Performance modelling of zeolite-based potentiometric sensors

Developing simple and cost-effective electrochemical sensors for widespread on-site application is of considerable practical importance in agriculture, environmental monitoring and food science. Among multiple sensing platforms, potentiometry is particularly effective in terms of simplicity, low cost and mass-production. This work is focused on a systematic analysis of the structure - performance relationship using chemometric techniques, which can be applied to sensor arrays with varying response patterns. The potentiometric sensitivity of zeolitemodified electrodes, containing thirteen synthetic and three natural zeolites, in aqueous solutions of Na+, K+, NH4+, Ca2+ and Mg2+ has been correlated with a range of zeolite characteristics using PCA and PLS modelling, thus demonstrating how structural and physical properties impact the performance of zeolite-modified potentiometric sensors. In addition to steric factors, e.g. zeolite pore size, the important characteristics governing the sensor performance are the Si/Al ratio and the presence of specific extraframework cations. For instance, K+ and Na+ show a strong effect on the potentiometric sensitivity towards Ca2+. The level of precision achieved by the PLS models indicates that semi-quantitative predictions are feasible. To improve the computational models, larger sets of data with a wider range of zeolite-modified sensors are necessary. The constituent materials of such sensors should have a set of well-defined properties, which can be controlled and tuned for a particular application. It is anticipated that synthetic rather than natural zeolites would satisfy such requirements.