Nieuwe materialen voor percolatiegebaseerde sensoren toepasbaar en resistent in een industriële omgeving

Water ingress in industrial structures (e.g., insulated pipelines) is potentially dangerous, and its detection before related damage appears is crucial. A promising solution for reliably detecting water ingress involves using percolation-based sensors, such as the PercoSens® sensor developed at KU Leuven. This resistive-like sensor exhibits highly non-linear sensitivity to water, which was achieved by coating a textile-based cable with a sensing composite material made of polyvinyl alcohol (PVA) and titanium carbonitride particles (TiCN). However, implementing this cable sensor in industry is difficult due to its poor flexibility, resulting from the brittleness of PVA. Therefore, this PhD thesis explored how the sensor's flexibility can be improved without compromising its sensitivity to water. Additionally, the TiCN-PVA material remains thermally stable only up to 170 °C, thereby challenging its integration in sandwich composite structures during manufacturing. Hence, this PhD thesis also aimed to improve the sensor's applicability at higher temperatures. Plasticizing PVA with propane-1,3-diol (PDO) significantly reduced its glass transition temperature (Tg), reducing brittleness and thereby enhancing the flexibility of the PercoSens® sensor. The highly non-linear sensitivity to water was also maintained, but only up to a critical PDO concentration. More importantly, cyclic bending tests in dry conditions and humidity cyclic tests revealed the PercoSens® sensor's poor durability, which was considerably improved using PDO. After determining the optimal PDO concentration, the plasticized sensor was scaled up and tested in insulated pipes over a 1.6-year pilot study, ultimately confirming its promising potential for detecting water ingress. An alternative heat-resistant percolation sensor, based on a carbon black (CB)-epoxy nanocomposite, was also investigated. Stencil-printed sensors exhibited humidity sensitivities ranging from 0.40 to 0.90%/%RH, significantly surpassing the existing benchmark (≤0.02%/%RH). Looking at the microstructure, this PhD thesis demonstrated that enhancing the water sensitivity of CB-epoxy nanocomposites requires proper dispersion of the CB fillers in epoxy using sonotrode ultrasonication, along with the strategic selection of a filler loading near the Electrical Percolation Threshold. However, CB-epoxy sensors cannot reliably detect water owing to the negligible increase of the electrical resistance on a logarithmic scale, making them substantially less sensitive than (PDO)-PercoSens® sensors.