Microprobe investigation of molecularly imprinted polymers

27 June 2013
Ranganathaiah Chikkakuntappa and Tenzin Pasang
A novel technique that measures tiny pores can improve physical selectivity in lock-and-key polymer applications.

What do a front door, the human immune system, a dog's nose, and molecularly imprinted polymers (MIPs) have in common? They all use a lock-and-key system to identify and select specific agents from numerous possibilities. Much like an antibody that binds only to a complementary antigen, MIPs act as artificial locks for specific molecules. The polymer receptor is designed to grab only ‘key’ chemicals. Because MIPs are artificially engineered, design freedom is expansive: keyholes are not limited to proteins, or even to naturally occurring functional groups.

MIPs have inspired a great deal of recent research because of their wide variety of applications.1 These include biosensors, solid-phase extraction, chromatography, and drug delivery systems. The molecular imprinting technique leaves cavities in the polymer matrix with affinity for a chosen ‘template’ molecule. The process usually involves initiating the polymerization of monomers in the presence of a template molecule that is extracted afterwards, leaving complementary cavities behind. Usually a high percentage of crosslinker (~50%) is incorporated to avoid strain in the polymer matrix.2 The extraction of template molecules from the polymer matrix leaves behind a cavity. The cavity corresponds to the extracted template molecule both physically (size and shape) and chemically (functional binding).

In all applications, selectivity of the imprinted polymer is critical. However, foreign molecules with functional groups similar to the template molecule can cause nonselective binding. Further complicating MIP selectivity, polymer microstructure has its own native cavities. These pores can act as non-selective binding sites for template molecules if they are of the right size. This possibility must be addressed if the MIP's physical selectivity is to be precise.3 Unfortunately, properties that increase selectivity can be at odds with properties that are otherwise desirable. A highly flexible polymer with accessible cavities allows fast release and uptake of template molecules, which is ideal for many applications, but the same properties increase the likelihood of non-selective binding.

Current methods of measuring polymer porosity are based on gas adsorption, such as the Brunauer-Emmett-Teller (BET) and Barrett-Joyner-Halenda (BJH) techniques. BET and BJH use adsorption-desorption isotherms of nitrogen molecules at critical pressure and liquefaction temperature (77K). These methods are limited by hysteresis between the isotherms' adsorption and desorption branches, and pore size calculations can differ considerably for the same kind of pores. These methods provide very little information about the micropores (<2nm) in MIPs in which small molecules are imprinted. We propose a novel method of positron annihilation spectroscopy to increase physical selectivity of MIPs through careful determination of the polymer porosity.

To demonstrate the method's utility, we have chosen to prepare MIPs with ethylene glycol dimethacrylate (EGDMA) as the crosslinking agent and methacrylic acid (MAA) as the monomer. This material has been investigated by several researchers in the past.2–4 We prepared three types of samples: non-imprinted polymers (NIPs), samples that included NIP and the template molecules (NIP+template), and MIP samples in which the template has been removed. In each set we prepared polymers with molar ratios of 1:2, 1:5, and 1:9 of the monomer MAA to the crosslinker EGDMA. We chose 4-chlorophenol (4-CP) as the template molecule.

Fourier transform IR spectroscopy (FTIR) confirmed that the template molecule imprinted successfully (see Figure 1). ‘NIP5’ indicates an NIP sample with a 5:1 ratio of crosslinker to monomer. The extra peaks in the FTIR scan of the NIP5+template sample (marked by lines at 648 and 825) indicate the presence of the template. The disappearance of these peaks in the MIP5 sample confirms the formation of a MIP with the successful extraction of template molecules.

Fourier transform IR spectroscopy (FTIR) scan of a non-imprinted polymer with a 5:1 ratio of crosslinker to monomer (NIP5) in red, NIP5+template molecules in green, and a molecularly imprinted polymer with the same crosslinker/monomer ratio (MIP5) in blue.

Positron lifetime measurements are usually employed to measure the types and densities of defects in solids. A positron is a subatomic particle with the same mass as an electron, but a positive charge. To make a positron lifetime measurement, positrons are injected into a solid material and the lifetimes of these pair-annihilating particles are measured. Because electron densities are lower near atomic voids or defects, positron lifetimes are longer in these areas and the particles accumulate. We conducted positron lifetime measurements on our polymer samples using a standard fast-fast coincidence system with two BaF2 scintillation detectors. (A coincidence system measures the time delay between two events or two different events within a certain time interval. ‘Fast-fast’ refers specifically to the two detectors we used.) The lifetime spectrometer had a time resolution of 220ps. All measurements were made at room temperature with source-sample sandwich geometry.

The term ‘free volume’ refers to space created by the movements of polymer chains. Free volume cavities occur due to chain folding and the molecular architecture of the segments.5 Consequently, these cavities evolve to different sizes. We used a computer program called PATFIT-88 to extract lifetime data from the positron annihilation lifetime spectra (PLS). PATFIT-88 analysis of the measured PLS provides an average cavity size assuming a spherical shape. To get the distribution of cavity sizes from the same measured spectrum, we used another computer program called CONTIN-PLS2.5 CONTIN-PLS2 generates positron annihilation rate probabilities from which we can determine cavity radius and the volumetric size distribution. The PLS results clearly illustrate changes due to different monomer and crosslinker molar ratios in the MIP.

We observed that as the ratio of crosslinker increases in NIP samples, the fractional free volume (the free volume content as a percentage) in the matrix decreases. As the template is extracted from the NIP+template sample, we observed a strange decrease in free volume for MIP samples with molar ratio 1:9, whereas in MIP samples with ratios of 1:2 and 1:5 we observed an increase in the fractional free volume, as expected. The decrease in fractional free volume for molar ratio 1:9 may be due to excess crosslinker, which is not desirable. The changes in observed fractional free volume should be due to cavities left behind by the template. Fractional free volume measurements after template removal from the NIP5+template sample was noticeably high. Therefore, we chose 1:5 as the optimum molar ratio for this particular MIP.

The free volume radius distribution f(R) obtained from the CONTIN-PLS 2 program in NIP5, NIP5+template, and MIP5 samples are shown in Figure 2. As is evident by the curves, for NIP5 we find that a large fraction of the cavities are larger than the template molecules (4-CP radius 2.99Å). These large cavities are inherent in the polymer microstructure. When the template is incorporated into the NIP, a few of these molecules may get into the free native pores and others will bind to sites through hydrogen bonding. Therefore, after imprinting, the cavity number is a combination of binding sites (removal of 4-CP from hydrogen-bond-formed sites) and non-binding sites (native pores of the size that accommodated template molecules). Since physical selectivity estimation is defined by the size and shape of the pores created in MIP5, this information must be taken into account. As demonstrated in Figure 2, the distribution curves for NIP5 and MIP5 are not identical. Each has a distinct height, width, and an under-curve area that represents the probability of observing such cavities. These curves illustrate the probable distribution function of different pore sizes. We used these differences along with other parameters to determine the physical selectivity, which was found to be about 85%.

The free volume radius distribution f(R) in NIP5, NIP5+template, and MIP5 samples. 4-CP: Template molecule 4-chlorophenol.

In summary, this work shows that PLS is a valuable and sensitive tool to study the molecular imprinting process. It also shows that lifetime spectrum analysis of free volume size distribution is better and more accurate than previously employed techniques such as BET and BJH. These methods cannot ascertain the physical selectivity efficiency due to the limitations discussed earlier. The distribution works well to describe the actual polymer microstructure, whereas average positron lifetime results fail to illustrate the pore structure responsible for the selectivity of the MIPs. The physical selectivity of MIPs depends on the size and shape of micropores. We demonstrated that PLS is a versatile tool in investigating precise physical selectivity properties of MIPs. Further studies should investigate other types and sizes of template molecules for physical selectivity of MIPs in real-time sensors.


Ranganathaiah Chikkakuntappa
Department of Studies in Physics University of Mysore

Ranganathaiah Chikkakuntappa is a professor. His areas of research include nuclear physics and positron annihilation spectroscopy studies in polymers and polymer-related materials.

Tenzin Pasang
Department of Studies in Physics University of Mysore

Tenzin Pasang is a PhD candidate working under the supervision of Ranganathaiah Chikkakuntappa.


  1. C. Alexandrer, H. S. Adersson, L. I. Adersson, R. J. Ansell, N. Kirsch, I. A. Nicholls, J. O'Mahony and M. J. Whitechomb, Molecular imprinting science and technology: a survey of the literature for the years up to and including 2003, J. Mol. Recognit. 19, pp. 106-180, 2006.

  2. A. D. Vaugahan, S. P. Sizemore and M. E. Byrne, Enhancing therapeutic loading and delaying transport via molecular imprinting and living/controlled polymerization, Polymer 48, pp. 74-81, 2007.

  3. Molecularly Imprinted Polymers Man-Made Mimics of Antibodies and Their Applications in Analytical Chemistry 2nd ed. ed., Amsterdam, Elsevier, 2003. ch. 2

  4. N. Holland, J. Frisby, E. Owens, H. Hughes, P. Dugga and P. McLoughlin, The influence of polymer morphology on the performance of molecularly imprinted polymers, Polymer 51, pp. 1578-1584, 2010.

  5. P. Ramya, C. Ranganathaiah and J. F. Williams, Experimental determination of interface widths in binary polymer blends from free volume measurements, Polymer 53, pp. 4539-4546, 2012.

DOI:  10.2417/spepro.004920