Chloride migration characteristics and reliability of reinforced concrete highway structures in Pennsylvania

Abstract

An experimental study was conducted to assess the chloride migration resistance of concrete used for bridge superstructure and substructure projects in the state of Pennsylvania. A total of 68 concrete mixes were sampled and tested to determine the chloride migration coefficient using NT Build 492. A subset of these samples was also examined using the rapid chloride migration and surface resistivity tests. The tested samples exhibit a wide variation in aggregate rock type, cementitious materials and mix proportions. The results show that the concrete migration coefficient varied from 0.54 × 10−12 m2/s for a concrete with 59.4% cement type I, 35% slag and 5.5% silica fume to 30.62 × 10−12 m2/s for a concrete with 100% cement type I, the average migration for all samples was 9.51 × 10−12 m2/s. A statistical analysis shows correlation between migration coefficient, compressive strength, and water/cement ratio but no correlation with aggregate absorption, air entrainment, or aggregate type. The results were utilized to evaluate the life-cycle performance of bridge decks constructed for the Pennsylvania Department of Transportation. Reliability analyses were carried out, where the limit state consisted in the chloride content in concrete at the depth of the reinforcement reaching the critical chloride content for the reinforcing steel corrosion initiation. The chloride diffusion over time was modeled with a probabilistic implementation of Fick’s 2nd law and the evolution of reliability was assessed. The impact of the chloride migration on the expected service life of superstructure components shows that 75% of tested concrete when used with uncoated reinforcement has a 10% likelihood of corrosion initiation within 15 years. This service life is extended to 30 years when epoxy coated reinforcement is used.