The molecular imprinting effect of propranolol and dibenzylamine as model templates: Binding strength and selectivity.

Recent studies have shown anomalies with the most studied non-covalent molecularly imprinted polymer, the propranolol imprinted one. This imprinted polymer, like many others, binds more template than the non-imprinted control polymer, but its selectivity in template adsorption is only slightly or not at all improved by imprinting, depending on the compound compared. The reasons for this anomaly are discovered here. Simple experiments show that acid homoassociation in the prepolymerisation complex is the likely cause of the anomaly. The specific conductivity of prepolymerization mixtures at different functional monomer to template ratios follows a pattern observed in homoassociating systems. Analysis of the optimal prepolymerization mixture shows that on average two molecules of the functional monomer are complexed to the basic template, even if the template lacks any other hydrogen bonding functional group than the amino group. Molecular modeling calculations provide the structure and stability of the homoassociated prepolymerization complexes. These results lead to a plausible interpretation of the anomaly, which may not be unique for the propranolol imprinted polymer, but may affect all imprinted polymers made for basic templates by using acidic functional monomers. The analytical applications of the new imprinting model are demonstrated.

The Selectivity of Polymers Imprinted with Amines.

One of the main reasons for making molecularly imprinted polymers (MIPs) has been that MIPs interact selectively with a specific target compound. This claim is investigated here with the example of a widely used type of noncovalent MIP, the MIP for the beta blocker propranolol. Adsorption isotherms of this MIP and of a nonimprinted control polymer (NIP), respectively, have been measured with a series of compounds in the porogen solvent acetonitrile. The results, visualized as "selectivity ladders", show that the MIP binds propranolol and many other amines better than the NIP does, but the selectivity of the MIP is actually inferior to that of the NIP. The selectivity of either polymer for propranolol is modest against many amines, but is remarkable with respect to other compounds. The contribution of imprinting towards selectivity can be better appreciated when three MIPs, made with different amine templates, are compared among themselves. Each MIP is seen to bind its own template slightly better than the other two MIPs do. In media different from the porogen, the selectivity patterns may change substantially. Propranolol seems to have properties that make it stand high on the selectivity scale in different solvents, albeit for different reasons.

New Methods to Study the Behavior of Molecularly Imprinted Polymers in Aprotic Solvents.

This work presents three new experimental methods for studying molecular imprinting. The electric conductivity measurements of the pre-polymerization mixture of amine templates in an aprotic solvent provide evidence of ionic dissociation of the pre-polymerization complexes. The displacement measurement of the template propranolol from its molecularly imprinted polymer (MIP) using a quaternary ammonium ion in toluene, shows that this MIP behaves as an ion exchanger even in a non-polar solvent. The same experiment also shows that template binding to the MIP from toluene involves ionic interaction. The third experimental method introduced here serves to study the models of template binding on MIPs. To this end the binding isotherm of propranolol (PR) has been measured on a polymer mixture consisting of non-imprinted control polymer (NIP) and a stronger binding acidic polymer, respectively. All three methods are suitable for studying several other imprinting systems.

Binding capacity of molecularly imprinted polymers and their nonimprinted analogs.

Molecularly imprinted polymers bind their target compounds at binding sites. The binding sites are typically based on some type of functional group, such as carboxyl group. The total amount of such functional groups and their distribution into available and unavailable groups is not well known. The total binding capacity is usually indirectly determined from adsorption isotherms, which are measured much below the theoretical binding capacity. This work shows that in a variety of differently prepared, methacrylic acid based molecularly imprinted and nonimprinted polymers, all carboxylic groups used for the polymer synthesis are retained in the polymer, 80-90% of them can be accessed by strong bases and essentially the same amount can be used for adsorption of weak bases. This high level of adsorption can only be achieved, however, if the adsorbed weak base is strong enough, if the polymer is sufficiently elastic and if the solvent does not compete too strongly for the binding sites. These results may explain why the maximum binding capacities obtained from isotherm measurements are usually not equal to the total amount of available binding sites. This study confirms the usefulness of nonimprinted polymers at high loadings.