Oregon DOT looks into self-curing concrete for decks

By: Bruce Johnson, Contributing Author

Like many state highway agencies, the Oregon DOT has had recurring instances of transverse deck cracking in girder bridges. ODOT has conducted previous research on the cause and found there are several contributing factors.

The agency has monitored construction practices to try to isolate the cause, but has found that it is difficult to isolate the cause to any single factor.

Based on previous work, ODOT has improved the standard construction specifications for curing over the years to reduce drying shrinkage as a cause. The latest improvements to the curing regime include continuous fogging during deck placement, mandating curing blankets be applied within 20 minutes of completing the finishing operations and using lighter polypropylene curing blanket material to reduce contractors’ concerns about damage to the freshly finished surface. ODOT also has reduced the strength requirement for deck concrete to encourage the use of less cement in the mix.

ODOT is still finding that occasional significant deck cracking is occurring. To try and address the curing issue further, ODOT is conducting research in the use of self-cured (or internally cured) deck concrete.

ODOT is trying to determine if concrete mixtures using some form of internal curing will help reduce the tendency of high-performance deck concrete to crack. Two methods are being investigated. Also, no matter how much cracking reduction is achieved, there may be other beneficial effects using internal or self-curing concrete mixtures. Internal curing using a self-curing admixture (SCA) or lightweight fine aggregate (LWFA) appears to produce a more dense mixture and potentially improve durability due to less absorption of chlorides and carbon dioxide. If cracking is reduced, additional research will be needed to quantify the additional advantages to using these mixes. Final results of the work on the SCA will be available by the end of 2010, and final results of the work on LWFA will be available by the end of 2011.

Heavy on the light

Some positive results have been reported from using porous coarse aggregates that hold moisture and then release it for the hydration process during curing. This form of internal curing may be one way to reduce cracking from drying shrinkage if a good local source of hard, porous coarse aggregate is available. Lacking that source, ODOT has elected to conduct research on the use of lightweight, porous fine aggregates. The research is under way now. The following are the research objectives:

• Develop a high-performance mix design using saturated LWFA that reduces autogenous shrinkage;
• Develop a short curing regime to complement use of LWFA mix;
• Conduct field trials; and
• Prepare guidelines for use of lightweight aggregate mix with reduced or no external curing.

The initial mix design planned for the saturated LWFA mix is as follows:

• Initial mix parameters, 30% Class F fly ash, 4% silica fume, water/cementitious material ratio of 0.37;
• Determine optimal amount of fine aggregate to replace with saturated LWFA for internal curing; and
• Initial estimate*: Mass = Cement Factor x Chemical Shrinkage x Max Hydration Max Hydration of LWFA x Absorption of LWFA.

Samples made with this mix design will undergo shrinkage testing, strength testing and visual crack surveys.

Shrinkage testing will be performed on both mortar samples and concrete samples. Concrete samples will have a portion of the fine aggregate replaced with saturated LWFA. Two innovative pieces of testing equipment will be used: a free deformation frame and a rigid cracking frame.

The research team has completed the literature review and is conducting lab testing now. It will be completed by March 2011. Field trials are planned for June–August 2011, and the final report will be available in September 2011.

Tried in Taiwan

The second method of achieving a self-curing concrete mix has been developed by Dr. Wen-Chen Jau of the National Chiao-Tung University in Taiwan. He has developed an SCA made of commonly available polymers that acts as a desiccant to hold moisture until needed for cement hydration during curing. His theory is that the demand for moisture during hydration is stronger than the admixture’s demand for moisture. Therefore, water retained by the admixture will be released and made available for curing internally. Also, internal curing using SCA or LWFA appears to produce a more dense mixture and potentially improve durability due to less absorption of chlorides and carbon dioxide.

The SCA developed by Dr. Jau has been used successfully in trial applications and in a few field applications in Taiwan, but there are some significant differences in the work completed to date in Taiwan. The first is that relative humidity in Taiwan is much higher than in many parts of Oregon. Although western Oregon has relatively high rainfall for about six months of a typical year, the other six months are relatively dry compared with Taiwan. Average humidity in many parts of Oregon can be as low as 30%.

The second difference is in the materials available for concrete mixes. Aggregates are different and the fly ash is significantly different. The research will be carried out in three phases to account for these differences. The results of each phase will determine whether or not the research will continue to subsequent phases.

The tasks for the SCA research at National Chiao-Tung University included developing the optimum SCA dosage rate using trial batches in Taiwan. Testing included the following elements:

1. Using locally available aggregates, fly ash and silica fume in Taiwan;
2. Curing using a controlled-environment chamber to simulate areas of low relative humidity common in Oregon; and
3. Using the ODOT high-performance deck concrete mix.

Testing on samples cured in a chamber with temperature and humidity control included strength, weight loss, shrinkage and cracking.

The initial testing in Taiwan found that the strength gain was less than without SCA and the shrinkage was excessive at over 400 microstrain. The research team determined that testing needed to continue to optimize dosage rate to increase strength and reduce shrinkage. Some revisions to the amount and constituent percentages of the SCA needed to be made. A Phase 1A, Phase 1B and Phase 1C were added before finding the optimal dosage and constituents of the SCA admixture.

In Phases 1 and 1A, the samples were tested with the cylinders (1) wrapped with plastic on the sides, but left exposed to the environment on the top and bottom to simulate the conditions in a concrete bridge deck, and (2) no wrapping to simulate the most demanding situation. Curing regimes included “no cure” (immediate exposure to the atmosphere) and “two-day wet cure, then exposure to the atmosphere.”

The desire was to achieve an SCA mixture that needed no wet curing. However, the tests could not achieve the desired maximum 200 microstrain shrinkage at the first stage. They also could not achieve a strength that was comparable to strength without SCA but with standard cure. Subsequent tests in Phases 1B and 1C based on the new formula of SCA, however, met the very strict condition, that is, no wrapping and maximum microstrain less than 200. Therefore, it was determined to conduct the additional testing in Oregon with a two-day cure, three-day cure and seven-day cure. Any one of these options would substantially reduce the time of construction for cast-in-place bridge decks.

The final results of Phase 1C testing were very encouraging and met the minimum requirements to move forward with Phase 2.

The following charts show the results of the final SCA dosage rate of 1% for strength, shrinkage and weight loss. It was found that the SCA is very useful in strength development, shrinkage reduction and weight-loss control. Based on these results the research was advanced to Phase 2, which included additional trial batches in Oregon using locally available materials.

Conclusions from Phase 1:

• The SCA does help in strength development;
• The SCA helps to distribute the moisture uniformly and evenly within the specimens;
• The test results show the performance of concrete specimens for different treatment on wrapping. Specimens with wrapping perform better than specimens without wrapping;
• Strength increases when drying shrinkage decreases;
• Strength increases when weight loss decreases;
• Weight loss increases when drying shrinkage increases; and
• The new SCA meets ODOT expectations (strength higher than 4,500 psi and strain less than 200 microstrain at 28 days).

Phase 2 is now under way in Oregon. The research team is conducting trial batching in Oregon using Oregon aggregates, fly ash and silica fume, and the SCA admixture developed for Oregon conditions in Taiwan.

Results from Phase 2 were available in March 2010. Based on those results the research will move into Phase 3. In Phase 3 an SCA high-performance deck mixture will be used to construct approach slabs in a project in eastern Oregon where conditions are dry (humidity less than 30%) and hot (temperature 90-100°F) in the summer. The mixture also will be used in a western Oregon project to test its performance in a condition with higher humidity and lower temperatures.

Cracking down

Deck cracking has been a recurring issue for many bridge decks. There are a number of causes, one of which is inadequate curing in field conditions. ODOT is committed to determining if concrete mixtures using some form of internal curing will help reduce the tendency of high-performance deck concrete to crack. Two methods are being investigated.

No matter how much cracking reduction is achieved, there may be other beneficial effects using internal-curing, or self-curing, concrete mixtures. Internal curing using SCA or LWFA will produce a more dense mixture and potentially improve durability due to less absorption of chlorides and carbon dioxide. If cracking is reduced, additional research will be needed to verify additional advantages to using these mixes. Final results of the work on the SCA admixture will be available by the end of 2010, and final results of the work on LWFA also will be available.