Introduction Engineered nanoparticles (ENPs) have found widespread use in modern technology, medicine and daily life. Ag-NPs for example are used as antimicrobials in several medical and cosmetic products, whereas Au-NPs have attracted great interest in medical applications, such as drug delivery systems. However, ENPs can potentially cause adverse toxic effects due to their small size and high surface to volume ratio and their increased utilization in different medical and consumer products leads to an increased exposure to humans. While a range of state-of-the-art technologies, such as ICP-MS, is available for the detection of ENPs, they usually require thorough sample preparation, highly trained personnel and expensive equipment. [1]To circumvent these limitations, the use of artificial recognition materials, as for instance molecularly imprinted polymers (MIPs), is a promising alternative. MIPs are synthesized via (radical) polymerization in the presence of template molecules. After template removal, cavities, which are complementary to the template in size, shape and chemical functionality, form in the material. These can rebind target molecules with high sensitivity and selectivity. [2] MIPs can be used as receptors for quartz-crystal microbalances (QCMs), which are mass-sensitive transducers that measure a change in mass via a change in resonant frequency of the oscillating quartz. [3]In this work, we report on developing an easy-to-use and low-cost MIP-based QCM sensor for detection of ENPs in aqueous matrices. Method Stamps with immobilized metal-based NPs were used to imprint template material on PU layers. Particle distribution and shape of NPs, immobilized on PDMS stamps, were analyzed using atomic force microscopy (AFM). MIPs were prepared by pressing NP-stamps onto PU layers. After polymerization overnight, stamps were removed and remaining NPs were washed off the surface. MIPs were characterized via AFM and QCM. In case of QCM, we utilized an in-house dual-electrode system. In that case, one of the two electrode pairs served as reference and was coated with the non-imprinted polymer (NIP) to compensate for external effects, such as temperature changes. Results and Discussion Two different imprinting strategies were tested for fabrication of NP MIPs: bulk and surface imprinting. Bulk imprinting did not result in cavity formation on the polymer surface as examined via AFM. Consequently, we could not observe any significant differences of sensor signals between MIP and NIP during QCM measurements. In contrast to this, surface imprinting led to more favorable outcomes. NP-stamps were used for that purpose and characterized using AFM to verify the presence of NPs on stamp surfaces. As shown in Figure 1a, NPs are distributed evenly across the surface. Figure 1b displays an AFM image of the corresponding imprinted PU layer, which clearly reveals cavities on the polymer surface, matching the template in size and shape, thus confirming successful NP imprinting. In addition, an image of the reference, the non-imprinted polymer (NIP), is given for comparison, bearing no binding sites on the polymer surface (Figure 1c).Frequency measurements of MIP and NIP coated QCMs resulted in reversible and concentration-dependent sensor responses with no cross-selectivity to larger NPs. Furthermore, frequency shifts of MIPs, caused upon injection of NP solution, were much higher than those of NIPs. Fist results on acrylate-based MIPs using self-initiating monomers seem promising. First QCM measurements revealed incorporation of NPs by MIPs, with a way higher sensor response compared to the NIP. However, there is still room for improvement regarding batch-to-batch reproducibility. Further characterization of these sensing materials is necessary and still ongoing. Conclusion and Outlook Imprinting of thin polymer-films using metal-based ENPs is a novel concept with considerable potential in the field of sensor technology. PU-based MIPs were prepared by surface imprinting and characterized using AFM, revealing corresponding NP cavities on the polymer surface and thus confirming successful imprinting of NPs. Moreover, frequency measurements of MIP coated QCMs led to sensor responses in a reversible and quantitative manner. In the near future, selectivity studies will be performed in order to assess binding responses as a function of NP size, stabilizer and core material. Moreover, surface morphology of MIPs will be further characterized using PeakForce QNM.Overall, this work demonstrates the potential use of MIPs for analysis of NPs in complex matrices, such as food, cosmetic or medical products. Acknowledgements This work has been funded by the Austrian Science Fund FWF, project number I 3568-N28, which we gratefully acknowledge.
Read full abstract