Evaluation of linear and nonlinear vibration methods to characterize induced microstructural damage in portland cement-based materials

The ability to characterize the microstructure of concrete provides insight to the material’s load bearing capacity and durability. Apart from semi-destructive methods such as coring, and exclusive laboratory techniques such as X-ray diffraction (XRD) and scanning electron microscopy (SEM), no practical laboratory methods that directly evaluate the microstructure of concrete exist. The main objective of the present study was to investigate the use of practical vibration methods to characterize the behavior of portland cement paste and mortar subjected to oven drying and alkali silica reaction (ASR) activity, respectively, and thus to infer and to distinguish the resulting microstructural changes. The effect of internal moisture content on the vibration responses also was investigated. Linear (resonant frequency, half power bandwidth) and nonlinear (fast dynamics, slow dynamics) vibration measurements, using both transverse and longitudinal modes of vibration, were applied in this study. Vibration excitation and sensing was carried out as described in ASTM C 215. The non-classical nonlinear fast dynamic response was measured using a moving window regime on a single transient time domain response. The nonlinear slow dynamic response was measured using a simple repeated impulse vibration procedure. Linear resonant frequency data were very consistent and reduced with increasing duration of oven heat exposure and alkali silica reaction time. Both of these environmental exposures are expected to promote material microstructural changes, including an increase in distributed microcrack volume with continuing environmental exposure. However, the change in resonant frequency was shown to be significantly affected by internal material moisture content, and the effects of moisture and damage (distributed material microcracking) cannot be clearly distinguished. A separate study on the effects of gentle drying and wetting showed that the linear resonant frequency response due to drying and wetting showed hysteretic behavior that is recoverable, and likely not a result of microstructural changes. The material moisture content of the outer skin region of the prismatic samples appears to influence the vibration response more than that in the core. Some nonlinear fast dynamic parameters and half power bandwidth too showed this hysteretic behavior with drying and wetting, but with considerable variation. It was expected that half power bandwidth would be able to characterize damage due to heat exposure and due to ASR, but results do not confirm this hypothesis. Fast dynamic measurement data, as extracted by the method proposed in this study, were unsuccessful in consistently and reliably tracking microstructural changes owing to the respective environmental exposures. It is believed that the input energy provided to excite the fast dynamic behavior of the specimens was insufficient and inconsistent. The proposed slow dynamic data show promise to characterize microstructural changes, but their variability is higher than that of linear resonant frequency measurements