Carbonation is common in old structures, badly built structures (particularly buildings) and reconstituted stone elements containing reinforcement, that often have a low cement content and are very porous. Carbonation is rare on modern highway bridges and other civil engineering structures where water/cement ratios are low, cement contents are high with good compaction and curing, and there is enough cover to prevent the carbonation front advancing into the concrete to the depth of the steel within the lifetime of the structure. On those structures exposed to sea water or de-icing salts, the chlorides usually penetrate to the reinforcement and cause corrosion long before carbonation becomes a problem. Wet/dry cycling on the concrete surface will accelerate carbonation by allowing carbon dioxide gas in during the dry cycle and then supplying the water to dissolve it in the wet cycle. This gives problems in some countries in tropical or semi-tropical regions where the cycling between wet and dry seasons seems to favour carbonation, e.g. Hong Kong and some Pacific Rim countries. When a repairer talks of repairing corrosion due to ‘low cover’ he usually means that the concrete has carbonated around the steel leading to corrosion. As the cover is low it was a quick process, perhaps within five years of construction. If the concrete were of the highest quality, carbonation may not have been possible and low cover might not have mattered. Carbonation is easy to detect and measure. A pH indicator, usually phenolphthalein in a solution of water and alcohol, will detect the change in pH across a freshly exposed concrete face. Phenolphthalein changes from colourless at low pH (carbonated zone) to pink at high pH (uncarbonated concrete). Measurements can be taken on concrete cores, fragments and down drilled holes. Care must be taken to prevent dust or water from contaminating the surface to be measured but the test, with the indicator sprayed on to the surface, is cheap and simple.

Like carbonation, the rate of chloride ingress is often approximated to the laws of diffusion, but chlorides present complications. The initial mechanism appears to be suction, especially when the surface is dry. Salt water is rapidly absorbed by dry concrete. There is then some capillary movement of the salt-laden water through the pores followed by ‘true’ diffusion. There are other opposing mechanisms that slow the chlorides down. These include chemical reaction to form chloro-aluminates and absorption on to the pore surfaces. The other problem with trying to predict the chloride penetration rate is defining the initial concentration, as chloride diffusion produces a concentration gradient not a ‘front’. In other words we can use the square root relationship for the carbonation front as the concrete either is or is not carbonated, but we cannot use it so easily for chlorides as there is no chloride ‘front’, but a concentration profile in the concrete. Chloride attack mechanism The chloride ion attacks the passive layer but, unlike carbonation, there is no overall drop in pH. (though inside pits hydrochloric acid can form, which can rapidly accelerate local dissolution of the bars). Chlorides act as catalysts to corrosion when there is sufficient concentration at the rebar surface to break down the passive layer. They are not consumed in the process but help to break down the passive layer of oxide on the steel and allow the corrosion process to proceed quickly. This makes chloride attack difficult to remedy as chlorides are hard to eliminate.

Concrete buried in soils or groundwater containing high levels of sulfate salts, particularly in the form of sodium, potassium or magnesium salts, may be subjected to sulfate attack under damp conditions. An expansive reaction occurs between the sulfates and the C3A phase to form calcium sulfoaluminate (ettringite) with consequent disruption to the matrix. Past experience has shown that sulfate attack is rare in concrete, only occurring with very low cement content concretes, with less than about 300 kg/m3 of cement. As a guide, levels of sulfate above about 4% of cement (expressed as S03) may indicate the possibility of sulfate attack, provided sufficient moisture is present. Sulfate attack requires prolonged exposure to damp conditions. However, there has been recent concern with another form of sulfate attack… Thaumasite Attack – A Form of Sulfate Attack This hit the news in 1998, when the foundations to a number of bridges in the USA were found to be suffering from serious erosion and crumbling of the outer part of the concrete in the foundations. The problem was diagnosed as being due to an unusual form of sulfate attack, known as thaumasite attack. For the problem to occur, a number of factors have to be present.

• A source of sulfate
• Water (usually plenty of moisture)
• A source of limestone (as aggregate, or filler, or possibly as fill or even carbonated groundwater)
• Low temperatures ( <15°C)

The combination of these factors can cause an unusual reaction between the cement, the lime and the sulfate, to form thaumasite, a sulfate mineral. The effect is to cause serious damage and softening of the exposed outer surface of the concrete.