While there are a number of industrial structural applications where the use of steel fibres to replace traditional reinforcement is justified under the ultimate limit state, the main justification of this replacement is, without any ambiguity, in the service limit state. In other words, it is mainly in the field of the behaviour of concrete structures in service, since steel fibres are mechanically superior to traditional rebars.
This reality, as observed experimentally and also physically, is related to control of cracking. It can be stated that for a given concrete structure and for the same service loading, the structure with steel fibres, with a dosage considered as ‘structural’, contains cracks much thinner and less spaced than the structure with traditional reinforcement.
This finding is perfectly explained by the fact that, due to their dimensions, especially their diameter, the fibres are mechanically more efficient than traditional rebars when cracks have openings of lower or equal to 500 microns. It is a question of ‘mechanical fitting’ between the diameter of the reinforcement, fibre or rebar, and the crack opening bridged by this reinforcement. This is why, for larger cracks of several millimeters, the rebars are mechanically more efficient than the fibres.
Returning to cracks appearing under service limit state loadings, and in addition to the question of thinness, two other issues that distinguish cracks in reinforced concrete (RC)structures can be added to cracks in steel-fibre reinforced concrete (SFRC) structures:
To sum up, the reality is that the cracks related to SFRC structures are much thinner, more tortuous and less continuous than the cracks related to RC structures. This reality has the consequence that, in a requested service structure, fluids (liquids and gases) penetrate more easily in a structure in SFRC than in a structure in RC.
Now consider the case of a structure under a chlorinated environment.
The presence of rebars or steel fibres is immediately synonymous with potential corrosion of these two types of reinforcements.The reality is that the fibres behave much better than the rebars when this corrosion phenomenon is considered. This superiority is based on two main points:
If there is corrosion, the volume of corrosion products generated by corrosion of rebars is sufficient to split the concrete surrounding the rebars, which is not the case for the corrosion of fibres. This splitting of the concrete around the rebars leads to two main consequences: the corrosion process accelerates and the bearing capacity of the structure decreases.
In the case of SFRC, the fine cracks generate two positive phenomena where the corrosion of fibres is concerned: the first is related to the ‘clogging’ of the cracks by corrosion products deposited within them and the second is related to the self-healing of these cracks by the continuation of the hydration process within them (this phenomenon does not exist for cracks of wider dimension, as is the case for cracks in RC structures.
Where structural design or choice between different technical solutions are concerned, the knowledge of this superiority of SFRC over RC, based on experience and physical evidence, is not always sufficient to convince the owners to choose SFRC rather than RC, especially in a chlorinated environment. They need to see quantitative proof, by calculation, in the frame of a design approach that the crack openings under service limit state loadings will be significantly smaller with a SFRC than with a RC design.
There are, today, methods or computational tools that provide access to the quantification of these cracks opening in the SFRC structures. To be more precise, there are two modelling approaches: a simplified approach used in design codes (Model Code of Fib (French, Italian, German) Recommendations) and a more sophisticated and physical approach using finite element simulations.
This is a well known and conventional approach that consists of balancing the efforts in a cracked section of the structure. It is an approach that has been developed historically for the design of RC structures. With regard to design of SFRC structures, it is based on knowledge of the post-cracking behaviour of the material in uniaxial tension, a behaviour which connects the tensile stress at the crack opening. This simplified approach has some important limitations:
There are numerical models that take account of cracking of SFRC structures. These models, which are fully validated [1,2], provide precise access to the crack spacing and opening in a SFRC structure which exhibits a hyperstatic or isostatic behaviour. It is important to emphasise that the majority of design codes allow the use of finite element modelling.
As a conclusion to this paper, it is important to keep in mind that it is demonstrated and has been proven, experimentally and physically, that SFRC structures are more durable in their service limit state situation than RC structures, and that this is the case more particularly in chlorinated environments. This superiority can be quantified by the use of powerful and validated numerical tools.
1. Tailhan, J.L., Rossi, P., Daviau-Desnoyers, D. Numerical modelling of cracking in steel fibre reinforced concrete (SFRC) structures. Cement and Concrete Composite, vol. 15, pp. 315-321, 2015.
2. Rossi, P., Daviau-Desnoyers, D., Tailhan, J.-L. Analysis of Cracking in Steel Fibre Reinforced Concrete (SFRC) Structures in Bending using Probabilistic Modelling. Structural Concrete, DOI: 10.1002/suco.20140008, 2015.