Development and Construction of Low-Cracking High-Performance Concrete (LC-HPC) Bridge Decks: Free Shrinkage Tests, Restrained Ring Tests, Construction Experience, and Crack Survey Results

Abstract

The development, construction, and evaluation of low-cracking highperformance concrete (LC-HPC) bridge decks are described based on laboratory test results and experiences gained during the construction of 13 LC-HPC bridge decks in Kansas, along with another deck bid under the LC-HPC specifications but for which the owner did not enforce the specification. This study is divided into four parts covering (1) an evaluation of the free shrinkage properties of LC-HPC candidate mixtures, (2) an investigation of the relationship between the evaporable water content in the cement paste and the free shrinkage of concrete, (3) a study of the restrained shrinkage performance of concrete using restrained ring tests, and (4) a description of the construction and preliminary evaluation of LC-HPC and control bridge decks constructed in Kansas. The first portion of the study involves evaluating the effects of the duration of curing, fly ash, and a shrinkage reducing admixture (SRA) on the free-shrinkage characteristics of concrete mixtures. The results indicate that an increase of curing period reduces free shrinkage. With 7 days of curing, concretes containing fly ash as a partial replacement for cement exhibit higher free shrinkage than concretes with 100% portland cement. When the curing period is increased to 14, 28, and 56 days, the adverse effect of adding fly ash on free shrinkage is minimized and finally reversed. The addition of an SRA significantly reduces free shrinkage for both the 100% portland cement mixture and the mixture containing fly ash. The second portion of the study investigates the relationship between the evaporable water content in the cement paste and the free shrinkage of concrete. A linear relationship between free shrinkage and evaporable water content in the cement paste is observed. For a given mixture, specimens cured for a longer period contain less evaporable water and exhibit lower free shrinkage and less weight loss in the free shrinkage specimens than those cured for a shorter period. The third portion of the study evaluates the cracking tendency of concrete mixtures using the restrained ring tests. Different concrete ring thicknesses and drying conditions have been tested. The results indicate that specimens with thinner concrete rings crack earlier than those with thicker concrete rings. Exposing specimens to severe drying conditions results in the earlier formation of cracks, although it does not result in increased crack width. Mixtures with a lower watercement (w/c) ratio crack earlier than mixtures with a higher w/c ratio. Concretes with a higher paste content crack earlier than concretes with a lower paste content. The final portion of the study details the development, construction, and preliminary performance (with most bridges at three years of age) of LC-HPC and control bridge decks in Kansas. The results indicate that the techniques embodied in the LC-HPC bridge deck specifications are easy to learn. Contractor personnel can be trained in a relatively short time. The techniques used for LC-HPC bridge decks are effective in reducing bridge deck cracking. The crack surveys indicate that LC-HPC bridge decks are performing much better than the control decks, with average crack densities reduced by about seventy five percent at three years of age. The factors that may affect bridge deck cracking are analyzed. The analyses indicate that an increase in paste content, slump, compressive strength, maximum daily air temperature, and daily air temperature range causes increased crack densities. Contractor techniques influence cracking.