Properly designed and constructed concrete pavements can provide 20 to 25 years of initial service life without significant maintenance. In recent years, however, cases of early-age problems and premature deterioration have resulted from use of incompatible materials. Highway professionals have noted instances of early loss of workability (early stiffening), delayed set (retardation), early-age cracking due to excessive shrinkage, and lack of proper air voids. These problems affect long-term performance and even construction productivity.

As concrete mix designs become more complex, the likelihood of incompatibility among materials increases with the number of ingredients added to the mix. The problem is compounded because not much is known about the factors that lead to incompatibility, and tests are lacking to determine the susceptibility of materials combinations to distress mechanisms.

Construction Technology Laboratories, Inc., was tasked with developing reliable tests to identify incompatible combinations that adversely affect fresh and hardened concrete at early ages. The materials include cement, supplementary cementitious materials, chemical admixtures, and aggregates.

"The project is intended to develop practical tests and tools for concrete producers, contractors, and State department of transportation personnel to use in mixture evaluation for adequate air entrainment, setting characteristics, and early-age cracking," says Dr. Colin Lobo, vice president of engineering at The National Ready Mixed Concrete Association and task leader for the project.

Because of the complexity of concrete, predicting incompatibility from tests of the individual materials is impossible. Combinations must be tested using methods that simulate conditions at job sites. Once test procedures are refined, the research team for the incompatibility of concrete materials project will develop protocols and guidelines for evaluation of materials for specific jobs. Some of these protocols are already established standards, while others are in the advanced stages of development.

The results will enable concrete producers and users to have greater confidence in the use of portland cement concrete for pavements and avoid use of marginal concrete for paving. The materials can be prequalified in laboratories to prevent incompatibility problems in the field, costly errors, and construction delays.

Early Stiffening

Loss of workability may be due to false set or early stiffening. Excessive calcium sulfate in the form of plaster in the cement results in false set whereas the uncontrolled early hydration of alluminates in the cementitious materials may cause early stiffening. Continued mixing of the concrete can overcome false set, whereas early stiffening is not reversible. Standard test methods for early stiffening of portland cement often are unreliable and do not incorporate consideration of supplementary cementing materials or chemical admixtures. The poor reliability of the test methods has been attributed to the mixing intensity specified in these test methods.

Early stiffening in the cement phase of concrete leads to loss of workability, as indicated by loss of slump. Loss of workability leads to difficulties in concrete placement and consolidation. When concrete is hard to place, contractors may add additional water, reducing both strength and durability and increasing the potential for shrinkage and cracking. The addition of admixtures (such as high-range water reducers) improves workability without these negative effects but adds to the cost of the concrete. High-range water reducers may retard setting, depending on the amount used. The tendency to early stiffening may be attributed not only to the individual cementitious materials, but also to interactions among the various cementitious materials and the chemical admixtures.

Retarded Concrete

From time to time, some paving projects experience problems with concrete setting being delayed by a few hours to more than 12. Although retardation is not a common phenomenon, the consequence is the inability to perform joint sawing in a timely manner, leading to unplanned cracking.

A common cause for delayed set is the incompatibility between the water reducer and cementitious materials, compounded frequently by cool weather conditions during placement of the concrete. The effect of the admixture dosage on cement hydration for a given placement temperature needs to be considered during the mix design. The admixture dosage needs to be optimized, and any excess that can lead to extended setting and poor development of early-age strength should be avoided.

Setting is identified with a certain stage in the development of the coagulational-crystallization network, while the process of hardening denotes the development of the much stronger and irreversible crystalline structure. Retarders, such as gypsum, and surface active agents, such as calcium lignosulphanate, influence the rate of formation of the coagulational structure and the speed and form of the crystallization of the alluminate hydration products. The retarders may combine with certain cement constituents to form insoluble metal organic complexes, which coat the cement grains and retard hydration.

Concrete Shrinkage

A host of factors can cause premature cracking. Shrinkage can occur in the fresh (plastic) or hardened concrete. The major cause of cracking from plastic shrinkage is thought to be development of tensile stress as water evaporates from the surface of the concrete, leaving the capillaries partially filled and creating a disjoining pressure due to surface tension. As the concrete has not yet hardened, its tensile strength is low, and cracks can develop, especially at the surface where drying is greater.

The risk may be higher in concretes that exhibit early stiffening, as the mix does not remain fluid long enough to allow a layer of bleed water to remain on the surface. Although bleeding is generally thought of as detrimental because it contributes to settlement cracking and weak surfaces, it is of some benefit with relation to the potential for plastic cracking evaporation from the surface cannot occur if it is covered with bleed water.

In the absence of bleeding, loss of water from the concrete surface can occur rapidly if precautions are not taken. These precautions include (1) strict adherence to specifications regarding evaporation rates and cessation of concrete placement if relative humidity is low and temperatures and wind speeds are high, (2) use of fog sprays, (3) use of evaporation retarding agents during and immediately after finishing, and (4) initiation of wet curing as soon as possible after finishing. Some of these precautions may not be practical in large pavement jobs, so careful testing and selection of materials that are least susceptible to plastic shrinkage cracking are important.

Cracking can occur somewhat later in the life of the pavement due to autogenous shrinkage, drying shrinkage, thermal effects, and external loads. Cracking occurs when the maximum principal tensile stress in the concrete exceeds the concrete tensile strength.

Air Void System

In recent years, synthetic air-entraining agents have caused problems in a number of paving projects, including loss of strength. These problems are associated with accumulations (coalescence) of air voids around the aggregate particles. The size and spacing of the air voids in the hardened concrete need to accommodate disruptive forces (stresses) due to freezing. A marginal air void system may result due to incompatibility between certain water reducers and air-entraining agents.

Literature Review

As part of the project's first phase, the team conducted an extensive review of literature pertinent to the three areas of interest: early stiffening and retardation (rheology), excessive shrinkage related to premature cracking, and air void characteristics. The team also sent a request for field experience information to State departments of transportation, FHWA, chapters of the American Concrete Pavement Association, paving contractors, cement companies, admixture suppliers, and consultants.

The rate of response to the informal survey of highway officials and other technical personnel was not good, and much of the information provided was anecdotal. Many field problems are not formally reported nor extensively analyzed, and in many instances corrective measures are put in place through a trial-and-error process. The survey did identify several incidences of problems possibly related to material incompatibility, and plans were developed to obtain materials samples from those projects.

Phase 2

Based on the literature review and survey, it is apparent that many common practices in the construction of highway pavements are less than ideal. Placement during hot weather, short concrete mixing times, transport in nonagitating trucks, excessive or insufficient vibration, and poor timing of control joint sawing all contribute to unsatisfactory performance. Some materials and combinations of materials clearly are more sensitive to these practices than others, such as combinations of cement with Class C fly ash, and cementitious materials with chemical admixtures.

In phase 2 of the study, the overall objective is to develop test methods to identify chemical and physical incompatibility of the materials. Phase 2 also is addressing the effect of external factors such as temperature and specific construction practices since these factors affect the degree of incompatibility.

Procedures to Determine Concrete Workability

Chemical incompatibility mainly originates from the chemistry of the ingredients used to make concrete. Physical incompatibility mostly depends on characteristics such as size, shape, and amount of the aggregates. The study will carry out tests on the paste system to evaluate chemical incompatibility and on the concrete to evaluate physical incompatibility.

Test procedures will include cement paste or mortar rheological measurements (rotational rheometer), penetration test (American Society for Testing and Materials C 191), heat evolution (conduction calorimetry), FHWA's vibrating slope apparatus, consolidation apparatus (new concept), and other supplementary tests.

In order to identify chemical incompatibility, paste microstructure should be similar to that typically found in concrete. Therefore, the team will prepare cement paste using a new high-shear mixing procedure that produces a typical paste. They will test the cement paste in a rotational rheometer to determine the rheological parameters (yield stress and plastic viscosity), setting characteristics, bleeding behavior, and air entrainment at several time intervals.

The results will dictate the combinations of materials to be tested. Concrete mixes will be tested at two temperatures, approximately 23°C (73°F) and 32°C (90°F), to simulate normal and hot weather conditions. The team will measure consolidation, flow, and setting properties of the mixes.

Heat evolution is directly correlated with setting characteristics. An adiabatic cell will be developed to monitor the heat evolution by embedding thermocouples into concrete. The two parameters that provide information on setting behavior are the slope of the heat evolution curve and the time at which the peak is reached. In general, a retarding concrete has a smaller slope compared with a normally setting concrete. The cell may be used to monitor the heat evolution both in the laboratory and the field.

To simulate the paving operation, concrete consolidation will be tested under vibration. Concrete properties will be measured using a vibrating slope apparatus on loan from FHWA and a prototype consolidation apparatus built for this work. The vibrating slope apparatus measures the rate of discharge of concrete from a chute placed at two different slopes. The consolidation apparatus measures the settling time of concrete under vibration.

Identifying Shrinkage Potential

The main factors that contribute to cracking include heat evolution, unrestrained shrinkage from the time of casting, and stiffness, as measured by the modulus of elasticity (mathematical ratio between stress and elongation). A test developed in this study and nondestructive tests suitable for field monitoring will measure the ability of the concrete to resist cracking from flexural and splitting tensile strengths. Measurements of initial and final setting times and compressive strengths will indicate when joints could be cut.

The test program involves a 24-hour period of intensive monitoring of the properties of the concrete immediately after casting. The purpose is to examine the development of the properties of interest at early ages. Such an intensive test program would not be incorporated into an evaluation of the materials and mix designs for actual projects, but it should indicate which measurements are the most critical.

The study will evaluate and refine test procedures such as the Japan Concrete Institute's (JCI) test for unrestrained shrinkage and analytical methods like HIPERPAV and American Concrete Institute 207 to predict potential for shrinkage under varying ambient conditions and different design and construction features.

All tests will be conducted at 32°C (90°F). Unplanned cracking related to shrinkage typically occurs at elevated temperatures. Some specimens will be exposed to dry laboratory air to examine the effects of delayed or neglected curing.

Procedures to Characterize Air Voids in Fresh Concrete

The incompatibility of concrete materials study will evaluate a suite of promising tests designed to characterize the system of air in fresh concrete and hardened concrete at early ages. The work will test a wide series of concrete mixtures with a range of air void characteristics, using a variety of admixtures, materials, and mixing conditions to simulate field conditions.

Phase 1 indicated that unstable air void systems are associated with the use of certain air-entraining admixtures, incompatibility of materials, construction practices, and field conditions (e.g., temperature and consolidation). Unstable and nonuniform air void systems may lead to unacceptable reduction in strength and durability. Selection of air-entraining admixtures is based primarily on the recent experience of the State highway departments in Delaware, Michigan, New Jersey, and South Dakota.

The testing program will evaluate the effect of air-entraining admixtures on the stability of air void systems prior to use in concrete and then will test the fresh concrete and hardened concrete at early ages. Test procedures evaluated, refined, or developed will include air void stability; foam index and drainage; air void analyzer; air content of freshly mixed concrete by pressure meter; unit weight, yield, and air content; and microscopic determination of parameters of the air void system in hardened concrete.

The recently introduced air void stability test is based on measurement of the characteristics of the air bubbles released from a paste sample. The size distribution and the spacing factor of the air bubbles are determined, and the air content and the spacing factor are deduced.

Summary of Phase 2 Testing

For each of the three areas of concern under study, problematic materials obtained from the field (from the 2000 and 2001 construction seasons) will be tested, or testing will be performed using simulated problematic materials with constituents or dosages known to cause problems. The focus of the incompatibility study is to develop test procedures that will identify problematic combinations of materials. It is not to study materials that cause problems. As such, the selection of materials will provide a range of conditions to verify, calibrate, and validate the various test procedures, under a range of test conditions such as temperature.

For each of the test procedures, the analysis phase will include the comparison of results for the problematic mixtures and the control mixtures. The research team will perform appropriate statistical tests to determine the ability of the test procedures to discriminate between normal concretes and problematic concretes.

The test procedures to be developed need to be rapid and simple so that they may be used both in the laboratory during the mix design phase as well as in the field during the construction phase.

References

1. Japan Concrete Institute Technical Committee on Autogenous Shrinkage of Concrete, Committee Report. Autogenous Shrinkage of Concrete, Ei-ichi Tazawa, Ed. London and New York: E & FN Spon, 1999.

2. Lea, F. M. The Chemistry of Cement and Concrete, 3rd edition, Chemical Publishing Company, Inc.

3. Magura, D.D. Air Void Analyzer Evaluation, FHWA Report No. FHWA-SA-96-062, Washington, DC: Federal Highway Administration, 1996.

4. Popovics, S. Fundamentals of Portland Cement Concrete: A Quantitative Approach, J ohn Wiley, 1982.

5. Wong, G. S., et al. Portland Cement Concrete Rheology and Workability—Final Report, FHWA Report No. FHWA-RD-00-025, Washington, DC: Federal Highway Administration, April 2001.

Dr. Colin Lobo, vice president of engineering at The National Ready Mixed Concrete Association of Silver Spring, MD, serves as task leader [manager] for the project. The Construction Technology Laboratories, Inc., (CTL) in Skokie, IL, is conducting the study. Dr. Shiraz Tayabji serves as the principal investigator for the CTL team. Other CTL key staff involved in the study include Dr. Rachel Detwiler and Dr. Mohamed Nagi. The National Institute of Standards and Technology (NIST) of Gaithersburg, MD, serves as the subcontractor to CTL and is responsible for the portion of the study related to workability. Dr. Chiara F. Ferraris serves as NIST's project manager.

Shiraz Tayabji, regional manager, Construction Technology Laboratories, Inc., with more than 25 years of experience in concrete pavement technology, has extensive experience with concrete pavement design, construction, and rehabilitation and with concrete as a construction material. Tayabji holds a B.Sc. degree in civil engineering from the University of East Africa (Nairobi, Kenya) and an M.S. and Ph.D. in civil engineering from the University of Illinois at Urbana. He is registered as a professional engineer in Delaware, Illinois, Maryland, New Jersey, Pennsylvania, and Virginia. He was chairman of the Transportation Research Board's Committee A2B04 on Pavement Rehabilitation (1997 to 2002), member and former chairman of the American Concrete Institute's Committee 325 on Concrete Pavements, and member of several committees of the American Society for Testing and Materials, including Committee D04.39NDT of Pavements, of which he is the vice chair. He serves as the president of the International Society for Concrete Pavements. He has authored or co-authored more than 40 papers published in professional journals and presented at national conferences.