Ontwikkeling van selectieve coatings voor optische sensoren

The sustainment of the environmental quality has over the last decade shifted into the centre of political, legislative and economic interest.In order to be able to take measures, the mapping of pollution problemsneeds to be taken care off in a precise, reliable and cost-effective way. Furthermore, monitoring is needed during the remediation, as a follow-up and as prevention. Therefore, the need for reliable in-situ monitoring methods arises. Also food safety and the efficient cheap monitoring of the potential harmfull components have become more and more important. The availability of appropriate chemical sensors would facilitate the screening of these pollutants or food components enormously. Within the upcoming field of chemical sensor technology, it is especially optical sensors that are gaining a lot of interest for environmental monitoring and food safety, as they are robust and inert against for instance electromagnetic interference. Optical sensing based on surface plasmon resonance (SPR), is widely used for studying bio-molecular interactions and bio-sensing. SPR is an optical technique that probes changes of refractive index induced by changes of mass within the near vicinity of the surface. SPR based platforms are also gaining more interest for the detection of chemicals such as volatile organic compounds or hydrocarbons.This study explores the potential of an innovative fibre optic SPR platform in combination with 3-D structured recognition layers to detect small molecules.Hereto different 3-D structures to create the selective layer were evaluated making use of different model systems. A first system evaluated was a zeolite based layer. The target analytes are hydrocarbons and selectivity is determined via the molecular sieve properties of the zeolite. A second system was a molecular imprinted polymer (MIP), a polymer with a molecular memory for a molecule present during synthesis, based on shape and interactions between functional groups of template and monomer. Such a MIP structure is more versatile than a zeolite structure, as it is tailor-made and can thus be used for a wide range ofmolecules. A MIP for the detection of six ergot alkaloids, small molecules around 400-500 g/mol, was used here as a model compound. Different ways of attaching the MIPs to the fibre optic sensor were explored: (i) direct attaching of the MIP beads, (ii) embedding of beadsin a polymeric matrix, and, (iii) direct coating with an imprinted polymeric layer. It was demonstrated that the third procedure resulted in the most sensitive sensor. Further, theinfluence of the different measuring environments (aqueous or solvent medium or the presence of barley matrix compounds) was evaluated. Finally, polymeric layers were evaluated for their ability to host the MIP beads. Porous polymer-based membranes were evaluated as a tool to prepare samples before actual sensing. When a membrane with an appropriate selectivityis used as a sample preparation method, compounds that woul interfere with the sensor detection could be already removed in anticipation from the matrix via the membrane filtration. In order to be able to separate compounds with only a small difference in molecular weight, a sharp molecular weight cut-off (MWCO) is neededHereto, the role of different membrane synthesisparameters was evaluated to manipulate the MWCO of a polysulfone (PSf) nanofiltration membrane. As no consensus has been reached in literature yet about a uniform solute/solvent system to determine the MWCO-curves for solvent filtrations, different systems were first proposed to correctly evaluate the PSf-membranes developed for these sensor pre-treatment purposes.The direct MIP coated FO-SPR has the potential to be used as a sensing system. However, improvements to sensitivity and applicability in a wide range of measurement environments still need to be done. The evaluation of the different phase inversion membrane synthesis parameters clearly indicated that the polymer properties (chain length, purity, MW, polydispersity) influence the membrane performance. Further, no uniform 'solute/solvent' system could be determined for SRNF membrane separations, as the influence of the hydrophobic/hydrophillic nature of solutes, solvents and membrane cannot be neglected.status: publishe