U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590
Federal Highway Administration Research and Technology
Coordinating, Developing, and Delivering Highway Transportation Innovations
|This report is an archived publication and may contain dated technical, contact, and link information|
Publication Number: FHWA-RD-02-099
Date: January 2005
Many of the problems associated with curing portland cement concrete pavements (PCCP) can be anticipated from knowledge of concrete materials, mixture proportions, and early age properties along with knowledge of probable climatic conditions during placing and in the several days after placing. Figure 1 summarizes major action items.
Properties of cementitious materials (cement and pozzolan) and chemical admixtures are important to consider in anticipating problems with curing. Variation in aggregate properties is probably less important (except possibly for lightweight aggregate, which is not commonly used in paving), although there may be subtle effects. None of the properties described in this section necessarily requires specific action when values deviate from the acceptable limits, but being aware of effects may help anticipate a problem.
The cement properties that are most important in determining curing requirements are strength gain, time of setting, and fineness. Most paving is made with types I, II, or I/II portland cement; guidance is found in publications from the American Association of State Highway and Transportation Officials (AASHTO M 85) (6) and the American Society for Testing and Materials (ASTM C 150).(7) Type V is used where soils are high in sulfate. The strength gain rates among types I, II, and I/II tend to all converge within a given geographic area, so the user really has very little choice in this property. Blended cements specified by AASHTO 240(8) and ASTM C 595(9) have strength-gain behaviors that are essentially equivalent to the M 85/C 150 types. ASTM C 1157, which has no AASHTO equivalent, is a general specification for hydraulic cement (portland and blended cements).(10) Requirements are based on performance properties, with little or no prescriptive specifications. Strength development of the various grades is essentially equivalent to C 150 types (e.g., type O is approximately equivalent to C 150 type I in performance).
The length of required curing of a concrete structure is sometimes directly tied to the strength-gain rate of the cementitious materials. In most guidance, the length of curing is either a prescribed amount of time or the time required to achieve a given strength of the concrete. The strength-gain rate of cementitious materials can affect the strength gain of concrete, but other variables are also involved, most notably the water-cement ratio. The strength-gain rate of cement also affects the amount of cement necessary in a concrete mixture to obtain a given strength in a required time interval. High cement content can result in large amounts of long-term drying shrinkage, particularly if the cement is well hydrated. Hydrated cement paste contributes strongly to drying shrinkage.
Mortar strengths of about 24 MPa at 3 days and 31 MPa at 7 days are most common for types I, II, and I/II cements. Strengths for type V cements are typically about 21 MPa at 3 days and 28 MPa at 7 days. Strengths of available cements can range from about 3.5 MPa less than these values to about 7 MPa higher, but these are less common. Within a geographical area, cement strengths among producers tend to converge on similar values. Some specifications that are based on fixed-time curing requirements cite the need for extra curing time of concrete made with type II cement. Before 1980, type II cement was usually made with a composition that gained strength at a significantly slower rate than type I cement. Typical 3-day mortar strengths were about 14 MPa. This is now rarely true except in custom-made cements, usually produced for mass concrete applications. Except when the optional heat of hydration requirement is cited, the only practical distinction between type I and type II cement has to do with sulfate resistance.
The principal direct effect of fineness on curing has to do with its effect on bleeding and, in concretes with a very low water-cement (w/c) ratio, on development of internal desiccation due to early consumption of mixing water. Modest bleeding tends to buffer the effects of early-age drying and help prevent PSC. Since finer cements tend to hydrate faster, they also generate more heat and potentially cause temperature gradients in the concrete depending on ambient conditions and curing procedures employed (see discussion of HIPERPAVTM in chapter 4.
Blaine fineness values for portland cements tend to range between 325 and 375 square meters per kilogram (m2/kg). Values higher than 400 m2/kg may indicate a problem with development of too little bleed water when drying conditions are high, and/or internal desiccation if water-cement ratios are less than about 0.40. Pozzolans are sometimes very fine and can contribute significantly to this problem. Silica fume is particularly noted for this property, but is rarely used in slip-form paving because of workability and cost issues. Slag can also be fine enough, particularly in grade 120, to have a detectable effect on water demand. Fly ashes are typically not so fine as to be problematic, although ultrafine products that may have some noticeable effect are being introduced into the market.
Class F pozzolan (AASHTO M 295,(11) ASTM C 618(12)) was the major type of pozzolan used in paving until recent years. The major effect of this class of pozzolan is that setting times are usually delayed by 1 to several hours, and strength gain can be retarded relative to concrete made without pozzolan. The major effect of delayed setting time is that the optimal time for applying final curing is also delayed, hence more time for occurrence of PSC. Slow strength gain can result in prolonged curing-time requirements, unless concrete temperatures are warm. These properties have typically limited the amount used in paving to about 20, by mass of total cementitious materials.
In the last 10 years, class C fly ash has become an abundant product in concrete construction. This class of fly ash is often popular in paving concrete because strength gain is higher than with class F pozzolan; however, setting times may be delayed by times similar to class F. Some of these materials contain chemical phases that hydrate very rapidly upon contact with water, and may tend to tie up the water in the concrete within a few minutes of mixing. This property normally causes some early stiffening.
Class N pozzolan is not commonly available, but some of the products available in the past have been very finely divided, giving good early strengths but seriously affecting water demand.
Water-reducing admixtures (WRAs) can have two effects on curing. One effect is that they facilitate reducing the w/c rating which impacts curing requirements as discussed below. The other effect involves the occasional case of cement-admixture interaction. Occasionally certain cements and certain WRAs interact badly, resulting in very rapid early hydration of the cement. This may result in rapid consumption of a significant amount of the free mixing water and substantially reduce or eliminate bleeding. Under certain drying conditions (described below), this will make the concrete more susceptible to plastic shrinkage cracking.
WRAs are sometimes advertised as being helpful in reducing drying shrinkage cracking. This effects stems from the fact that if the water-cementitious materials (w/cm) ratios are low enough, most of the mixing water is either chemically bound or tightly bound as surface water in gel pores, and not available to evaporate and cause shrinkage. Unfortunately, when taken to extremes, this overconsumption of mixing water creates internal desiccation, which is similar in effects to atmospheric drying.
The w/cm ratio, total cementitious materials content, and percentage of cement replaced by pozzolan are the three most important mixture-design variables affecting curing requirements, as discussed in the following paragraphs.
The amount of bleeding is highly dependent on the w/cm ratio. Small to moderate bleeding is effective in buffering excessive drying when concrete is in the plastic state and susceptible to PSC. Excessive bleeding can be detrimental because it tends to result in deposition of a layer of weak material on the surface of the concrete. The w/cm ratio of paving concrete mixtures is rarely high enough to result in this problem.
The relationship between bleed rate and w/c ratio is approximately linear. The empirically developed equation shown in figure 2 approximately relates the average rate of bleeding, in kilograms per square meter per hour (kg/m2/h), to the w/cm ratio.(13) T is pavement thickness in centimeters.
Figure 2. Equation. Bleeding rate from water-cement ratio.
Paving concretes tend to have w/cm ratio between 0.38 and 0.48. For 30-cm thick pavement, bleeding would then range from about 0.13 to 0.28 kg/m2/h. These are lower average bleeding rates than found in more general-use concretes, which range from about 0.5 to 1.5 kg/m2/h. The result is that paving concretes are more susceptible to losing excessive or unsafe amounts of bleed water to evaporation. ACI 308(4) states that drying conditions of less than 0.5 kg/m2/h represent a mild threat to most concrete. A safer upper limit for paving would be about 0.3 kg/m2/h. Recommendations on how to evaluate danger of excessive drying are described below.
The cementitious materials content of paving concrete typically ranges from 325 to 385 kg/m3. It is relatively common practice to compensate for slow strength gain, particularly when using flexural strength as the design property, by adding more cement. High cementitious materials contents, particularly if cementitious materials are very finely divided, tend to contribute to a reduced amount of bleeding.
A major effect of high cementitious materials contents is long-term drying shrinkage, even if the concrete is well cured. Since hydration ties up free water and results in a volume decrease, hydrated cement paste is the major component of concrete causing drying shrinkage. Long-term drying shrinkage is almost totally dependent on the fraction of hydrated cement in the concrete.
High cement contents can also contribute significant heat of hydration, particularly if the concrete is placed several hours before the hottest part of the day. Portland cement typically reaches its most intense hydration period (and hence heating) 2-4 hours after time of initial setting. Given that time of setting is typically 2-4 hours after mixing, the period of peak heat evolution is approximately 4-8 hours after placing. Temperaturerelated problems begin after the concrete reaches peak temperature. As the concrete begins to cool, temperature stresses go from compressive to tensile (a situation where concrete is relatively weak).
In general, cement contents as low as compatible with adequate strength gain and durability are beneficial in reducing effects of drying shrinkage and thermal heating effects.
Retardation of early strength gain is strongly related to the amount of pozzolan replacement of portland cement, particularly if Class F pozzolan is used. Class C pozzolan tends to make a strength contribution at an earlier age than Class F pozzolan. The AASHTO Guide Specifications for Highway Construction recommends 3 days' extra curing if substantial (greater than 10) replacement amounts of pozzolan are used.(14) However, preconstruction mixture verification studies should be used to verify this effect. Calculations based on maturity concepts can help anticipate required curing time. The strength gain rate of pozzolan concretes reported to be more temperature sensitive than pure PPCs.(15) If the temperature is expected to be in the 5-15 °C range, then some preliminary exploration of strength gain using maturity calculations may help quantify potential delays in strength gain. See chapter 5 for a discussion of the maturity method.
It is important to determine the bleeding behavior of concrete intended for use in paving because this indicates the amount of water that can be safely lost to evaporation. Plotting bleeding over time allows one to identify potentially critical intervals during the bleeding period, which occurs between placing and initial time of setting. Concrete ceases to bleed after the time of initial setting.
Bleeding of job concrete is easily measured during mixture verification testing. The basic method is described in AASHTO T 158(2) and ASTM C 232,(16) but several modifications make the data more useful for the present purposes. The standard test method stipulates using a unit-weight bucket as a test apparatus.
The procedure calls for fabricating a test specimen from job concrete using the same procedures as used in making strength cylinders (AASHTO T 23,(17) ASTM C 31(18)). Make the specimen about the same height as the thickness of the pavement. If the pavement is to be placed on a porous base, then a layer of sand in the bottom of the mold can be used to simulate this drainage potential. Control evaporative losses by keeping the container covered except when taking measurements. About every 30 minutes between fabrication and time of initial setting, tilt the cylinder slightly to one side and let the bleed water collect for about 5 minutes. Draw off the bleed water with a syringe or medicine dropper and measure, either by volume using a small graduated cylinder (5-10 milliliters (mL)) or by weighing. Making a slight depression on the downhill side of the specimen surface will facilitate collecting and drawing off the bleed water.
Calculate the average bleed rate over each time interval using the equation shown in figure 3.
Figure 3. Equation. Time-averaged bleed rate.
V = the amount of bleed water (in kg)
A = the surface area of the specimen (m2)
t = time (h)
Units of bleeding are kg/m2/h for that specific thickness of pavement
Plotting the amount of bleed during each time interval gives a time profile of bleeding. Periods when bleeding is less than 0.3 kg/m2/h may be potentially critical periods. However, the level of criticality depends on drying conditions. Figure 4, using data found in volume II, (13) shows such a plot for a paving mixture.
Figure 4. Graph. Plot of bleed water formation v. time for a typical paving mixture.
For this concrete, bleeding rates are low during the first hour and again immediately before time of setting, which occurred at 5 hours. Even at the peak, the bleeding rate was less than the 0.5 kg/m2/h cited in ACI 308(4) as a limit below which caution should be exercised. Additional information on interpreting such data and accounting for drying conditions is found later in this chapter.
Time of initial setting is an important property in paving practice because it indicates bleeding is complete and final curing procedures can be initiated. This detail is not usually part of standard guidance on the start of final curing. Application of final curing is usually directed to start when final finishing is complete and the surface sheen is gone. In conventional concreting, final finishing is typically not executed until about the time of initial setting. In slip-form paving, final finishing is usually completed within a few minutes of placing the concrete, well before the time of initial setting and the end of the bleeding period. If bleeding rates are low relative to evaporation rates, then loss of surface sheen will appear rather soon after placing, suggesting that final curing should be initiated even though bleeding is continuing.
Starting final curing before the time of initial setting can lead to several problems. With water and sheet curing, the surface can be damaged due to lack of strength. Water tends to wash out fines, and sheet materials can mar the surface. With curing compounds, continued bleeding under an applied membrane can lead either to poor membrane formation (and loss of critical mixing water) or to spalling of surface mortar. See chapter 4 for a discussion of this phenomenon.
Time of setting is measured as described in AASHTO T 197(3) and ASTM C 403,(19) and is conveniently done during mixture verification work prior to the start of construction. The time of setting is strongly affected by the concrete temperature, and therefore the field time of setting will differ from the laboratory-determined time if the two temperatures differ. This is important in field application, since in-place concrete temperatures can differ significantly from laboratory concrete temperatures, and the effect can be substantial. Laboratory values can be adjusted for actual concrete temperature using the following equation.(13)
Figure 5. Equation. Time of setting-adjustment for concrete temperature.
TOS = time of setting at temperature of in-place concrete, same units as in standard test
TOSStdTemp = time of setting under standard conditions, any units
CT = temperature of in-place concrete, K
StdTemp = temperature of concrete during laboratory test, K
R = constant
The constant, R, can be determined empirically, but a value of 5,000 Kelvins (K) works well. This equation can be programmed into a spreadsheet to simplify the calculation for use in exploratory work.
It is important to be able to anticipate likely drying conditions immediately after placing to determine whether water in excess of bleed water is likely to be lost, making the concrete vulnerable to PSC.
A nomograph ACI has been found to be reasonably accurate in estimating drying conditions for inputs of wind velocity (0.5 m above the concrete surface), concrete temperature, air temperature, and relative humidity of the air above the concrete.
The range of probable drying conditions can be forecast for a given locale based on typical range of weather conditions and projected concrete temperatures. Drying rates of greater than 0.3 kg/m2/h may present a problem for paving concrete, depending on bleeding rates during the same time (see below).
The information in the nomograph can be represented by the equation shown in figure 6. This equation can be programmed into a spreadsheet to simplify the calculation. The nomograph from ACI 308 is shown in figure 7.(4)
Figure 6. Equation. Evaporation rate of bleed water-effect of environmental conditions.
ER = evaporation rate (kg/m2/h)
WS = the wind speed (m/s)
CT = concrete temperature (°C)
AT = air temperature (°C)
RH = relative humidity (%)
Figure 7. Chart. Evaporation rate nomograph from ACI 308.(4)
It is very instructive to explore the effects of ranges of environmental conditions expected in a given construction location on the evaporation of the bleed water. This information, along with bleeding data and time of setting, allow the engineer to anticipate critical conditions. Wind and concrete temperature are usually found to be the most critical variables. The temperature of freshly placed concrete is a property over which a producer has some control by adjusting the temperature of concrete-making materials. ACI 305 R contains equations that relate the temperature of materials to temperature of concrete.(5)
Anticipating thermal stress conditions on the job can be complicated because of the many variables involved. The Federal Highway Administration (FHWA) has developed a software program called HIPERPAVTM that allows the user to enter plausible data on concrete and site conditions and get thermal-stress output back, in the form of warnings on critical times after placing when cracks may develop (see chapter 4). This program also includes an evaporation-rate calculator similar to the results obtained from the equation shown in figure 6.
Comparing bleeding behavior with probable drying conditions will identify potential critical points during construction. The time of initial setting indicates the end of this critical period. Figure 8 shows the cumulative bleed rate calculated from the data shown in figure 4 plotted along with the cumulative evaporation rate, assuming a constant evaporation rate of 0.30 kg/m2/h.
In this example, evaporation rates exceed bleeding rates for the first hour after placing and again after about 3.5 hours. The time of setting is about 5.2 hours. These two periods represent critical periods from a PSC perspective. Sometimes concrete will endure the first critical period because the mixture is plastic enough to adjust to evaporative losses by simply shrinking into a thinner placement. However, the period after about 3.5 hours may result in cracking because the concrete may have developed some stiffness at this point, and cannot adjust to the loss of water by simply reducing volume.
Figure 8. Graph. Plot of cumulative bleed and cumulative evaporation v. time.
Standard guidance recommends that when evaporation exceeds bleeding, something must be done to reduce evaporation rates. Standard remedies include use of fogging and wind breaks. Neither of these methods is particularly useful for large paving projects. Three practices are potentially useful in paving large areas. One is to shift paving operations to a time of day when the drying conditions are less severe. Nighttime placement is often attractive because relative humidity is usually higher than during the day.
Another effective option is to reduce the temperature of the concrete at placing. This is a very strong variable affecting evaporation rates. ACI 305 R gives guidance on calculating placing temperature (also discussed in chapter 3). This calculation can be used to explore the amount of benefit expected from cooling concrete ingredients. For example, figure 9 shows the effect of reducing the concrete temperature by 5 °C (using data from the nomograph in figure 3) on evaporation rates. Evaporation rates become less critical as a result of this adjustment.
Figure 9. Graph. Effect of reducing concrete placing temperature from 30 °C to 25 °C.
Still another approach is to use evaporation reducers. Evaporation reducers can reduce evaporation rates by as much as 65 percent. (13) Figure 10 shows the effect of a 50 percent reduction in evaporation, using the data shown in figure 4. The cumulative effect is similar to the reducing concrete placing temperature by 5 °C. Currently, there are no test methods or specifications for evaporation reducers, and the user must rely on the manufacturer's guidance.
Figure 10. Graph. Effect of reducing evaporation by 50 percent by using an evaporation reducer.
A limited laboratory investigation demonstrated that evaporation reducers can reduce evaporation rates by an amount ranging from 0 to 65 percent. Evaporation reducers must be reapplied if the time of setting is extended and drying rates are high. The approximate application interval is discussed in chapter 3.
In conclusion, reducing concrete placing temperature and use of evaporation reducers can have a relatively strong influence on potential for plastic shrinkage cracking.