Chloride Ingress in a Swedish Road Environment

Rapport, 2002

Concrete blocks have been exposed up to five winters at a field station along the motorway Rv40 west of Borås, some 60 km from the Swedish west coast. The concrete mixes are made from a number of Nordic cements and other types of binders (fly ash, silica fume and blast furnace slag) and admixtures. The concrete blocks have been periodically investigated for chloride ingress, frost and salt attack, moisture content and reinforcement corrosion. In this report the chloride ingress measured for three of the exposed concrete mixes, with w/b=0.40 and 0.75 and with an addition of silica fume, is presented. The exposure environment has been documented with data from a meteorological station at the field station together with information about spread of de-icing salt and records of water, snow and ice around the blocks. The chloride ingress is measured as chloride ingress profiles, where the quotient between the chloride and calcium content is given in depth-intervals from the surface. The results show the following: • Chloride ingress during a year. There is a build up of chlorides in the concrete during a winter. The build up seems to be more rapid for vertical surfaces but the maximum ingress depths are higher for horizontal surfaces. During the summer there is a redistribution of chlorides, with a transport into the concrete and a decrease of the surface chloride content. • Effect of age at the first exposure to chlorides. The age of the concrete at first exposure to chlorides seems to have a decisive effect on the achieved chloride ingress. If the concrete is young at the first exposure to chlorides the result will be a comparable higher chloride ingress, while a lower chloride ingress is achieved if the concrete is older at the first exposure to chlorides. The reason for this is probably that the pore-structure of the concrete is not fully densified, since the cement hydration has not reached its final level. The results indicate that a difference in age of one to two weeks at the first exposure to chlorides may give a significant different chloride ingress. • Development in time for chloride ingress. The major part of the chloride ingress takes place during the first winter. After that only a slow increase in the chloride ingress has been observed. If an apparent diffusion coefficient, DF2, is determined special attention should be paid to the definition of the exposure time (total time or time of exposure to chlorides?). • Effect of surface orientation on chloride ingress. The surface orientation (horizontal/vertical) does not seem to have any significance on the chloride ingress into concrete with low w/b (w/b=0.40). However, for concrete with high w/b (w/b=0.75) the chloride ingress and the penetration depths are higher for vertical surfaces. • Effect of concrete composition. The concrete composition has a significant influence on the achieved chloride ingress, where, as expected, the w/b is the dominating factor. A low w/b gives a low chloride ingress and vice-versa. However, the effect on chloride ingress from use of silica-fume in the concrete is unclear – sometimes silica-fume concrete has lower and sometimes higher chloride ingress compared to Portland cement concrete. Probable explanations could be the slower degree of reaction and lower chloride binding in silica-fume concrete compared to Portland cement concrete, which means that the age of the concrete at first exposure to chlorides is important. It is obvious that more research is needed to better explain the effect of silica-fume in concrete.