Modelling of self-sustainable microbial fuel cell type oil sensors based on restricted oxygen transfer and two-population competition

Abstract

Oil leaks during oil industrial chain pose threats to the ecosystem. The microbial fuel cell-type oil sensor has been developed for early warning of such issues. Oil contacting with the sensor restricts oxygen availability and triggers correlative signal anomaly which serves as indicative of the oil presence. To extend its application for the real world, modelling of the sensor is required to pre-describe the signal behavior under unknown conditions. Therefore, by integrating Butler-Volmer, restricted oxygen transfer (ROT) and Monod equations, a dynamic ROT-MFC model with sufficient substrate precondition was developed. The ROT-MFC model was trained on the experimental single-oil-shock test (R 2 = 0.996) and validated by the experimental sequential-shocktest (R 2 = 0.998). Numerical analysis of the trained ROT-MFC model indicates that the single-shock detection has higher sensitivity (≥40.6 mV/detection) and the sequential-shocks detection spends a shorter response time (≤2.2 h). Besides, the sequential-shocks detection with proper strategy is more applicable due to flexible options on detection limit and working range. The model was further evolved into the TPC-ROT-MFC model by introducing a two-population competition (TPC) theory to describe performance under limited substrate conditions. Results indicate a critical substrate concentration range (42.1 to 62.8 mg-COD/L) for dividing baseline steadiness, and that the impact of substrate concentration on anodic charge transfer coefficient soars when the substrate concentration lessens furtherly. This sensor model is relatively easy to implement and may enhance practical use for design and operation.