Online corrosion monitoring in industrial boilers

Fireside corrosion of evaporator walls results from various reaction mechanisms between the furnace wall, ash deposits and the flue gas atmosphere. Despite the moderate steam parameters, waste incineration plants in particular are strongly affected by corrosion due to the high chlorine content of the combusted waste. As a result, the ash deposits contain molten chloride and sulfate eutectics at elevated temperatures. Since molten salts are electrolytes, it is possible to investigate such corrosion phenomena using electrochemical methods. This allows an immediate detection of corrosion-relevant operational changes such as air ratio and fuel changes. In order to monitor fireside corrosion in combustion plants, an online corrosion sensor was developed by the Institute for Energy Systems and Technology of TU Darmstadt. In the course of a revision, six corrosion sensors were installed in the membrane wall of the waste-to-energy plant in Berlin-Ruhleben, whereby two sensors are placed in each of the three vertical passes. The results show clear differences in the corrosion activity of the different passes. The highest corrosion current occurs in the first pass, where short-term fluctuations of the evaporator wall and flue gas temperature cause strong fluctuations of the measured corrosion current. In the second pass, the measured corrosion signal is characterized by the growth and detachment of the deposits. The spontaneous partial loosening of the deposits leads to a sudden increase in surface temperature and heat flux density, resulting in a sharp increase in the corrosion signal. In the third pass, the measured corrosion current is negligible and indicates no significant material loss. A correlation analysis identifies the ratio between primary and secondary air as the most significant operational influence on the occurring corrosion. At a constant total airflow, higher secondary air leads to a lower flue gas temperature and lower temperature fluctuations in the first pass, reducing corrosion. After a service time of nine months, the sensors were removed from the plant and the electrodes weighed. Based on recorded data and weighed electrode mass loss, an empirical approach for quantifying the corrosion current density in form of an online corrosion rate [mm/1000 h] is developed. The deposits on the electrodes are investigated using Incident Light Microscopy (ILM) and Scanning Electron Microscopy with Energy Dispersive X-ray Spectroscopy (SEM/EDX) in order to identify occurring corrosion mechanisms.