Evaluation of modulus of elasticity and fatigue behavior of alkali-activated concrete and portland cement concrete using different testing methods

Abstract

Different types of structures are subjected to dynamic loads that can cause fatigue deterioration, such as wind towers, bridges, roads and marine structures. Such structures often require the use of Portland Cement Concrete (PCC) in structural components, such as foundations. Amongst the existing alternatives to PCC, Alkali-Activated Concretes (AAC) may bear lower CO2 emissions during their production. Properties such as fatigue strength, modulus of elasticity and Poisson's ratio affect the material's behavior under varying loads. While Poisson's is usually an assumed value (0.20), modulus of elasticity is commonly determined for PCC through static tests. However, such properties can also be obtained using cyclic/dynamic loading. Fatigue necessarily requires varying loads to be tested. In this context, the general objective of this work is to evaluate modulus of elasticity and fatigue behaviors for Portland cement concrete and alkali-activated concrete. This work uses different types of tests (compressive strength, static modulus, quasi-static complex modulus, impact resonance, P-wave and S-wave ultrasound and fatigue), with variation of their parameters (frequency, age, temperature, specimen size and loading) to characterize the mechanical properties of AAC (2 concretes) and PCC (2 other concretes). Furthermore, finite element models incorporating the viscoelastic behavior of both concretes were developed to assess concretes' self-heating during cyclic fatigue loading based on thermal properties, to calculate surface temperature and compare to experimental data. No significant difference on modulus or strength for tests using different specimen sizes for all four concretes was observed. Amongst the modulus of elasticity tests, static modulus presented the lowest values and impact resonance modulus presented the highest values for all the concrete formulations, probably due to differences on strain levels and material nonlinearity. Combined P-wave and S-wave ultrasonic tests provided Poisson’s ratios higher than the commonly adopted value of 0.20. AAC specimens presented higher phase angles (around 2.3°) than PCC specimens (around 1.1°) and modulus of elasticity increased up to 10.0% with the increasing frequency (from 0.1Hz to 25Hz). Finally, the finite element method overestimated initial surface temperatures by up to 1.6°C for AAC and 0.6°C for PCC, with total increases of 6.7 ºC and 4.0ºC, respectively, with film coefficient values adopted in order to match final temperature increase between model and experiment. More work is necessary to understand the measured phase angle values, since they seem high when used to fit viscoelastic models and predict creep behavior. This work is the result of a partnership between UFC, at Fortaleza, Brazil and ENTPE, at Vaulx-en-Velin, France.