Nanostructured materials for sensor applications

Abstract

The utilisation of zeolites as infrared spectroscopy-based gas sensors and as potentiometric sensors in zeolite modified electrodes is reported in this thesis.
Commercial zeolites (NaX, NaY, MOR, FER, BEA-12, BEA-19 and ZSM-5) and prepared Sn-BEA zeolite have been modified by CuSO4, Cu(NO3)2 and Cu(CH3COO)2 as copper sources to obtain room-temperature CO sensors. Cu+ forms stable complexes with CO at room temperature that can be observed by IR spectroscopy. In preliminary screening, MOR impregnated by Cu(NO3)2 showed the highest response to 1mbar of CO in a vacuum and from ~10 ppm to 5000 ppm under flow conditions. After further testing, water interference presented a significant problem to sensor performance. Therefore, hydrophobic Sn-BEA (Si/Al >1500, Si/Sn = 64) was prepared. Cu(NO3)2 – impregnated Sn-BEA demonstrated resistance to water interference in a humid environment (~1500 ppm of H2O) while maintaining the sensing properties.
Hydrophobic Sn-BEA and dealuminated BEA zeolite were prepared from a commercial BEA-19 precursor to obtain a room-temperature infrared spectroscopy-based exhaust gas fumes sensor. Due to fluoride-assisted synthesis, Sn-BEA has high hydrophobicity and a nearly defect-free structure. Both materials have been subjected to CO, CO2, NO and NO2 in Ar flow containing 100 ppm of H2O. Dealuminated BEA showed a response to CO2 and NO2, while Sn-BEA showed selective response only to NO2. Also, Sn-BEA demonstrated two types of response that could be used for determining both the current and the cumulative concentration of NO2.
The new concept of zeolite-modified electrodes called “Ion Sensitive Pencil” has been established. Zeolite and graphite were combined in a uniform mixture (40:60 wt.% ratio) and pressed by a hydraulic press (4 tonnes) to form a pellet, which was subsequently used to draw electrodes. Heulandite zeolites from three natural deposits and 11 commercial zeolites (NaX, KX, NaY, KY, NaA, KA, BEA-12, BEA-19, ZSM-5, MOR, FER, LTL) have been used to detect 9 cations (Na+, K+, NH4+, Ca2+, Mg2+, Mn2+, Zn2+, Ni2+, Cu2+). Most sensors exhibited near-Nernstian responses to alkali and alkaline earth cations, while the responses to transitionmetal cations were relatively low. Zeolites demonstrated inferior selectivities and sensitivities as compared to classical ion-selective electrodes, although the production is simpler and less expensive. Therefore, the zeolite-containing electrodes were used in a multisensor array to quantify Mg2+ content in a model plant fertilizer solution (Na+, K+, NH4 +, Mg2+). The relationship between the responses was determined using the partial least squares method. The R2 coefficient in cross-validation was 0.83 and the root mean square error of cross-validation was 0.34 log[Mg2+].
The structure-performance relationship of ion-sensitive pencils has been determined using chemometrics. Principal component analysis has been used to visualise the variation of selected zeolite properties (Si/Al; Al/(Si+Al); Pore Size (Å); Largest Channel, MR; Channel Network; Extraframework Cations (Na+, K+, NH4+, Ca2+, Mg2+ or Fe3+); Crystallite Size (XRD); Particle Size (SEM-TEM); SiOH Intensity). Partial least squares regression was used to relate zeolite properties with sensor responses to Na+, K+, NH4+, Ca2+ and Mg2+. Pore size, largest channel and Si/Al ratio affect the sensor performance the most. Additionally, it was found that the presence of K+ and Na+ as extraframework cations affect the sensitivity towards Ca2+. The R2 coefficients in calibration models were 0.81-0.96 and root mean square errors of cross-validation for the presented models were 4.1-12.4 mV/dec.