An underground concrete structure in Japan was severely damaged. The structure was located in a soil that contained considerable hydrogen sulfide and the effects of various microorganisms could not be disregarded. In this study, several types of bacteria in the soil around the structure were isolated and cultured, and simulation tests of mortar deterioration were performed to clarify effects of bacteria on concrete deterioration. Hydrogen sulfide-producing bacteria or sulfur-oxidizing bacteria or both were used. Calcium ion was dissolved out of mortar soaked in the culture medium inoculated with hydrogen sulfide-producing bacteria. The value of pH was decreased and the quantity of hydrogen sulfide was increased in the culture medium as growth of the bacteria proceeded. Since this dissolution of calcium ion from mortar was not observed in the control medium, metabolites of these bacteria could be one of the reasons. From the results of TG-DTA, CaCO 3 content in the mortar surface increased and Ca(OH) 2 content decreased. This is because carbon dioxide produced by the bacteria caused carbonation of the mortar. It has already been established that concrete can be severely damaged by anaerobic bacteria, but this study suggests that concrete can also be damaged by metabolites of aerobic bacteria.

For the calculation of civil engineering structures, designers employ the mechanical aspect underestimating the physicochemical phenomena in connection with the hydration of cement paste. Although the mechanical approach is widely sufficient for classical structures, this is not the case for large structures like dams, where thermophysical phenomena play a leading part. After a short analysis of the degradation observed on a roller compacted concrete dam, showing the importance of the control of hydration effects on mass concrete, the authors present a thermomechanical model able to describe the main evolutions of concrete properties with aging. Application to the Riou dam shows the ability of the approach to simulate temperature, strains, and stresses and, as a consequence, the risk of damage for the structure. Cracks in the middle of the dam are properly represented. This approach permits determination of the position and number of construction joints and setting the schedule of construction as thickness of concrete layers or maximum delay between two layers.

The authors indicate that, in addition to steel corrosion, calcium chloride can act specifically as an aggressive agent for concrete through the formation of calcium oxychloride (3CaO CaCl 2 15H 2O). This product is formed by reaction of CaCl 2 diffusing through the cover and Ca(OH) 2 produced by cement hydration. The main purpose of the paper was to study the influence of the cementitious system (portland cement with and without a pozzolanic addition) on the damaging effect caused by CaCl 2 used as a deicing agent. To block both steel corrosion and concrete deterioration, the reduction of the water-cement ratio in the concrete mix should be accompanied by the utilization of slag cement or pozzolanic cement. The slag content should be at least 50 percent of the cement, but silica fume (> 15 percent by weight of cement) instead of fly ash (30 percent) is preferred.

There are currently several commercially available corrosion-inhibiting chemical admixtures for reinforced concrete. The two most commonly used are a calcium nitrite-based admixture that inhibits corrosion by reacting to ferrous ions to form a protective ferric oxide film; and a water-based organic admixture, consisting primarily of amines and esters, which functions by reducing chloride ion ingress into concrete and by forming a coating on the surface of the embedded steel. Because of the different mechanisms by which these two inhibitors function, comparative accelerated time-to-corrosion evaluations were performed to obtain a measure of their effectiveness relative to one another in both uncracked and cracked reinforced concrete specimens. Concrete of moderate to low quality was used. The calcium nitrite inhibitor was evaluated at dosages of 10, 20, and 30 L/m 3 and the water-based organic inhibitor was evaluated at a dosage of 5 L/m 3. Companion untreated concrete specimens, which served as controls, were also evaluated. The duration of the cracked concrete time-to-corrosion evaluation was 23 weeks and was 48 weeks for the uncracked concrete. The time-to-corrosion evaluations indicate that both inhibitors were effective. The water-based organic inhibitor was particularly effective in the cracked concrete time-to-corrosion evaluation relative to the calcium nitrite inhibitor. The degree of effectiveness of the calcium nitrite increased with increasing dosage.

Corrosion characteristics of steel embedded in hardened cement pastes and mortars were investigated by using data obtained from potentiodynamic anodic polarization and polarization resistance techniques. One normal portland cement and one fly ash were used. The dosages of fly ash as cement replacement material were 0, 20, 40, and 60 percent. The results indicate that there was no negative effect of pozzolanic reaction of fly ash on steel passivation, even at a high replacement dosage of 60 percent, and after 2 years of curing. In fact, with prolonged curing, steel embedded in fly ash-blended cement pastes was found to have a higher degree of passivation with greater stability than that embedded in plain cement paste. Chloride binding capacity of 40 percent fly ash-blended cement paste, as indicated by the measured corrosion rate of steel, was found to be very effective after 3 days of curing. In accelerated carbonation condition, corrosion rates of steel were initially high in fly ash-blended cement mortars. When fully carbonated, the results indicate that the corrosion rate of steel can be higher in plain cement in comparison to 40 percent fly ash blend. In the case of chloride penetration, the corrosion rate of steel was found to be consistently less than that of equivalent plain cement when compared on an equal water-binder ratio.