Predicting fracture of mortar beams under three-point bending using non-extensive statistical modeling of electric emissions

• Pressure Stimulated Currents (PSC) in cement based materials under mechanical load.
• Tsallis entropy as predictor of the expected failure.
• Study of the PSC relaxation as qq-exponential in cement based materials under mechanical load.
• Evaluation of the mechanical status of cement based materials subjected to three-point bending technique.
• Non-extensive statistical physics modeling of PSC emissions during mechanical loading.

Weak electric signals termed as “Pressure Stimulated Currents, PSC” are generated and detected while cement based materials are found under mechanical load, related to the creation of cracks and the consequent evolution of cracks’ network in the bulk of the specimen. During the experiment a set of cement mortar beams of rectangular cross-section were subjected to Three-Point Bending (3PB). For each one of the specimens an abrupt mechanical load step was applied, increased from the low load level (Lo)(Lo) to a high final value (Lh)(Lh), where LhLh was different for each specimen and it was maintained constant for long time. The temporal behavior of the recorded PSC show that during the load increase a spike-like PSC emission was recorded and consequently a relaxation of the PSC, after reaching its final value, follows. The relaxation process of the PSC was studied using non-extensive statistical physics (NESP) based on Tsallis entropy equation. The behavior of the Tsallis qq parameter was studied in relaxation PSCs in order to investigate its potential use as an index for monitoring the crack evolution process with a potential use in non-destructive laboratory testing of cement-based specimens of unknown internal damage level.The dependence of the qq-parameter on the LhLh (when Lh<0.8LfLh<0.8Lf), where LfLf represents the 3PB strength of the specimen, shows an increase on the qq value when the specimens are subjected to gradually higher bending loadings and reaches a maximum value close to 1.4 when the applied LhLh becomes higher than 0.8Lf0.8Lf. While the applied LhLh becomes higher than 0.9Lf0.9Lf the value of the qq-parameter gradually decreases. This analysis of the experimental data manifests that the value of the entropic index qq obtains a characteristic decrease while reaching the ultimate strength of the specimen, and thus could be used as a forerunner of the expected failure.