Impedance-Based Structural Health Monitoring to Detect Corrosion

Abstract

Corrosion begins as moisture penetrates the protective barrier of a surface, starting an electrochemical process which over time leads to surface pitting. The combined action of mechanical stresses and corrosion induced pitting reduces structural integrity as the pits enlarge to form nucleation sites for surface cracks, which propagate into through-thickness cracks. In most cases, the total mass loss due to corrosion within the structure is small; however, significant reductions in mechanical strength and fatigue life can occur in the corroded material leading to advanced crack growth rates or fast fracture. Since the structural damage due to localized corrosion pitting is small and the crack growth rates may be large, traditional inspections methods and "find it and fix it" maintenance approaches may lead to catastrophic mechanical failures. Therefore, precise structural health monitoring of pre-crack surface corrosion is paramount to understanding and predicting the effect corrosion has on the fatigue life and integrity of a structure. In this first third of this study, the impedance method was experimentally tested to detect and the onset and growth of the earliest stages of pre-crack surface corrosion in beam and plate like structures. Experimental results indicate the impedance method is an effective detection tool for corrosion induced structural damage in plates and beams. For corrosion surface coverages less than 1.5% and pit depths of less than 25 microns (light corrosion), the impedance method could successfully detect corrosion on plates and beams at distances up to 150 cm from the sensor location. Since the impedance method is a proven tool for corrosion detection, it makes sense to determine how well the method can quantify and track key corrosion variables like location, pit depth, and surface coverage. In order to make fatigue life adjustments for corroded structures it is necessary to quantify those variables. Thus, the second portion of this study uses the impedance method to quantify corrosion location, pit depth, and location. Three separate tests are conducted on beam-like structures to determine how well the damage metrics from the impedance method correlate to the key corrosion variables. From the three tests, it is found that the impedance method correlates best with the changes in corrosion pit depth, so if combined with data from routine maintenance it would be possible to use the impedance method data in a predictive or tracking manner. The impedance method can be correlated to location and surface coverage changes, but the relationship is not as strong. Other NDE techniques like Lamb Waves could use the same sensors to quantify corrosion location, and perhaps surface coverage. The impedance method can detect and quantify pre-crack surface corrosion which leads to shortened fatigue life in structures; however, the sensors must be robust enough to withstand corrosive environments. The last portion of this study tests the following: corrosive effect on Lead Zirconate Titnate (PZT) and Macro Fiber Composites (MFC) sensors, Kapton protected MFC actuators for corrosion detection, and determines if corrosion damage can be sensed on the side of the structure opposite the damage. Sensor recommendations regarding the use of piezoelectric sensors in corrosive environments are made.