Development of Mechanical Properties of Concrete with Time - Experimental and Numerical Study

Abstract

The performance of concrete is usually governed by its "strength" and "stiffness" properties. As most of the concrete structures are usually designed to last in service for a period of about half a century, it is important to understand the development of these "strength" and "stiffness" properties over the course of time. Based on the standard codes and practices, these properties are assumed to continuously increase with age assuming standard conditions (moist curing environment and temperature of 20° C), or at least remain constant in case of other curing conditions. However, few studies [27] [13] have reported a reduction in the trend of the strength and stiffness properties with time for the ordinary concrete of different strength mixes, when exposed to drying at different relative humidities after an initial moist curing period. Moreover, such decrease in trend has also been reported for a Geopolymer concrete upon drying at controlled lab conditions (20° C and 50%RH) after an initial moist curing period of 28 days [35]. It is not clear if the reduction in the trend observed is only temporary as a result of the imposed shrinkage deformations and eigen stresses, or if it is a permanent reduction as a result of microcracking. In this regard, it is aimed to understand the uncertainty in the development of the strength and stiffness properties of particularly the ordinary concrete mixes, which have been used across the world for a long time. Accordingly, it is hypothesized that a similar reduction upon drying in the trend of mechanical properties as reported by Prinsse [35] for a Geopolymer might also occur for the ordinary concrete mixes of various grades, which are cured and tested in the same way as the aforementioned Geopolymer concrete. Based on that, parameters such as the curing conditions, specimen size and the grade of concrete mix are assumed to influence the trend and a comprehensive literature study is performed in order to understand the influence of these assumed parameters on the mechanical properties - namely the compressive strength, splitting strength and elastic modulus over time. From the literature study, it is understood that there is some uncertainity in understanding the trend which is aggravated by the complex inter-dependency between the assumed parameters. Accordingly, an experimental program is set up to firstly verify whether there is a similar reduction in trend like the Geopolymer concrete and secondly to give more clarity on the complex inter-dependency between the assumed parameters. Further, the development of eigen stresses is understood to play a huge role in the trend of the measured mechanical properties over time and their influence might vary for the different tests. However, due to its temporary occurrence and larger testing frequency days, they are difficult to capture in the experiments. Accordingly, a FEM tool known as "FEMMASSE" is used, which can simulate the concrete behaviour by incorporating the heat and moisture models. Apart from considering the material effects like hydration, FEMMASSE can also capture the eigen stresses developed as a result of the thermal and moisture gradients. The software mostly finds application in the design of bridges and tunnels. Although, in this research study, the main purpose of using FEMMASSE is to firstly capture the shrinkage induced eigen stresses by simulating the tests at smaller intervals and secondly single out the influence of the studied parameters in order to give more clarity on the obtained experimental results. The obtained experimental results indicate no major reduction in the trend of the studied mechanical properties in time as compared to the Geopolymer concrete. With regard to the studied parameters, it is found that for the chosen curing regime (28DM regime), the trend of both compressive and splitting strength is seen to be affected by the development of the eigen stresses. Interestingly, for the trend of splitting strength, a general pattern of increasing and stabilizing behaviour is observed which seems to be dependent also on the specimen size. However, due to the lack of knowledge of the actual magnitude of the eigen stresses coupled with other phenomena like the size effect and current state of hydration, it becomes challenging to explain the trend merely on the basis of experimental results. In this regard, the use of numerical tool FEMMASSE is shown to help understand the experimental results at greater depth. Owing to the limitations of the software, only the tensile tests are simulated. Initially, it is understood that the attainment of the hygral equilibrium is dependent on the specimen size and the type of concrete,with the smaller specimen sizes and lower strength mixes reaching the equlibrium relatively quickly. The model is then extended to simulate the tensile tests - splitting and flexural and direct tension tests. It is concluded that drying after a moist curing period of 28 days affects the these tensile tests differently. In case of the splitting tests, drying results in the temporary increase of the strength owing to the apparent prestressing at the core of the specimen, with the gradual stabilization of the trend as the specimen attains moisture equilibrium. This trend of temporary increase and gradual stabilization is dependent on the specimen size with the smaller specimens stabilizing at a relatively quicker time due to the faster attainment of the moisture equilibrium. However, in case of the flexural and direct tensile test, drying results in the temporary reduction in the flexural strength due to the reduction in the tensile capacity at the surface, followed by an increasing trend as the specimen starts attaining hygral equilibrium. This indicates that the surface failure tests like the direct tension and flexural are negatively affected (temporarily) by drying as compared to the interior failure tests like splitting test which is positively affected (temporarily) by drying. From the research study, it is evident that there are no major reduction in the trend of strength and stiffness properties over time obtained experimentally until the period of 155 days. However, the presence of the eigen stresses, which not captured in the experiments, does influence the trend of the measured material properties over time. As long as these eigen stresses are present, the material strength determined experimentally might be deceptive, as evident from the simulations of the tensile tests performed in the study. It is understood that unless there is hygral equilibrium (no eigen stresses) across the specimen, the measured material properties might underestimate the actual material strength in case of the flexural and direct tension tests and overestimate in case of the splitting tensile tests. In the engineering practice, especially in mass concrete structures, the eigen stresses could be present throughout the service life of the structure. Thus, the experimentally obtained material strength used for designing the structure might be deceptive due to the influence of the eigen stresses.