition could be achieved at least by one of the two ways of treatment.

Task 4.3: Development of MIP sensors
Various novel methods of surface-confined grafting for the preparation of molecularly imprinted polymeric ultra thin films onto sensor surfaces and their application as thin film coatings ( SPR type sensing for triazines ) and in fluorescence sensing are presented in this Task. This work has been carried out by CU using as target class of compounds triazines (as model template analogue for atrazine) and beta-estradiol.
E2 is one of the most common endocrine disruptor known with severe adverse effect on human and aquatic life. Therefore, it is chosen as one of the potential targets in WATERMIM project. To date, several methods have been proposed for the detection of E2 including gas chromatography, high performance liquid chromatography, immunoassays etc. However, those methods are somewhat high-cost and complicated. Compared with these methods, electrochemical method is fast, economic and simple. Therefore, the electrochemical determination of E2 attracts great interests. While, the detection limits are still not satisfied. Hence, researches for the development of more sensitive electrochemical sensors are still on.
In the present work, a bifunctional monomer was used for construction of electrochemical sensor for E2. Conducting films were prepared on the surface of electrodes (Au electrode) by electropolymerization of the aniline moiety of functional monomer. A layer of MIP was photochemically grafted over the polyaniline, via iniferter activation of the methacrylamide groups. Experimental parameters such as deposition time, cyclic volammetric (CV) scan cycles, and conditions for accumulation were optimized. The detection limit of this method was 6.86x10-7 M. Further, hybrid electrode was successfully used to analyze E2 in water samples without complex pre-treatment which demonstrated selectivity towards interferents which are alfa-estradiol, progesterone, estriols. These results reveal that hybrid MIP is an effective electrochemical technique to determine E2 real-time in complicated matrix.
Also, at CU Antibodies from the Flask Imprinted Dendrimer Core-Shell Nanoparticles were developed to be applied sensors. Antibodies, in common with other biomolecules, suffer from poor stability to extremes of temperature, pH and humidity. Mimicking the functional properties of antibodies in polymeric nanoparticles would be highly advantageous for both scientific and economic reasons. The present work describes an approach to the preparation of molecularly imprinted nanoparticles that have properties (affinity, size), similar to antibodies, using dendrimers as a graftable core. Surface amine groups of the dendrimer were modified with UV-labile groups to prepare a macroiniferter. Photo-initiated living polymerisation was performed to graft a molecularly imprinted polymer (MIP) shell. The synthesized nanoparticles were fractionated using affinity separation. Blank nanoparticles (control) were similarly prepared, but in the absence of the template.
Scanning electron microscope (SEM) observation showed that MIP and blank nanoparticles have similar shape, size (~200 nm) and composition. Dynamic light scattering measurements (DLS) and transmission electron microscopy (TEM) confirmed the narrow size distributions and the organic nature of the synthesized nanoparticles. For the first time cubic organic NPs are reported. Cubic NPs were an unexpected result and the origin of this morphology could be the subject of future study, however this shape may be an advantage where close-packing of NP-based material is required, such as in dense coatings. CU demonstrated very good specificity and selectivity of the synthesized material. Fluorescent-core cubic MIP NPs were prepared in an innovative concept. They demonstrated excellent selectivity and affinity after just 10 min incubation time. Advantages such as precise control of the number of the fluorescent labels per particle and its polymer shielding are reported for the first time as a new technique. NPs prepared in this way are promising materials to replace antibodies in sensors and immunoassays and in drug delivery and diagnostics. Triazine derivative was used for model. The limit of detection for acetoguanamine was calculated to be 3 × 10-8 M. Another area of interest for CU was the application of above synthesised denrimer core nanoparticles MIPs in making fluorescence sensor.
Also, at LNU a facile route to PEI films immobilized on gold has been developed. It was demonstrated that the surface coverage of the adsorbed polymers films can be tuned by varying the ionic strength of the adsorption buffer solution. Formation of PEI film with hyperbranched structures was confirmed with spectroscopic (RAIRS, ellipsometry) and microscopy (AFM) measurements. The polymer films thus prepared are capable of resisting the adsorption of the template molecule at 1 mg/ml concentration, and possessed anti-fouling even after a period of six months from production. Studies are underway to enhance the thickness and surface coverage of the PEI films to provide a better capture matrix. The adsorption of ferric and ferrous ions and humic acids is to be investigated in future work so as to establish the anti-fouling properties of the films in an even broader context. The present study reveals possibility of using PEI thin films, for generating anti-biofouling coatings with potential use in conjunction with the development of sensors. Preliminary studies on the entrapment of polymer nanoparticles under electrochemical polymerisation conditions indicates that this polymer system should prove useful as a generic platform for the rapid and cost-effective immobilisation of MIP-nano-particles in stable thin films on sensor surfaces in particular QCM.

Task 4.4 Monitoring and benchmark studies
Experimental on filtration/adsorption experiments by benchmark testing, in continuous or batchwise mode, and HPLC analysis realized at three WATERMIM Partners (e.g. CERTH, vTI and LNU). MIP technology was compared with other common adsorption materials, such as activated carbon and C18 column packing material. A comparison of the binding capacity for different analytes (mixture), with similar structure to the template used for the synthesis of MIPs, was also realized. Additionally, an effort was placed in order to verify the maximum binding capacity of activated carbon and to definitize the conclusions regarding the comparison of the two technologies (cost analysis). It was strongly suggested to still consider this as potential application, since the MIP technology was considered as more environmental friendly and easier at handling (because of the regeneration ability by the user himself). On the other hand, it was agreed that MIPs are comparable with other commercially available materials, such as C18, thus indicating the potential application of MIPs for the development of novel analytical tools.
It was agreed that the most promising materials physical formats are: Films developed by CU for environmental monitoring and sensing, Composite membranes embedded with nanoparticles developed by CERTH, USTUTT, MIPTech for separation, Grafted ceramic filters developed by CERTH, CU, Keranor for separation/filtration/monitoring and Composite nanofibers developed by ULUND for environmental monitoring and sensing.
It can be concluded that significant progress has been achieved through the WATERMIM project on the functional behaviour of MIPs in aqueous environment. Even if the developed MIPs have lower binding capacity as compared with activated carbon, their capacity is comparable with other commercial available materials, such as C18. This result, in combination with the fact the MIPs can also be regenerated gives to the delveloped technology within the WATERMIM project an extra advantage in comparison with traditional separation methods. The selectivity of the developed MIPs has been also tested in different ways. UDO has investigated MIPs performance with river and waste water using microextraction by packed sorbents in the Watermim project. In that test, a huge difference has been observed between MIPs and NIPs, thus proving the high selectivity of MIPs. Moreover, different compounds have been tested at the same time, showing a high specificity of the MIPs towards the selected compounds. The combination of the capacity and selectivity is essential for the use of MIPs in industrial environments. If the selectivity is extremely high, then the capacity is no longer a critical issue for process streams that contain a very low concentration of target species. Regarding the cost of the developed MIPs, further work for the estimation is needed. Also, a more thorough examination of the benefit from using MIPs related to the regeneration ability of the materials, which can be done without any product replacement but only with appropriate washing procedures, is necessary. Regarding the applicability, it should be noted that huge steps have been made within the project lifetime in order to bring the imprinting technology one step closer to the water treatment applications.

Potential Impact:
The main technological task of the WATERMIM project was the development of MIPs for selective recognition and separation of Synthetic Organic Compounds (SOCs) such as pesticides, pharmaceutically active compounds and endocrine disruptors. The ultimate goal was the development of new technologies for the remediation of the already contaminated water addressing the limitations of the traditional water purification techniques, such as ozonation, UV radiation, membrane filtration, and activated carbon adsorption. Indeed, many technical challenges and commercially attractive separation problems can not be solved with the presently available purification techniques, since they are non-selective and bring a secondary pollution. For this reason, the isolation and purification of high-value products and/or the efficient removal of critical contaminants from complex mixtures is of great importance. To this direction, innovative developments for the synthesis and preparation of molecularly imprinted materials have been focused on the optimization of membrane/filter morphologies in order to develop specialized, tailor-made products with high selectivity (i.e. below 0.1ppb) along with sufficient performance stability and low cost. Such novel membranes developed for separations are expected to find applications also in other fields such as analytics, screening and sensing systems.
After the realization of a series of experiments an Internal Critical Assessment was realized focusing on the Nano- and Micro-Particle and Membranes Technologies and the combination of the materials produced within WATERMIM project with the existed ones (i.e. application of ceramic or polymeric substrates) in order to develop novel systems for water treatment. The scaling-up, the regeneration, the cost analysis, the stabilization were critical issues. The cost, the simplicity, the applicability and the efficiency of the systems were estimated where possible for better evaluation of the results.
For the effective implementation of the exploitation of the WATERMIM project results, a Business Plan for the Commercialization of MIPs has been set up, after the valuable contribution of the Exploitation Strategy and Innovation Consultant and under the supervision of Project Officer. The WATERMIM consortium has also been motivated for the consideration of creating a spin-off company for MIP systems. To this direction, a critical examination of the project results was first realized. It was decided that a more proactive, focused approach would be to contact individual potential users in order to assess their specific interests in MIP. Thus, an effort has been placed to first identify the potential end users, based on the technology that the WATERMIM Partners have in their hands.
The main target was in all cases to emphasize on the fact that the new water-related environmental legislation in Europe clearly defines a need for regular monitoring of wide, and even growing, range of organic substances down to low nanogram per litre levels (2000/60/EC, 2000; 98/83/EC, 1998). Since many of the so-called priority substances (2000/60/EC, 2000) to be monitored in water are rather difficult to be analysed, WATERMIM as a consortium had put a considerable effort to propose a range of techniques and methods for the monitoring of the regulated compounds trying to contribute by these means in the implementation of the Water Framework Directive (WFD) (2000/60/EC, 2000).
It should be noted that huge steps have been made within the project lifetime in order to bring the imprinting technology one step closer to the water treatment applications. Even if it is quite difficult at the moment to apply directly MIPs in large scale water purification processes, it is foreseen that due to environmental reasons, new materials are needed for niche applications in water treatment industry, such as sensing and monitoring systems and a market in this area is thought to be already available for selective and cost effective materials, such as MIPs. Apart from water treatment industry, pharmaceutical and biomedical industry could also be supported by innovative diagnostic, analytical and separation tools based on imprinting technology. These tools could further form the basis for new standards which can support commercialization, since new water-related environmental legislation in Europe clearly defines the need for even stricter regulations. To this direction, WATERMIM has set the basis for the establishment of a new range of techniques and methods for the monitoring of the regulated compounds at very low levels and a number of applications are foreseen for the coming period utilizing procedures/formats, which are considered as the most promising exploitable results, and include:
Optimization of MIP composition by combinatorial screening (UDO)
Prototype of sandwich-type nanoparticles-based membranes (CERTH, USTUTT, MIPTech, UDO)
Prototype of nanoparticles-embedded membranes (USTUTT)
Immobilization of MIP nanoparticles using Click chemistry (ULUND)
Prototype of MIP films on/in ceramic substrates (KERANOR, CERTH, CU, UDO)
Grafting of polymers, over various surfaces with polyaniline-derivatives monomers (CU)
Immobilization of thin-films of MIP using novel iniferter and grafting approach (CU)
Prototype of nanofiber-based membranes for sensors (ULUND)
Prototype of composite MIP adsorbent/catalyst systems via various techniques (vTI, USTUTT, CERTH, MIPTech, Keranor)
MIP-based sensors for the detection of water pollutants (HIK/LNU)
Polymer beads for atrazine adsorption (MIPTech)
Dissemination of knowledge outside the consortium has been made via the participation in conferences focused on biosensors, nanotechnology, regulations for water, etc. Also, a specialized International Workshop devoted on the development of MIP technologies has been held having at the Organizing Committee and the Scientific Advisory Board of the Conference Prof. Sellergren, Prof. Nicholls and Prof. Ye, WATERMIM Project Steering Committee representatives. To this event the WATERMIM project had numerous representatives for the announcement of project results to other scientists working in the area of molecular imprinting. Moreover, non-specialized lectures to community central points (i.e. universities, school groups) has been realized so as to help students and the general public to understand the scientific principles and the importance of the combination of the molecular imprinting and membrane technology in today's society. Also, a specialized Workshop devoted on the Recent Developments of Nanotechnology-based Methods for Water Purification and detoxification, was realized in Greece and hosted by CERTH. To this event 15 EU funded projects had numerous representatives for the announcement of their project results to other scientists working in the area of water treatment.
Moreover, interviews at the Press Programme at the EuroNanoForum 2009 Conference on 2-3 June in Prague and for NanoTV project (content research stage of the FP7) conducted by the Institute of Nanotechnology (http://www.nano.org.uk) enabled the successful dissemination of project scope/results to specialised audiences.
Another target for WATERMIM project was to intrigue the enthusiasm of new students to explore the frontiers of materials science and to encourage young scientists not only to think about the potential of novel advanced materials, such as molecularly imprinted materials, as a means to help solve real problems, but also to work in this area. To this direction, several PhD theses related to the WATERMIM project have been initiated and 9 PhD students are expected to take their Diploma, on the following topics:
(CERTH): Synthesis of Hybrid Membranes for Water Purification
(ULUND): Building functional devices using molecularly imprinted nanoparticles
(CU): Development of MIP nanoparticles and nanolayers by living free-radical polymerisation
(HiK/LNU): Towards the Rational Design of Functional Materials
(HiK/LNU): Molecular Imprinting: Polymer Structure - Function Relationships
(USTUTT):Development of molecularly imprinted polymers for the use in aqueous media against endocrine disruptors EDC (Working title).
(USTUTT): Preparation and characterisation of biomimetically prepared molecularly imprinted particels for selective peptide adsorption
(USTUTT): Preparation of biofunctionalised nanoparticle layers
(vTI): Catalytic decomposition of pollutions

Also, the web site that has been created to provide useful information to the general audience. More specifically, the website informed about the scientific and technological objectives of the project as well as about its partners and could be used as a tool for contacts with companies/universities outside the consortium. (http://lpre.cperi.certh.gr/watermim/)
Finally, dissemination of knowledge has been fulfilled by submitting papers to scientific journals. Selected papers are:
Cubic Molecularly Imprinted Polymer Nanoparticles with a Fluorescent Core, Petya K. Ivanova-Mitseva, Antonio Guerreiro, Elena V. Piletska, Michael. J. Witcombe Zhaoxia Zhou, Petar A. Mitsev,. Frank Davis, Sergey A. Piletsky, Angew Chem,International Edition, 51, 2012, 1-5
Model study on the aqueous-phase hydrodechlorination of clopyralid on noble metal catalysts, Ulf Prüße, Catalysis Communications, No 14, Elsevier, 2011, 96 - 100
Clickable molecularly imprinted nanoparticles, Ye et al., Chem. Commun., 47, 2011, 6096-6098
Molecular imprinting in Pickering emulsions: a new insight into molecular recognition in water, Ye et al., Chem. Commun., 47, 2011, 10359-10361
Molecularly imprinted magnetic materials prepared from modular and clickable nanoparticles, Ye et al., J. Mater. Chem., 22, 2012, 7427-7433
Imprinted polymer beads enabling direct and selective molecular separation in water, Ye et al., Soft Matt., DOI:10.1039/C2SM25574J
Mechanisms underlying molecularly imprinted polymer molecular memory and the role of crosslinker: resolving debate on the nature of template recognition in phenylalanine anilide imprinted polymers, G.D. Olsson et al, Journal of Molecular Recognition, 25, 2012, 69-73
An improved grafting technique for producing imprinted thin film composite beads, Mahadeo R. Halhalli, Carla S. A. Aureliano Eric Schillinger, Claudia Sulitzky, M. Magdalena Titirici and Börje Sellergren, Polym. Chem., 3, 2012, 1033-1042
Selective determination of estrogenic compounds in water by microextraction by packed sorbents and a molecularly imprinted polymer coupled with large volume injection-in-port-derivatization gas chromatography-mass spectrometry, Prieto A, Vallejo A, Zuloaga O, Paschke A, Sellergen B, Schillinger E, Schrader S, Möder M, Anal. Chim. Acta, 703, 2012, 41-45.
Also, two patent applications have been submitted within the project lifetime by Prof. Lei Ye (ULUND) and Prof. Sellergren (UDO).
Finally, an extra effort has been placed on the continuous and effective cooperation between the coordinators of the projects funded under the call: NMP-2008-1.2-2 Nanotechnologies for water treatment. The aim of this action was to support research and technological development in the field of water treatment gained within each project, by applying the developed nano-engineered materials to promising separation, purification and detoxification technologies. The consortium actively contributed to the information exchange between themselves and other consortia resulted from above mentioned call. Also, a 2nd Joint Dissemination Workshop of the Nano4water Cluster was organized by CERTH on 24-25 April 2012, in Greece. The website (http://www.nano4water.eu) of the nano4water cluster that has been set up as an electronic tool for the direct interaction among the projects is updated and includes the ppt presentations of the workshop after the written permission of the participants. The outcome of this workshop was further presented by the WATERMIM Project Coordinator, Prof. Kiparissides, in June 2012, in Aarhus at Industrial Technologies 2012. The WATERMIM consortium has also agreed with the publication of a special issue in Journal of Industrial and Engineering Chemistry including selected papers from the 2nd Joint Dissemination Workshop of the Nano4water Cluster.

List of Websites:
http://lpre.cperi.certh.gr/watermim/

Final Report Summary - WATERMIM (Water Treatment by Molecularly Imprinted Materials)

Udostępnij tę stronę

Pobierz