Corrosion and shear cracking are frequently observed near supports of pretensioned bridge girders in coastal climates, so non-linear finite element analysis was used to study the effect of corrosion on shear performance in a real case study. Varying degrees of corrosion and various locations (top strands, bottom strands, vertical stirrups and girder-slab interface) were considered. The analyses included construction phases, concrete creep and shrinkage, and the effects of corrosion on the properties of prestressing and reinforcing steel, concrete, and the bond between concrete and reinforcement. The study shows that high (20%) corrosion in the bottom layer of strands can modify the failure mode from concrete crushing in the web to strand slippage or crushing of concrete in the support zone with limited cracking. Although severe strand corrosion significantly compromises girder capacity and ductility, failure occurs only when there is overloading. The predicted failure mode was also sensitive to material parameters, in particular the corrosion-induced crack widths used for modelling the reduced concrete strength. Nevertheless, some similarities were noticed between observed and predicted cracking occurrence. For moderate corrosion (10%), girder capacity was limited by strand fracture, but extensive flexure and shear cracking would appear before failure. 20% corrosion in the vertical stirrups in the web seems to have potentially smaller effect on the shear capacity than 20% corrosion in the strands in support, while corrosion in the top strands or stirrups in the girder-slab interface did not affect the girder capacity.

@en

Alkali–silica reaction (ASR) in concrete causes expansion and degradation of the material, which might give adverse structural consequences. From the structural engineer point view, the greatest concern is if ASR leads to loss of structural integrity. Two natural questions arise when assessing existing concrete structures affected by ASR: (1) how to calculate the ASR-induced stresses, and (2), when the ASR-induced stresses are calculated, what is the residual capacity when accounting for the material deterioration caused by ASR? This study aimed to contribute in answering the first question. The ASR-induced stresses can be calculated in a structural analysis that includes a concrete material model that incorporates the effects of ASR on the material behaviour, i.e. expansion and material deterioration. Many such models exist. However, these models rely on predefined field variables, e.g. moisture and temperature, which are (generally speaking) unknowns for an existing structure. Consequently, structural analysis of ASR-affected concrete structures involves dealing with unknown field variables. From this background, we developed a material model and a suitable structural analysis method. The material model relies on only one predefined, howbeit unknown, field variable —the free ASR expansion. The structural analysis method is based on solving an inverse problem, which is to back-calculate the free ASR expansion field from a set of measured displacements. The material model and the structural analysis method were applied in a structural analysis of an ordinary reinforced beam bridge in Norway. Then, the imposed deformations and stresses due to ASR were investigated to increase the understanding of the structural consequences of ASR in ordinary reinforced continuous beam bridges.

@en

Material modelling, from the micro to the macro level, of concrete affected by alkali–silica reaction (ASR) has been devoted a lot of research. However, the application of the material models in structural analyses of reinforced concrete (RC) structures, showing the structural implications/consequences of ASR, has got little attention in the literature. This paper aims to show the relevance of the constitutive model on the calculated load effects—induced by ASR—in statically indeterminate beam structures. For the purpose of the study, a three-span RC beam, inspired by a real bridge in Norway, is analysed. The RC beam is modelled using Euler–Bernoulli beam theory, and numerical solutions are obtained with the finite element method. The effects of ASR on the concrete are accounted for in an expansion based (macro) constitutive model, which also accounts for cracking, creep and compressive non-linearity. In this way, ASR gives an imposed deformation similar to thermal dilation and shrinkage, for which structural effects have been widely studied. As imposed strain gradients tend to cause higher load effects than uniform strains, the effect of ASR gradients, owing to e.g. a moisture gradient, is addressed. It is shown that linear structural analyses (using a linear material model), give conservative results (the greatest load effects) when an ASR strain gradient is imposed. Among the non-linear material effects investigated, it is shown that stress dependent ASR expansion and concrete cracking are important to consider. The stress dependency of the ASR expansion is shown to have a smoothing effect on the imposed ASR strain field, and as a result, reduces the load effects induced by ASR, while cracking results in crack/plastic hinges releasing the stresses in the system.@en

The alkali silica reaction (ASR) in concrete causes internal localized swelling and micro cracking, which result in expansion and correlated deterioration of the concrete material. The stress state of the concrete is known to affect expansion due to ASR, with an anisotropic stress state giving rise to anisotropic expansion. Similarly, the orientation and extent of micro cracking have a directional effect on the concrete mechanical behaviour. This research studied the effect of sustained uniaxial compressive stress on the evolution of the mechanical behaviour of concrete in compression. Concrete cubes of 230 mm side length were uniaxially restrained and stored in accelerated conditions, with cores drilled in two directions for mechanical tests: a cyclic test in compression, i.e. a stiffness damage test (SDT) and a complete stress-strain test. A clear directional dependency of the mechanical characteristics was found. Furthermore, the results indicate that reduction in modulus of elasticity is well correlated with the expansion in the test direction. On the other hand, the damage indices obtained from the SDT merely relate to the expansion, which puts in question the SDT's ability to predict ASR expansion in stressed concrete and therefore in concrete structures.

@en

The most common methods for detecting chloride-induced corrosion in concrete bridges are half-cell potential (HCP) mapping, electrical resistivity (ER) measurements, and chloride concentration testing, combined with visual inspection and cover measurements. However, studies on corrosion detection in pretensioned structures are rare. To investigate the applicability and accuracy of the above methods for corrosion detection in pretensioned bridge girders, we measured pretensioned I-shaped girders exposed to the Norwegian coastal climate for 33 years. We found that, even combined, the above methods can only reliably identify general areas with various probabilities of corrosion. Despite severe concrete cracking and high chloride content, only small corrosion spots were found in strands. Because HCP cannot distinguish corrosion probability in the closely spaced strands from other electrically connected bars, the actual condition of individual strands can be found only when concrete cover is locally removed. Wet concrete with high chloride content and accordingly low HCP and low ER was found only in or near the girder support zones, which can therefore be considered the areas most susceptible to chloride-induced corrosion. We conclude by proposing a procedure for the inspection and assessment of pretensioned girders in a marine environment.

@en

This study was based on findings from the Norwegian Public Roads Administration's Bridge Management System and field investigations on corrosion damage in pretensioned Norwegian standard I-beam (NIB) girders in 227 coastal climate bridges. The main durability design parameters are summarized and related to regulations over the last 80 years. Environmental exposure is discussed in the light of the global, local, and micro climate. We found that 51% of the bridges have girder corrosion damage. The percentage is highest for bridges built when the cover thickness required was lowest. Cover thickness below that required (resulting from production faults) caused 74% of corrosion damage. Most of the severe chloride-induced corrosion damage in bridges was found in the inner NIB girders, particularly in the support-zones and their vicinity. This can be due to interaction between geometry and exposure. Corrosion of reinforcement in support-zones can impact structural behavior, particularly NIB girder shear performance.

@en