Date of Graduation

2002

Document Type

Thesis

Degree Type

MS

Committee Chair

Julio F. Davalos

Abstract

To alleviate the deterioration of concrete structures, High-Performance Concrete (HPC) is extensively used, particularly for bridge decks, due to its favorable durability characteristics. However, there is a problem with the high cracking tendency of HPC due to its high early shrinkage, low water-cementitious ratio, brittleness and low creep. Thus, the HPC advantages of high compressive strength and low permeability are somewhat offset by cracking problems, which reduce the service life of bridge decks. To improve this problem, the goal of this study is to modify conventional HPC mixtures with admixtures or modifiers to obtain low-cracking or no-cracking formulations using West Virginia aggregates. Using a normal concrete (NC) as a bench-mark, a total of four distinct HPC mixtures were studied: conventional HPC termed Class H by the WVDOH, HPC with shrinkage reducing admixture (HPC-SRA), HPC with latex (HPC-L), and HPC with high latex content or Overlay Latex-Modified Concrete (OLMC) as used by the WVDOH. By maintaining constant aggregate-paste volume ratio, the performance of these mixtures was evaluated through a number of tests: compressive strength, split tensile strength, free shrinkage, restrained shrinkage with ring specimens at early age, cracking tendency by measuring crack onset and width, chloride permeability, and creep for a few cases. Following ACI 363, the results showed that HPC-L and OLMC are less brittle than the other mixtures. Free-shrinkage results indicated that shrinkage-reducing admixture (HPC-SRA) minimized shrinkage significantly, followed by the addition of latex (HPC-L and OLMC). In relation to a proposed shrinkage model developed in this study, particularly for HPC-SRA and OLMC, the existing ACI 209 prediction overestimated the values, except for NC and HPC. For cracking response of restrained shrinkage, OLMC performed the best with no cracks after 90 days, followed by cracking onset for HPC-L at 60 days and HPC-SRA at 33 days. The nearly negligible free shrinkage of HPC-SRA did not prevent crack formation during restrained shrinkage. The onset of cracks for both NC and HPC occurred early, at 18 and 27 days respectively, and HPC exhibited the highest number of cracks than all other mixtures; this behavior of HPC is explained by simply analyzing the results for tensile strength, shrinkage and creep. Overall, the performance of OLMC and HPC-L was best in terms of strength, chloride permeability, and shrinkage and cracking. This study will provide better focus to continued in-depth research into mixture optimization to achieve nearly crack-free HPC mixtures under field restrained shrinkage conditions.

Download

Share

COinS