John J. Roller, CTLGroup and Robert N. Bruce, Tulane University
In conjunction with construction of the Rigolets Pass Bridge, the state initiated a research program with the objective of monitoring the structural behavior of one of the two high performance concrete (HPC) bridge spans (Span 43). Material property studies for four of the HPC girders incorporated in Span 43 (Girders 43A, 43B, 43C, and 43D) were included in the program. The same girders were instrumented with strain gages to measure long-term deformations and with reference points to monitor long-term deflections.
The HPC girders were fabricated by Gulf Coast Pre-Stress (GCP) located in Pass Christian, MS. Material property tests were performed on specimens representing concrete placed in the midspan region of the HPC girders. Concrete cylinders used for compressive strength and modulus of elasticity tests were “match-cured” to match the temperature in the lower flange of each corresponding girder from the time of initial placement until strand release. Other “field-cured” specimens were covered with plastic and stored adjacent to the casting bed. All the match- and field-cured specimens were stripped from the molds just prior to release of the prestressing strands.
Compressive Strength
Concrete compressive strength tests were performed in accordance with ASTM C39 at strand release and at ages of 7, 28, and 90 days. Average measured concrete compressive strength values for the four HPC girders of Span 43 are shown in Fig. 1. Three cylinders for each girder were tested at each test age. As indicated by the data presented in Fig. 1, the concrete used in the four HPC girders exhibited very similar compressive strength values at all test ages.
Modulus of Elasticity
Concrete modulus of elasticity tests were performed in accordance with ASTM C469 at the same ages as the concrete compressive strengths. Concrete modulus of elasticity versus compressive strength data for the four HPC girders of Span 43 are presented in Fig. 2 for all test ages. The line shown in Fig. 2 represents the relationship between concrete compressive strength and modulus of elasticity given by the Ec = 33,000K1wc1.5 
expression from Article 5.4.2.4 of the AASHTO LRFD Bridge Design Specifications, where K1 is taken as 1.0 and wc is taken as 0.145 kip/ft3 (2323 kg/m3). As indicated by the data presented in Fig. 2, the AASHTO LRFD relationship between compressive strength and modulus of elasticity appears to be reasonably consistent with the measured data for the strength levels investigated.
Creep and Shrinkage
Tests to determine creep and shrinkage properties for the girder concrete were performed in accordance with ASTM C512 on field-cured 6×12-in. (152×305-mm) cylinders representing concrete placed in the midspan region of one HPC girder (Girder 43D). Creep and shrinkage tests starting at ages of 3 and 100 days were performed under ambient conditions of 73ºF (23ºC) and 50% relative humidity. For both test ages, the target applied load used for creep testing corresponded to 40% of the measured concrete compressive strength at that age. The target applied stresses were 3300 psi (23 MPa) for the 3-day tests and 4880 psi (34 MPa) for the 100-day tests. Measured creep coefficient, defined as the ratio of creep strain to initial strain, and shrinkage data for tests starting at concrete ages of 3 and 100 days are shown in Figs. 3 and 4, respectively.
Corresponding calculated values for tests starting at 3 days determined using provisions from the AASHTO LRFD Bridge Design Specifications are included in Figs. 3 and 4. According to Article 5.4.2.3 of the AASHTO LRFD Bridge Design Specifications, when mix-specific data are not available, estimates of creep and shrinkage may be made using the provisions of Articles 5.4.2.3.2 and 5.4.2.3.3, respectively. Article 5.4.2.3.2 includes an equation for calculating creep coefficient for various ages after initial loading. Article 5.4.2.3.3 includes an equation for calculating shrinkage at various concrete ages.
Based on the data shown in Figs. 3 and 4, it is apparent that the AASHTO LRFD Bridge Design Specifications provisions for estimating creep and shrinkage (when mix-specific data are not available) did not correlate well with the measured data. The final measured creep coefficient value for the 3-day age of loading was approximately twice as great as the corresponding calculated AASHTO LRFD value. The final measured shrinkage value for the tests starting at 3 days was approximately 75% of the corresponding calculated AASHTO LRFD value. Consequently, for the HPC placed in the midspan region of Girder 43D, the provisions of Articles 5.4.2.3.2 and 5.4.2.3.3 of the AASHTO LRFD Bridge Design Specifications underestimated creep coefficient and overestimated shrinkage. As stated in the Commentary to Article 5.4.2.3.1:
“Without specific physical tests or prior experience with the materials, the use of empirical methods referenced in these Specifications cannot be expected to yield results with errors less than ±50 percent.”
Further Information
Further details from this research program are available in the project report (FHWA/LA/08-437), which can be obtained through the Louisiana Transportation Research Center (LTRC). Work on this project was performed jointly by Tulane University Department of Civil and Environmental Engineering, CTLGroup, and Henry G. Russell, Inc. under the sponsorship of the LTRC and in cooperation with the Louisiana Department of Transportation & Development.