The authors study the effect of water migration and carbonation progress on the corrosion rate of reinforcing bars, focusing on the spalling of cover concrete of railway viaducts. As a result, when un-carbonation depth (=cover depth – carbonation depth) becomes less than 10 mm with supplied water, the risk that the spalling of cover concrete occurs is high. On the other hand, even if un-carbonation depth is less than 10 mm, when the water is not supplied, the risk that the spalling of cover concrete occurs is low. In addition, when the water is supplied, the corrosion rate of the reinforcing bars is 0.2 to 9.3*10^-3 mm/year. On the other hand, even if un-carbonation depth is less than 10 mm, when the water is not supplied, the corrosion rate is 0.2 to 2.7*10^-3 mm/year.

 Steel bars in concrete structures usually maintain a passivation state due to high pH in the concrete pore solution. However, it is known that the passivation film of a steel bar in concrete is destroyed by chloride attack in a marine environment. In the past studies, the condition of passivation film break down is defined by the the ratio of chloride ion concentration to hydroxide ion concentration when the passive film breaks, but this value varies depending on each researcher. In this paper, the theoretical model of passivation film break down was constructed based on point defect dynamics and it was suggested that the condition of the passivation film break down was influenced by a steel bar potential. From the experiment results, it was found that the relationship between the steel bar potential and the the ratio of chloride ion concentration to hydroxide ion concentration when the passive film breaks is an exponential relation. Therefore, it is estimated that the difference of the steel bar potential was the cause of value varies depending on each researcher.

 This study aimed to investigate explosive spalling behavior of prestressed concrete (PC) beams subjected to fire test by heating condition as a parameter. A fire test was performed by performed by exposing PC beam specimens to one-side heating. The specimens after the heating test were then tasted under static loading to determine their load capacity empirically. Explosive spalling behavior of PC beams was further investigated by using a method proposed and estimated the proof strength of PC beam after heating test.

 The result showed it would be possible to evaluate the behavior of PC members during the early stage of explosive spalling by taking the effect of initial compressive strain due to prestressing force into account. Also showed that be able to evaluate the proof strength of PC beam after fire damage by considering reduction of introduction prestress due to explosion and heating effect.

 The static loading experiments and three-dimensional finite element (FE) analyses were conducted to make clear the effects of multiple holes which were introduced to recover structural performances of deteriorated reinforced concrete (RC) members in the shear load carrying mechanisms. The experiments and FE analyses were conducted by using the specimen and analytical model of sound RC beams without stirrups. The multiple holes were introduced in the beams by embedding polyvinyl chloride pipes at the time of specimen casting. As a result of experiment, increase in the loading capacities of the RC beams with multiple holes arrayed along a spline curve compared with the control beam was confirmed. To make clear the reasons, analyses based on theoretical decomposition method of arch mechanism and beam mechanism were applied to the experimental results. The results indicated that the loading capacities of the beams with the multiple holes increased by the predominance of arch mechanism with increase in the strength of beam mechanisms. As a result of FE analysis, it was made clear that predominance of the arch mechanism in case of beams with multiple holes arrayed along a spline curve was caused by expansion of the minimum principal stress distributing region. The results also indicated that the expansion was caused by the single hole which located at the boundary of the vertical stress contributed region, and that the crack inducing path coincided with the most effective crack path for the predominance of the arch mechanism due to the single hole.

 This research aims to propose an evaluation method of shear capacity for RC T-beams that takes into account the compression flange. The static bending tests and numerical analyses by 3D non-linear FEM were conducted for RC beams with the various width and height of compression flange, shear span ratio, and shear reinforcement ratio. The experimental and numerical results show that the compression flange of RC T-beams with shear reinforcement resists the progress of the diagonal cracks and contributes the shear resistance. The shear capacities of RC T-beams increase with the increase in the width and height of the compression flange. On the other hand, the shear capacity carried by shear reinforcement is not affected by the compression flange. Finally, this study proposes a new equation for the shear capacity carried by the compression flange.

 The authors found that frost damage of RC did not occur isotropically because of the presence of reinforcing bars, and consequently, mechanical responses of RC elements were significantly influenced by loading directions. There-fore, anisotropic damage models of frost-damaged RC elements were developed and introduced in three-dimensional non-linear finite element analysis (3D-NLFEA) to simulate static failure behavior of RC beams subjected to freeze-thaw action. In the analysis, the anisotropic expansion in a section of RC beams was evaluated by a coupled equation of heat and moisture transfer. The calculated strains were given to 3D-NLFEA as initial strains, and then the damage introduced by the initial strains were considered in the anisotropic damage models. The analysis method could successfully predict not only load-deflection responses but also failure modes which were different from that of RC beams without frost damage.