Michael J. Abrahams, Parsons Brinckerhoff Quade & Douglas, Inc.
The Cooper River Bridge project involves replacement of the adjacent Grace Memorial and Pearman Bridges that cross the Cooper River between Charleston and Mount Pleasant, SC. The new crossing will have an overall length of approximately 3 miles (4.8 km). It includes interchanges on both sides of the river and a main span of 1546 ft (471 m) that will be the longest cablestayed span constructed in North America, when opened to traffic in 2005. The new crossing will provide eight lanes of traffic and a 12-ft (3.7-m) wide sidewalk, replacing the now inadequate five lanes on the existing Grace and Pearman Bridges. The design was challenging due to the need to develop a cost effective concept that accommodated the high seismicity of the Charleston area, exposure to hurricanes, and the potential for ship collision. Charleston is the second busiest port on the East Coast.
The project design criteria called for a 100-year service life and required the design-build team to develop an appropriate solution. Although the design criteria indicated minimum concrete cover requirements, the type of reinforcing steel —coated or uncoated—and the type of concrete and mixture proportions were left to the design-build team. The decision to allow the team to develop their own solution for the project’s corrosion protection system provided a good opportunity for the engineers, material suppliers, and contractors to work together in determining a cost-effective solution for the 100-year service life—a unique advantage of the design-build process.
While the superstructure of the cablestayed main span, high-level approach spans, and curved interchange ramps use structural steel with a composite concrete deck, all other portions of the superstructure use precast, prestressed concrete bulb-tee girders with a composite concrete deck. The substructure uses cast-inplace reinforced concrete piers, with reinforced or post-tensioned pier caps, supported on large diameter drilled shafts. Except for the cable-stayed main span piers and abutments, there are no footings as the pier columns are supported directly by the drilled shafts.
Epoxy-coated reinforcement was not used because of its cost and concerns about its effectiveness in a marine environment. While a service life analysis suggested some improvement if epoxycoated reinforcement was used, the 100-year service life was obtained without the need for the epoxy coating. There was also a concern about bond of the epoxycoated reinforcement to the concrete in the plastic hinge zones of the structure. Charleston is an area of high seismicity and the structure has several plastic hinge zones. The longer lap splices required with epoxy-coated reinforcement would have made the reinforcing bar cages in each tower lift heavier and more difficult to handle. A further concern was additional time that might have been spent in the field if disputes arose about the acceptable level of damage to the epoxy coating.
The use of high performance concrete (HPC) provided an opportunity to achieve the 100-year service life. However, silica fume was not utilized in the HPC mixes because of concerns about its cost and difficulties with finishing.(1)
Consideration was also given to corrosion inhibitors, which would appear to be effective in extending the bridge’s service life. But at the dosages rates of 2 to 4 gal/cu yd (10 to 20 L/cu m) recommended by the suppliers, and with the large quantity of concrete required, they were not cost effective. Typical pier columns have a 6 to 8 ft (1.8 to 2.4 m) diameter, and adding corrosion inhibitor to such large placements became very expensive. There was also a concern that the inclusion of a corrosion inhibitor would reduce the concrete setting times, particularly in warm weather.
A performance based approach was used to establish the requirements for the concrete. Application of Fick’s law of diffusion to the design mix indicated that it would take 95 years for the chlorides to reach a corrosion initiation threshold value of 2 lb/cu yd (1.2 kg/cu m) at the level of the reinforcement. An additional 5 years for the propagation period provided the 100-years service life.(2-5) Initial studies were conducted to establish the salinity of the Cooper River at the bridge site. Several reports document an average salinity of 25 parts per thousand. For reference, sea water is 35 parts per thousand. Salinity varies during each tide cycle and is also affected by the volume of the river flow. Therefore, an average value was used in the diffusion analysis. The analysis also incorporated the effect of the average monthly temperature in Charleston as temperature affects the rate of diffusion.
Studies were done to evaluate the chloride contents for the adjacent Pearman and Grace Bridges. It is notable that the older Grace Bridge built in 1929 tended to have lower chloride concentrations than the newer Pearman Bridge built in 1966, as shown in Table 1.
For portions above the splash zone, very little data are available on chloride exposure rates, so considerable judgment was required in estimating the rate to be used in the analysis. Based on the relationship between concrete chloride permeability in coulombs and its diffusion coefficient recommended by Pfeifer,(6) the maximum permeabilities needed to provide a 100-year service life for various depths of cover at various locations were determined as listed in Table 2.
Local concrete suppliers were asked to develop and price concrete mixes that met the chloride permeability criteria as well as the water-cementitious materials ratio (w/cm), slump, maximum aggregate size, and volumes of concrete. The two suppliers who did develop mixes used different approaches. One supplier used slag cement for partial replacement of portland cement, while the other supplier used Class F carbon burn-out fly ash. This fly ash has a lighter color than typical fly ash as most of the carbon has been removed and the loss on ignition is typically less than 3 percent. Based on quotes submitted by the concrete suppliers, fly ash concrete was the most economical. Two typical concrete mix proportions used on the project are shown in Table 3.
As concrete with fly ash is slow in developing its impermeability, typically a year,(7) it is not practical to wait that long to conduct the permeability tests, nor is it practical to prevent exposure of the substructure elements to chlorides for one year. To address the testing issue, the sample is tested per AASHTO T 277 at 28 days after accelerated curing of the sample at 73°F (23°C) for 7 days and then at 100°F (38°C) for 21 days. This has been demonstrated to conservatively predict results of 1-year old concrete.(8) In the field, the concrete is protected from ingress of chlorides for the first year by applying a sealer shortly after the forms are stripped. Thereafter, the concrete itself achieves the desired low permeability.
The use of fly ash in concrete has other advantages. The large placements, which qualify as mass concrete, require special measures to control heat build up. This is mitigated by the inclusion of fly ash, thus reducing the contractor’s costs in cooling mass concrete placements.
There have been several instances during construction when the actual concrete cover was observed to be less than the specified values including the allowable placing tolerances, which are typically 1/2 in. (13 mm). In these instances, a silane or siloxane sealer meeting the criteria in NCHRP 244 was applied.(9) NCHRP 244 demonstrated that the use of those types of sealers would extend the service life for vertical surfaces. Hence, when a reduced cover does occur, the service life has been analyzed as described above using the reduced cover but adding in the extension to service life provided by the sealer. With this approach, we have been able to demonstrate that the 100-year criterion is still satisfied.
Further Information
For further information, contact the author at [email protected] or 212-465-5185
References
- Petrou, M. F., Harries, K. A., and Schroeder, G. E., “Field Investigation of High Performance Concrete Bridge Decks in South Carolina,” Transportation Research Record 1770, Design of Structures, Transportation Research Board, Washington, DC, 2001, pp. 12-19.
- Weyers, R. E., “Service Life Model for Concrete Structures in Chloride Laden Environments,” ACI Materials Journal, Vol. 95, No. 4, July-August 1998, pp. 445-453.
- Amey, S. L., Johnson, D. A., Miltenberger, M. A., and Farzan, H., “Predicting the Service Life of Concrete Marine Structures: An Environmental Methodology,” ACI Structural Journal, Vol. 95, No. 2, March-April 1998, pp. 205-214.
- ACI Committee 365, “Service-Life Prediction – State-of-the-Art Report (ACI 365.1R)”, American Concrete Institute, Farmington Hills, MI, 2000, 44 pp.
- Berger, R. H., “Extending the Service Life of Existing Bridges by Increasing Their Load Carrying Capacity,” FHWA, U.S. Department of Transportation, Report No. FHWA-RD-78-133, 1978, 152 pp.
- Pfeifer, D. W., “High Performance Concrete and Reinforcing Steel with a 100-Year Service Life,” PCI Journal, Vol. 45, No 3, May-June 2000, pp. 46-54.
- Ozyildirim, H. C., “Fabricating and Testing Low-Permeability Concrete for Transportation Structures,” VTRC 99-R6, Virginia Transportation Research Council, Charlottesville, VA, August 1998, 18 pp.
- Ozyildirim, H. C., “Effects of Temperature on the Development of Low Permeability in Concretes,” VTRC 98-R14, Virginia Transportation Research Council, Charlottesville, VA, February 1998, 24 pp.
- Zemajtis, J. and Weyers, R. E., “Concrete Bridge Service Life Extension Using Sealers in Chloride-Laden Environments,” Transportation Research Record 1561, Maintenance and Management of Highway Structures, Transportation Research Board, Washington, DC, 1996, pp. 1-5.