One in a billion

Selective separation is an important requirement in many areas of chemistry and biology. Molecular imprinting is a technique that yields selective sorbents and aspires to mimic the extreme selectivity achieved by many biological systems [1]. The technique involves synthesis of a highly cross-linked polymer in presence of the chosen metal ion or a molecule (or a suitable analogue of the molecule) as a template, and functional monomers that can complex with the template. The geometry of the complex formed is fixed within the polymer matrix by the cross-linkers that polymerize around the complex. The template is removed after polymerization, thereby yielding a polymer with a preset configuration that can selectively accommodate that particular molecule or metal ion. The imprinting technique has been used extensively to achieve selectivity among molecules, metal ions and also in selective catalysis.

While synthesizing metal-imprinted polymers, two approaches (Fig.1) are followed. In the first method (A), a mixture of ligands containing polymerisable groups (functional monomers), metal salt, and the cross-linker are taken together and subjected to polymerization after allowing sufficient time for complexation of the metal ion with the complexing monomers. The polymer formed is then treated with acid for metal ion removal to get the metal ion imprinted polymer. The second method (B) involves two steps. In the first step, the metal ion is complexed with the ligand functional monomer and the complex is isolated and purified. The purified complex is then subjected to crosslinking polymerization to get the polymer, which on metal ion removal gives the metal ion imprinted polymer. The second method ensures a single characterized complex is used which can ensure uniformity of the sites formed. However, such a two-step method is not necessary in all cases, especially when the metal ion has high complexing ability with the ligand chosen and forms predominantly one type of complex.

Generally, unlike the conventional ion exchange resins, high amounts of cross-linking agents, as much as 90 to 95 mol%, are used in synthesizing imprinted polymers. This leads to reduced capacity due to low amounts of functional monomers present, but increased selectivity due to the rigidity provided. The success of a metal ion imprinted polymer depends very much on the functional monomers (ligands) chosen, which should form a complex stable enough to retain the geometry but labile enough to facilitate removal of the template.

Nuclear applications

Selectivity is an important aspect in effective handling of radioactive waste. Extensive amount of work has been done on selective ion exchangers for nuclear applications. Unlike ion exchangers, imprinted polymers can offer the very high selectivity required in nuclear applications. Hence, attempts, though very limited due to the narrow field of application, have been made to synthesize metal ion-imprinted polymers targeted towards nuclear applications. One such early attempt was an imprinted polymer for selective separation of uranyl (U02)2+ ions by Bae et al [2]. Here, vinyl benzoic acid was used as the complexing functional monomer. The resultant polymer was shown to be suitable for quantitative pre-concentration of uranyl ion from a complex matrix, such as seawater. At a pH of 3.0, 100% recovery was obtained. Saunders et al [3] reported a uranyl imprinted polymer which showed selectivity for uranyl ion against Fe (III), Cu (II) and Th (IV) ions. They have used 2-chloroacrylic acid as the functional monomer for complexing the uranyl ion and ethylene glycol dimethacrylate (EGDMA) as the crosslinking agent. Preetha et al [4] detailed a uranyl imprinted polymer synthesized using mixture of functional monomers, namely salicylaldoxime (SALO) and vinyl pyridine. The work demonstrated the feasibility of using the uranyl imprinted polymer to remove uranyl ions selectively in the range 5µg - 300 mg present in 500 mL of synthetic nuclear power reactor effluent containing a host of other inorganic species. Also, lanthanide separation by ion chromatography using imprinted resins has been attempted [5]. A list of reports on imprinted polymers with relevance to nuclear applications is given in Table 1.

An area of significance in nuclear industry is effective handling of radioactive waste generated during the periodic clean-up operations of the plant: decontamination. Recently, we have reported an imprinted polymer capable of reducing the volume of active waste generated during such decontaminations by selective sorption of cobalt under typical decontamination conditions [6]. During decontamination, complexing reagents are circulated through the system in solution form to bring out the corrosion products, and thereby release and trap the active ions. Typically, this process leads to large excess of ferrous ions which are predominantly non-radioactive along with a very small amount, at sub-ppb levels, of radioactive cobaltous ions. These are trapped by the ion exchange resins. It is very well known that the ferrous and cobaltous ions have similar chemical behaviour at a macro scale and largely react in an identical manner to the conventional ion exchange resins.

Both are trapped by the cation exchange resins, leading to a spread of radioactivity throughout the whole volume of the resin used for trapping the metal ions. The quantity of resin used for this operation is decided by the amount of iron, and it is significant. Hence, with a resin that could trap only cobaltous ions, the volume of radioactive resin required could be minimized by trapping all the cobaltous ions using a small molecularly imprinted polymer column, leaving excess ferrous ions for the regular ion exchange column. This is important because cobalt activity is of great concern in nuclear industry because of its high gamma energy. It is with this objective that the imprinted polymer was synthesized.

Cobalt-imprinted polymer synthesis

The polymer was synthesised by polymerising vinylbenzyl iminodiacetic acid complexed to cobaltous ions in presence of excess crosslinking agent, namely ethylene glycol dimethacrylate. The scheme is shown in Fig. 2. The polymer was synthesized through both methods shown in Fig. 1 and was found to exhibit similar properties. Sorption by the imprinted polymer was tested under typical decontamination conditions using active cobalt present in a mixture of complexing agents (NTA, ascorbic acid and citric acid) with a large excess of ferrous ions. This resulted in a capacity of 44.0 µCi/g (± 0.7) for the cobaltous ions (initial solution activity: 2 µCi) and exclusion of ferrous ions (initial concentration: 4mM). The imprinted polymer also showed selectivity against Cr (III) and Ni (II) under complexing conditions. This polymer was synthesized essentially with an objective of solving the problem of handling excess waste associated with the dilute chemical decontamination in PHWRs. However, the universality of the problem of dealing with radioactive cobalt and the polymer’s selectivity for cobalt over other transition metal ions under complexing conditions widen its appeal.

Although there have been many theoretical studies on molecularly-imprinted polymers, there are not many reports on detailed theoretical investigations specifically for metal ion-imprinted polymers. Shamsipur et al [7] used ab-initio calculations to deduce the stable configuration of the complex formed by copper ions with the free functional ligand that is to be used in the synthesis of copper imprinted polymer. We have used theoretical studies to explore the selectivity shown by the cobalt imprinted polymer synthesized. Though ferrous and cobaltous ions behave in similar fashion, the polymer excludes ferrous ions. This was investigated by calculating the formation energies of the complexes formed in isolation and within the polymer matrix, that is, when the complex is attached to the cross-linkers [8]. These calculations have been performed using the ab-initio density functional theory code SIESTA (Spanish Initiative for Electronic Simulations with Thousands of Atoms). The calculated values were found to be 0.078eV for Fe2+ and -0.477eV for Co2+ complex with the functional monomers attached to the cross linkers. The formation energy values indicated that the iron complex is destabilized in presence of the crosslinking units attached to the functional ligand, whereas the cobaltous ion complexes are not. This prevents the iron leaving the complexing solution to complex with the polymeric sorbent. With these initial theoretical studies as the basis, further theoretical studies exploring different competing conditions in the solution vis-a-vis polymeric sorbent are to be performed as well. Through prudent use of such calculations, eventually one should be able to accurately predict suitable monomers for a particular application through these calculations. This is a long-term objective with respect to the theoretical studies.

Molecularly- and metal ion-imprinted polymers have started seeing the light of commercialization thus far, mainly for small scale analytical applications. Nevertheless, the robustness of these polymers together with their low cost are attractive not only for analytical scale separations but also for preparative or process scale separations. This can concern removal of toxic or unwanted byproducts from process streams in the food industry, wastewater treatment, trace heavy metal pollutant removal from water, or as outlined here, radioactive waste handling. However, when it comes to applications such as in the nuclear industry, where large volumes have to be handled, an important task will be to produce the polymers in formats compatible with high flow conditions. Laboratory studies indicating a fast mass transfer to these beads is promising in this regard. Provided that the required capacity is not very high, and that scalable production methods are available, large-scale applications of imprinted polymers can be expected in the near future. Imprinting technology is therefore likely to play an important role in the present century’s separation science.

Anupkumar Bhaskarapillai, Narasimhan V Sevilimedu, Water and Steam Chemistry Division, Bhaba Atomic Research Centre Facilities, Kalpakkam – 603102, India. Börje Sellergren, INFU, Technische Universität Dortmund, 44221 Dortmund, Germany