ABSTRACT

Figure 4.1 Schematic representation for the generation of MIPs: (a) Complex formation either by noncovalent interactions or by covalent bonds; (b) Copolymerization; (c) Removal of the template; and (d) Template rebinding. Depending on the interactions between the template and

functional monomers involved in the imprinting and rebinding

steps, molecular imprinting has been realized in three different ways through the covalent, noncovalent, and semicovalent approaches. The covalent approach was pioneered by Wulff’s group [6], where the functional monomers form complexes with template molecules via reversible covalent bonds (such as boronate ester, ketal and acetal, or Schiff base) prior to the polymerization, and the subsequent rebinding of the templates to the imprinted polymers also takes place through the formation of covalent bonds between them. The noncovalent approach was introduced by Mosbach and coworkers [3, 8], which utilizes only noncovalent interactions (such as hydrogen bonds, ionic interactions, hydrophobic interactions, and metal-ion chelating interactions) for both the molecular imprinting process and the subsequent template rebinding. The semicovalent approach (also called hybrid approach) was developed by Whitcombe et al. [5], where a covalently attached template was utilized in the imprinting process while the template rebinding step is noncovalent. In comparison, the noncovalent approach is more flexible in terms of the MIP preparation due to the absence of complicated synthetic chemistry and the broad selection of functional monomers and possible target molecules available; it has thus been more commonly used nowadays. In addition, the imprinted polymers prepared by the noncovalent imprinting approach show much faster rebinding kinetics than those prepared by the covalent approach, which makes them particularly suitable for applications involving their use as stationary phases in HPLC system and as sensors. However, it is worth mentioning that the covalent approach usually produces a more homogeneous population of binding sites than the noncovalent approach does because of the greater stability of the covalent bonds, as confirmed by Shimizu and coworkers [9, 10]. Since the success of the molecular imprinting process is very much dependent on the monomer-template interaction, the suitable choice of functional monomers is the most important characteristic in targeting a template molecule. So far, many types of functional monomers have been studied, and Fig. 4.2 presents some of the functional monomers used in the noncovalent molecular imprinting process. Among them, methacrylic acid has been the most widely used one. It can interact with amines via ionic interaction and with amides, carbamates, and carboxyls via hydrogen bonds. For achieving stronger ionic interaction, 2-/4-vinylpyridine is normally used for a template with carboxyl functionality [11-13]. Since the noncovalent

interactions between the functional monomers and templates are usually too weak for stable complexes to be formed, an excess of functional monomers needs to be added to shift the equilibrium toward the complex formation. Trifluoromethylacrylic acid [14] and acrylamide [15] have proven to be capable of forming more stable complexes with templates than methacrylic acid does. In addition, certain functional monomers, such as those containing the amidine group (e.g., N,N’-diethyl-4-vinylbenzamidine) [16] or guanidine group (e.g., 9-(guanidinomethyl)-10-vinylanthracene) [17], can even form very stable complexes with templates containing carboxyl groups in a stoichiometric ratio. The combined use of vinylpyridine and methacrylic acid has been found to be particularly useful for the preparation of MIPs against carboxylic acids, resulting in improved recognition capabilities as compared with MIPs prepared by use of only one of the functional monomers [13].