John Lawler and Paul Krauss, Wiss, Janney, Elstner Associates, Inc. and Craig Abernathy, Montana Department of Transportation
Concrete bridge decks in Montana are subjected to severe service conditions. Potential deterioration mechanisms include corrosion of the reinforcing steel and scaling of the concrete surface resulting from deicing salt applications, freezing and thawing distress, cracking due to thermal and humidity extremes during and after construction, and other materials-related problems. To overcome these challenges, the Montana Department of Transportation (MDT) funded research to develop high performance concrete (HPC) mixtures to optimize the durability of bridge decks. A previous article discussed a field trial of an HPC bridge in Montana.* A parallel investigation, conducted to identify the best combination of materials available in Montana for use in HPC, is summarized in this article.(1)
Performance objectives for durable concrete decks were used to design an experimental program focused on Montana’s environmental conditions and local materials. Testing included plastic concrete properties, slump loss, setting characteristics, air-void system parameters, electrical conductivity, strength, chloride penetration resistance, freezing and thawing resistance, scaling resistance, and drying shrinkage.
Three important concepts were recognized at the initiation of this project: 1) supplementary cementitious materials (SCMs) and other concrete raw materials vary significantly depending on geographic location, 2) consideration of the raw materials themselves is important since they may impact durability, and 3) concrete production capabilities vary widely with producers or location.
Optimizing HPC for durability typically involves the use of SCMs because they improve workability and deterioration resistance. However, since some of these supplementary materials are byproducts of other industries, the SCM properties can be inherently variable. Therefore, generalizations about the best combination of SCMs and aggregates cannot be made. Rather, the most effective solution must be determined by testing with locally available materials.
The raw materials investigated included four aggregate sources, a Type I/II portland cement, and a range of SCMs, including Class C and Class F fly ashes, ground granulated blast-furnace slag (slag), high-reactivity metakaolin, and silica fume, from Montana or a neighboring state. To ensure that the aggregates do not limit the concrete durability, aggregates from four sources throughout the State were evaluated for alkali-silica reactivity. The aggregate test program, conducted using ASTM C 1260 and C 1293 procedures and petrographic examinations, suggested that three of the four aggregate sources are susceptible to potentially deleterious alkali-silica reaction (ASR), and the fourth may also be marginally at risk. However, combining the most reactive of these aggregate sources with SCMs produced concretes that experienced little or no expansion in modified C 1293 testing. This suggests that while ASR is a potential limiting factor on service life, it may be mitigated effectively in the HPC mixtures.
Concrete testing was conducted in three rounds, each targeted at a slightly different type of HPC. The first round examined mixture combinations that have historically demonstrated good performance as reported in the literature and based on the experience of the investigators. Since the mixes that performed best in the first round were complex containing fly ash, slag, and silica fume, the second round quantified the performance of pre-combined blended cements that enabled similar combinations. These portland cement blends included a slag blend, a Class C fly ash blend, and a calcined-clay blend. The third round examined easy-to-produce mixtures that have statewide application. The first two rounds were conducted using an aggregate from the Yellowstone River Valley (YRV), while the third used an aggregate source from Western Montana (WM).
In evaluating the best performer, judgments must be made about the relative importance of the desired properties. The greatest cause of deterioration in Montana bridge decks is expected to be corrosion of reinforcing steel initiated by chloride ions from deicing salts. Therefore, given acceptable performance in the other tested properties, the highest emphasis was placed on chloride penetration resistance. However, some material combinations having very good chloride resistance resulted in concrete with high scaling or shrinkage making them less desirable. Based on the 14 specific mixtures evaluated, the combinations of tested SCMs that produced the best overall performance are listed in the table.
In this test program, concretes produced using the Western Montana aggregate (mainly quartzite and sandstone) demonstrated better performance in terms of strength, resistance to chloride penetration, and scaling resistance than that measured with the Yellowstone River Valley aggregate (mainly basalt and granite). However, the shrinkage tended to be higher. The influence of the raw materials and the importance of testing each mix containing specific materials were clearly demonstrated since changing the aggregate source had a greater effect on performance than modifications to the cementitious materials. The best mix combination was different for each aggregate source. This implies that the character of the paste-aggregate interfacial transition zone, as affected by aggregate type, is of utmost importance.
Reference
- Lawler, J. S., Krauss, P. D., and Abernathy, C., “Development of High-Performance Concrete Mixtures for Durable Bridge Decks in Montana Using Locally Available Materials,” Publication SP-288, American Concrete Institute, 2005, pp. 883-902.
*See HPC Bridge Views Issue No. 35, September/October 2004.