Self-Consolidating Concrete for Cast-in-Place Bridge Components (2016)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

B-1 Proposed Guidelines for Use of Self-Consolidating Concrete in Cast-in-Place Bridge Components These proposed guidelines are the recommendations of the NCHRP Project 18-16 staff at the University of Nebraska-Lincoln. These guidelines have not been approved by NCHRP or any AASHTO committee nor formally accepted for adoption by AASHTO. A t t A c h m e n t B

B-2 c o n t e n t s B-4 Introduction B-4 B.1 Guidelines for Selection of Constituent Materials B-4 B.1.1 General B-4 B.1.2 Cement, Supplementary Cementitious Materials, and Fillers B-4 B.1.3 Fine Aggregates and Coarse Aggregates B-4 B.1.4 Chemical Admixtures B-4 B.1.4.1 High-Range Water-Reducing Admixtures B-5 B.1.4.2 Air-Entraining Admixtures B-5 B.1.4.3 Viscosity-Modifying Admixtures B-5 B.1.4.4 Workability Retaining Admixtures B-5 B.2 Guidelines for Mix Proportioning B-5 B.2.1 General B-5 B.2.2 Workability Targets B-6 B.2.3 Proportioning Approach B-8 B.2.4 Quality Assurance B-8 B.3 Guidelines for Testing Fresh Concrete B-8 B.3.1 General B-8 B.3.2 Rheology B-9 B.3.3 Filling Ability B-9 B.3.4 Passing Ability B-10 B.3.5 Static Stability B-10 B.3.6 Dynamic Stability B-10 B.3.7 Heat of Hydration B-10 B.3.8 Pumpability B-11 B.3.9 Time of Setting B-11 B.3.10 Workability Retention B-11 B.3.11 Formwork Pressure B-11 B.4 Guidelines for Hardened Properties B-11 B.4.1 General B-11 B.4.2 Compressive Strength B-12 B.4.3 Modulus of Elasticity B-12 B.4.4 Splitting Tensile Strength B-12 B.4.5 Modulus of Rupture B-12 B.4.6 Bond to Deformed Reinforcing Steel Bars B-12 B.4.7 Shear Resistance B-13 B.4.8 Drying Shrinkage B-13 B.4.9 Restrained Shrinkage B-13 B.4.10 Creep B-13 B.4.11 Durability Properties

B-3 B-14 B.5 Guidelines for Production and Construction B-14 B.5.1 General B-14 B.5.2 Quality Control of Constituent Materials B-14 B.5.3 Mixing Procedures B-14 B.5.4 Plant Quality Assurance B-14 B.5.5 Transportation Procedures B-15 B.5.6 Site Quality Assurance B-15 B.5.7 Placement Methods B-16 B.5.8 Formwork Considerations B-17 B.5.9 Finishing Techniques B-17 B.5.10 Curing Methods

B-4 Introduction Self-consolidating concrete (SCC) is highly flowable, non-segregating concrete that can spread into place, fill the formwork, and encapsulate the reinforcement without any mechanical consolidation (ACI 237R-07). The use of SCC in cast-in-place bridge construction is limited due to the lack of design and construction guidelines and concerns about certain design and construction issues that are perceived to influence the structural integrity of the bridge system. Therefore, these guidelines were developed to address the factors that signifi- cantly influence the design, constructability, and performance of cast-in-place concrete bridge components using SCC. These guidelines provide highway agencies with the information nec- essary for considering cast-in-place SCC to expedite construc- tion and yield economic and other benefits. B.1 Guidelines for Selection of Constituent Materials B.1.1 General The proper and uniform selection of constituent materials is essential to ensure the satisfactory performance of SCC in both fresh and hardened conditions. All constituent materials shall follow the requirements of AASHTO LRFD design speci- fications (2014) and construction specifications (2010). When requirements are not available, engineering authorization is needed for incorporating any constituent materials. Changes in materials or their proportions shall be monitored continu- ally to avoid any adverse effects on the performance of SCC. B.1.2 Cement, Supplementary Cementitious Materials, and Fillers All portland cements that conform to the requirements of AASHTO M 85 can be used for the production of SCC for cast- in-place bridge components. Blended hydraulic cements that conform to the requirements of AASHTO M 240 can be also used. Types IP (portland-pozzolan cement), IS (portland blast- furnace slag cement), and PLC (portland-limestone cement) can be used. When types I, II or III cements are used, replace- ments with supplementary cementitious materials (SCMs) and fillers are recommended as they enhance the fresh and hardened SCC properties. Fly ash shall conform to the requirements of AASHTO M 295, ground granulated blast-furnace slag (GGBFS) shall conform to the requirements of AASHTO M 302, and silica fume shall conform to the requirements of AASHTO M 307. Limestone powder approved for use in concrete with average particle size of 11 µm or less can be also used as a filler. For cast- in-place bridge components with specified compressive strength from 4 ksi–6 ksi, SCC with total cementitious materials rang- ing from 658 lb/yd3–797 lb/yd3 is recommended. Cement replacements of 25% with Class C fly ash, 25% Class F fly ash, 30% GGBFS, or combination of 20% Class F fly ash in addi- tion to 15% limestone powder are recommended. B.1.3 Fine Aggregates and Coarse Aggregates Fine aggregates shall conform to the requirements of AASHTO M 6 and coarse aggregates shall conform to the requirements of AASHTO M 80. Well-graded combined aggregates are recommended for the production of SCC for cast-in-place bridge components. Natural gravel or crushed stone can be used as coarse aggregate, while natural or manu- factured sand can be used as fine aggregate. Coarse aggregate with nominal maximum size of aggregate (NMSA) greater than ¾ in. is not recommended for use in SCC. NMSA of coarse aggregate shall be determined based on the geometric characteristics of the component and its reinforcement spac- ing. The moisture content, water absorption, and gradation of the aggregate shall be continually monitored to ensure the con- sistency of SCC production and performance. If any changes in aggregate source or properties are observed, a field trial should be mandatory for determining the suitability of that aggregate for the project. B.1.4 Chemical Admixtures Chemical admixtures are mainly used in SCC for cast-in- place bridge components to reduce water content, provide air entrainment, improve viscosity, and enhance workability reten- tion. In special circumstances, other chemical admixtures are used to accelerate strength development, retard setting time, reduce drying shrinkage, and protect against reinforcement corrosion. The performance of chemical admixtures depends on the types and proportions of constituent materials, tem- perature, and compatibility among different admixtures. Trial batches using the plant conditions and materials utilized in production are needed to evaluate the performance of chemi- cal admixtures. Also, when no standard specifications exist, admixtures supplier should be consulted to ensure an admix- ture’s suitability for the application. B.1.4.1 High-Range Water-Reducing Admixtures The use of high-range water-reducing admixtures (HRWRAs) Type F (water reducing, high range) or G (water reducing, high range, and retarding) that conform to the requirements of ASTM C 494 or ASTM C 1017 is necessary to achieve the required flowability of fresh SCC for cast-in-place bridge construction. An HRWRA can be combined with Type A (water reducing), D (water reducing and retarding),

B-5 or E (water reducing and accelerating) in different dosages to achieve the target flowability. B.1.4.2 Air-Entraining Admixtures Air-entraining admixtures that conform to the require- ments of AASHTO M 154 are commonly used in cast-in- place bridge construction to generate an air void system that enhances the durability of bridge components especially with respect to their freeze and thaw resistance B.1.4.3 Viscosity-Modifying Admixtures Viscosity-modifying admixtures (VMAs) can be used in SCC to improve its stability especially when a large aggre- gate (size ¾ in.) and/or a high water/cementitious material ratio (more than 0.4) is used. A VMA should be used only to enhance the stability of SCC and not to correct the perfor- mance of a poorly designed SCC that is already segregating. Also, large dosages of a VMA may negatively affect the flow- ability of SCC, which may result in a higher demand for an HRWRA to achieve the required flowability. B.1.4.4 Workability Retaining Admixtures Workability retaining admixtures (WRAs) are used in SCC to maintain its fresh characteristics throughout the trans- porting, placing and finishing operations without adversely affecting its time of setting and hardened properties. A WRA should be added to the mixture at the plant and may reduce the demand for an HRWRA. B.2 Guidelines for Mix Proportioning B.2.1 General For proportioning SCC, the target values/ranges for SCC properties in both fresh and hardened conditions need to be identified. Workability target values/ranges are iden- tified based on the geometric characteristics of the cast- in-place bridge component as well as the production and placement conditions. Target values/ranges of hardened properties, including mechanical, visco-elastic, and durabil- ity properties, are usually identified by the bridge/materials engineer in the project specifications. The properties of the available and approved constituent materials also play an important role in proportioning SCC. Figure B-1 shows the general steps of proportioning SCC mix. In the follow- ing sections, the common target values/ranges are sum- marized and the approach used for mix proportioning is discussed. B.2.2 Workability Targets The workability targets are presented in terms of the three main performance properties of fresh SCC: filling ability (i.e., fluidity or deformability), passing ability, and stability (i.e., segregation resistance). Two classes are defined for each property as shown in Table B-1 for simplification. These definitions are based on the literature (EFNARC, 2005; ACI, 2007, and Daczko, 2012). To determine which workability target value/range applies to a specific bridge component, the decision tree shown in Figure B-2 is used. This decision tree provides Figure B-1. General steps of proportioning SCC mix for cast-in-place bridge components.

B-6 guidelines on workability targets based on the geometric characteristics of the bridge component. The three-digit identification shown at the bottom of the tree represents the target workability with respect to filling ability (FA), segrega- tion resistance (SR), and passing ability (PA) classes respec- tively. For example, 111 means FA1, SR1, and PA1. Table B-2 shows examples of common substructure and superstruc- ture bridge components, their geometric characteristics, and the corresponding target workability determined using these guidelines. Also, Table B-3 shows quantitative guidelines for defining the “low” and “high” values of each of the geomet- ric characteristics (EFNARC, 2005; and Daczko, 2012). The selected target workability classes should be revised to consider production and placement conditions. For example, lower FA and high SR are needed when high energy mixing and placement methods are used. The final target workability is used in proportioning SCC mix as presented in the next section. B.2.3 Proportioning Approach Several approaches can be used in proportioning SCC mixes (Okamura and Ozawa, 1995; EFNARC, 2002; Bui, Akkaya, and Shah, 2002; PCI 2003; GRACE, 2005; ACI 237, 2007; Koehler and Fowler, 2007; Domone, 2009; Kheder and Jadiri, 2010). The International Center for Aggregates Research (ICAR) mixture proportioning procedure devel- oped by Koehler and Fowler (2007) is recommended for cast-in-place SCC as this procedure considers the effect of aggregate gradation, shape, and angularity in mix pro- portioning. In addition, this procedure only requires con- ducting simple and standard tests to identify necessary Figure B-2. Decision tree used to determine workability targets. Workability Property Class Value/Range Application Filling Ability (FA) FA1 22 in. ≤ Slump Flow < 26 in. Simple Sections FA2 26 in. ≤ Slump Flow ≤ 30 in. Complex Sections or high formed surface quality Segregation Resistance (SR) SR1 10% < Column Segregation ≤ 15% 0.5 in. < Penetration ≤ 1 in. VSI = 1 Short or shallow components SR2 Column Segregation ≤ 10% Penetration ≤ 0.5 in. VSI = 0 Long or deep components Passing Ability (PA) PA1 80% > Filling Capacity ≥ 70% 2 in. < J-Ring ∆D ≤ 4 in. 0.6 in. < J-Ring ∆H ≤ 0.8 in. Wide spacing between reinforcing bars PA2 Filling Capacity ≥ 80% J-Ring ∆D ≤ 2 in. J-Ring ∆H ≤ 0.6 in. Narrow spacing between reinforcing bars Table B-1. Classes of SCC workability properties and their definitions for cast-in-place bridge components.

B-7 parameters. This method can be used in combination with ACI 211.1-91 Table 6.3.3 to estimate mixing water require- ments for different NMSA; and ACI 237R-07 Tables 4.1 and 4.2 to verify that the powder content, powder volume, and aggregate volume are within the recommended ranges for SCC. Below are the steps followed to proportion an SCC mix for a given workability class (e.g., 212) using the rec- ommended procedure. 1. Select the NMSA based on the PA and SR classes as follows: a. NMSA is ¾ in. for PA1 and SR1 b. NMSA is ½ in. for PA2 and SR1 or PA1 and SR2 c. NMSA is 3⁄8 in. for PA2 and SR2 In addition, NMSA should not exceed 1⁄5 of the narrowest component dimension and ½ of the smallest clear spacing between bars. Component Category Bridge Component Component Geometric Characteristics SCC Workability Targets Length Depth Thickness Shape Intricacy Formed Surface Quality Level of Reinforce- ment Proposed Property Classes* ID Su bs tr uc tu re Footing Low Low High Low Low Low FA1, SR1, PA1 111 Pile Cap Low Low High Low Low High FA1, SR1, PA2 112 Wing Wall Low Low High Low Low Low FA1, SR1, PA1 111 Abutment Wall High High High Low Low Low FA1, SR2, PA1 121 Pier Wall Low High High High High Low FA2, SR2, PA1 221 Pier Column Low High High Low High High FA2, SR2, PA2 222 Strut or Tie Low Low High Low High Low FA2, SR1, PA1 211 Pier Cap Low Low High Low High High FA2,SR1, PA2 212 Su pe rs tr uc tu re Box Girder High Low Low High High High FA2, SR2, PA2 222 Stringer Low Low High Low Low High FA1, SR1, PA2 112 Floor Beam Low Low High Low Low Low FA1, SR1, PA1 111 Girder High Low Low Low High High FA2,SR2, PA2 222 Arch High High High Low High Low FA2, SR2, PA1 221 * For deep/long components, SR1 could be acceptable if free-fall height/free-travel distance are controlled (e.g., tremie pipe). Table B-2. Workability targets for example of cast-in-place bridge components. Component Geometric Characteristic Class Value/Definition Length Low ≤ 33 ft High > 33 ft Depth Low ≤ 16 ft High > 16 ft Thickness Low ≤ 8 in. High > 8 in. Shape Intricacy Low Concrete flows in a single direction High Concrete flow around corners and cutouts Formed Surface Quality Low Unexposed to the traveling public High Exposed to the traveling public Level of Reinforcement Low Large spacing between bars (≥ 3 in.) High Small spacing between bars (< 3 in.) Table B-3. Definitions of the geometric characteristics of cast-in-place bridge components.

B-8 2. Determine the optimal gradation of the combined coarse and fine aggregates that results in the highest density. The ratio of sand to aggregate (S/A) can be changed to achieve optimal gradation using the 0.45 power curve. As a starting point, use S/A = 0.45 and change in the range from 0.4 to 0.5 until the optimal gradation is achieved. In the case of gap grading aggregate, more than two types of coarse aggregates can be used. Another method to determine the optimal S/A ratio is the use of predefined limits for the minimum and maximum percent retained on each sieve. 3. Determine the shape and angularity rating (RS-A) of the blended aggregate using the guidelines published in Table 6 of the ICAR 108-1 report (Koehler and Fowler, 2007). 4. Calculate the minimum percentage of paste that achieves the flowability of the mixture made of the blended aggregate. This is calculated as follows: Min Vpaste = 1 - (1 - %voids)(1 - %spacing), where, %voids is the per- centage of voids calculated using the dry-rodded unit weight of the blended aggregate, S/A ratio, and specific gravity of fine and coarse aggregates. The %spacing is calculated based on the shape and angularity rating as follows: 8 + 2(RS-A - 1). Selected paste volume percent- age shall be within the range (34% to 40%) recommended by ACI 237R-07. 5. Subtract the target air content (e.g., 6%) to get the volume of powder and water. Powder content can be estimated based on the target FA class (650–750 lb/cy for FA1, and >750 lb/cy for FA2). Strength requirements do not usually control the design of SCC mixes for cast-in-place applica- tions. To determine powder content more precisely, water content for SCC mixtures is estimated using Table 6.3.3 of ACI 211.1-91 for different NMSA and assuming a 1 to 2 in. slump in air entrained concrete (305 lb/cy for 3⁄8 in. NMSA; 295 lb/cy for ½ in. NMSA; and 280 lb/cy for ¾ in. NMSA). Water-to-powder ratio (w/p) shall be between 0.37 and 0.44 for cast-in-place SCC. 6. Select the type and amount of SCMs and mineral fillers based on availability and project requirements. Recom- mended SCM and filler percentages are 25% Class F fly ash, 25% Class C fly ash, 30% slag, and 20% Class F fly ash + 15% limestone powder. Limestone powder should be included in the total powder content and W/P ratio calculations. This is because several earlier studies have indicated that there is a synergistic effect of ground limestone that is reacting with the C3A in the system to enhance the reactivity of the remaining constituents, such as cement and fly ash (Cost et al., 2012; Beeralin- gegowda and Gundakalle, 2013; and Bucher, 2009). 7. Using the absolute volume method, determine the quan- tities of fine and coarse aggregate. According to ACI 237, the absolute volume of coarse aggregate should be 28% to 38%. B.2.4 Quality Assurance Once the SCC mix proportioning is completed, at least three trial batches are needed to verify the properties of the fresh and hardened SCC. For cast-in-place bridge components, the target 28-day compressive strength usually ranges from 4 ksi to 6 ksi, and the target air content usually ranges from 4.5% to 7.5%. These targets, in addition to any other requirements in the proj- ect specifications, shall be checked for each trial batch prior to concrete production in the plant setting. Necessary adjustments shall be made to ensure that all requirements are fulfilled. Trial batches are also used to determine the dosage of HRWRA (typically polycarboxylate based), VMA (if needed), and WRA (if needed) that achieve the workability targets. Guidelines for adjustments that might be needed when workability targets are not met can be found in Table 5 of the ICAR 108-1 report (Koehler and Fowler, 2007). In addition, the robustness of SCC mixes shall be evaluated by investigating the effect of minor variations in the water content (8–16 lb/cy) on the workability properties of the designed SCC mix. B.3 Guidelines for Testing Fresh Concrete B.3.1 General Properly designed SCC should have adequate workability in its fresh state to allow placement without mechanical con- solidation while maintaining its stability to ensure satisfactory performance in the hardened state. Workability requirements for successful casting of SCC include good deformability (FA), PA, and adequate SR. In terms of fundamental rheology, SCC is characterized by a low yield stress to ensure high deformability and a moderate plastic viscosity to maintain homogeneous sus- pension of coarse aggregate, hence avoiding segregation and blockage during flow and ensuring good passing and filling abilities. Several empirical test methods have been developed and used to evaluate the workability properties of SCC. Other fresh and early-age concrete properties, such as air content, heat of hydration, and formwork pressure, also need to be checked to ensure the constructability, durability, and strength of SCC. B.3.2 Rheology Rheology is the science of deformation and flow of matter. The two key parameters used to describe the rheology of SCC are: yield stress (t0), which represents the amount of shear stress required to cause concrete to deform (flow); and plastic

B-9 viscosity (µp), which describes the ease/resistance of flow at a certain shear stress. A high yield stress results in low FA, while a high plastic viscosity results in difficult placement and slow flow of SCC. Slump flow and T50 are good indicators of yield stress and plastic viscosity, respectively. A properly designed SCC should have lower yield stress than conventional concrete to achieve the target FA, and adequate viscosity to ensure SR. Concrete/mortar rheometers are used to determine yield stress and plastic viscosity by plotting the relationship between shear stress and shear rate for a given mixture assuming the Bingham model (ACI 237R, 2007). This test should be conducted in the lab as it requires qualified personnel to operate the rheometer and interpret the data (not suitable for site use). Another rheological property that is important to describe the behavior of SCC is thixotropy, which is the reversible time- dependent increase in viscosity when concrete is at rest (i.e., stiffening or build-up) and decrease in viscosity (i.e., break- down) when subjected to adequate shearing force (agitated). A high thixotropic SCC has several advantages, such as high static stability and reduced lateral pressure on forms. On the other hand, it is not favorable for multi-lift castings as it could result in pour lines (lift lines) when the time between successive cast- ings is relatively long. Thixotropy can be evaluated in the lab using the rheometer or on the site using the portable vane test (Omran, Naji, and Khayat, 2011). The yield stress, viscosity, and thixotropy of SCC have a significant effect on several fresh concrete properties, such as FA, PA, stability, pumpability, formwork pressure, and work- ability retention. B.3.3 Filling Ability FA (deformability or flowability) describes the ability of the SCC to flow into and completely fill all spaces within the form- work under its own weight without any mechanical consoli- dation. Different levels of flowability might be needed based on the geometry of the component, the required quality of the formed surface, and the method of placement (placement energy, location of placement point, and spacing between placement points). The slump flow test is a common procedure used to deter- mine the horizontal free-flow characteristics of SCC in the absence of obstructions (AASHTO T 347). The procedure is based on AASHTO T 119 standards for determining the slump of conventional concrete. The test is easy to perform either at a concrete plant or on a job site, repeatable, reproducible, and can be performed by single operator. This test evaluates the capability of the concrete to deform under its own weight and the time needed for the concrete to spread 20 in. (T50). It should be noted that the results of two slump flow tests on the same batch properly conducted by the same operator should not dif- fer by more than 3 in. (ASTM C1611). Other non-standard test methods for evaluating the FA of SCC include the V-funnel test and Orimet test (EFNARC, 2002). Two levels of FA are recommended for cast-in-place bridge components: moderate FA (22–26 in.), and high FA (26–30 in.). T50 of 1 to 6 sec is generally acceptable for civil engineering applications (EFNARC, 2002). Examples of the required level of FA for different bridge components are shown in Table B-2. B.3.4 Passing Ability PA describes the ability of SCC to pass among obstacles (e.g., reinforcements) and narrow spacing in the formwork without segregation and blockage. Different levels of PA might be needed based on the geometry of the component, level of reinforcement intensity and spacing, and method of placement. The J-ring test can be used to characterize the passing ability of fresh SCC with NMSA of up to 1 in. (AASHTO T 345). When SCC is placed in forms containing steel reinforcement, the mix- ture should remain cohesive, and the aggregates should not separate from the paste fraction of the mixture when it flows between obstacles. This is a critical characteristic of the mixture when it is used in highly congested reinforced structures. The difference between the J-ring slump flow and the unconfined slump flow is an indication of the degree to which the pas- sage of SCC through reinforcing bars is restricted. Another measurement of the J-ring test is the height difference of concrete inside and outside of the J-ring. The greater the difference in height inside and outside of the J-ring, the less the PA. Two levels of PA are recommended for cast- in-place bridge components: moderate PA (2–4 in. diam- eter difference and 0.6–0.8 in. height difference; and high PA (<2 in. diameter different and 0.6 in. height difference). The NMSA is determined based on the required level of PA. Examples of the required level of PA for different bridge components are shown in Table B-2. The J-ring test is easy to perform either at a concrete plant or on a job site. The test is repeatable, reproducible, and can be performed by one operator. Other test methods for evalu- ating the PA of SCC include the caisson test (AASHTO T 349). The caisson test evaluates the filling capacity of SCC, which describes the FA and PA of SCC with up to ¾ in. NMSA. The test is designed specifically for sections with reinforce- ment bars/strands that are at 2 in. spacing vertically and hori- zontally. The test is difficult to perform with one operator and requires calculation to determine the filling capacity of SCC; therefore, it is not recommended for on-site use. Filling capacity of 80% or more is recommended for components that require high PA, while filling capacity of 70% to 80% is acceptable for components that require moderate PA. Other non-standard test methods include the L-box test and U-box test (EFNARC, 2002).

B-10 B.3.5 Static Stability Static stability describes the ability of SCC to maintain homogeneous distribution of its various constituents while being in the forms (at rest). Static stability refers to the resis- tance of SCC to bleeding, segregation, and surface settlement from the end of casting until setting. Different levels of stability might be required based on the geometry of the component and placement method. The visual stability index (VSI) is commonly used to visu- ally determine the apparent stability of the slump flow patty (AASHTO T 351). This test method is simple and can be per- formed by a single operator at the same time the slump flow test is performed. The VSI is a qualitative rating that is used to compare batches of the same or similar SCC mixtures with respect to the tendency to bleeding and uniformity of aggre- gate distribution. A similar test can be performed on hardened SCC, known as the hardened visual stability index (HVSI) (AASHTO PP 058), to determine the relative stability of SCC batches by comparing the cut planes of hardened concrete cylinders. A quantitative rating is assigned based on the uni- formity of aggregate distribution and the thickness of the top mortar layer. Two levels of static stability are recommended for cast-in- place bridge components: moderate stability (VSI = 1 and HVSI = 2) and high stability (VSI = 0 and HVSI = 0). Other test methods for the quantitative assessment of SCC static stability include the column segregation test (ASTM C1610) and rapid penetration test (ASTM C1712). The col- umn segregation test is suitable for laboratory use to determine the potential static segregation of a SCC mixture by measuring the difference in coarse aggregate content in the top and bot- tom portions of a vertical cylindrical specimen that simulates SCC in a vertical form. The test is difficult for a single opera- tor to perform, time consuming, and requires a special appa- ratus. The rapid penetration test provides a simple and quick method to evaluate static stability indirectly by measuring the penetration of a specific cylinder into the SCC placed in the slump cone before conducting the slump test. Guidelines for classifying SCC static stability using these test methods are that penetration less than 0.5 in. and column segregation less than 10% indicates high stability, while penetration from 0.5 to 1.0 in. and column segregation from 10% to 15% indicates moderate stability. Other non-standard tests include the sieve segregation test (De Schutter, 2005) and surface settlement test (Khayat and Mitchell, 2009) B.3.6 Dynamic Stability Dynamic stability describes the ability of SCC to maintain homogeneous distribution of its various constituents during mixing and placement (free fall and flow). Adequate dynamic stability is required when SCC has to travel a long distance in a horizontal and/or vertical direction before reaching its final position and filling the form. Example applications include girders, walls, and arches. Some of the FA and PA test methods can be used to indi- cate the dynamic stability of SCC. No standard test method is currently available to specifically evaluate the dynamic stability of SCC. The flow trough test was developed to determine the dynamic segregation index (DSI) of SCC by measuring the dif- ference in the weight of coarse aggregate in two samples taken before and after flowing in a 6 ft long apparatus (Lange et al., 2008). Modifications to the flow trough were made and indi- cated a better performance in evaluated dynamic stability of SCC. The higher the DSI, the lower the resistance to dynamic segregation. Few experiments have indicated acceptable per- formance when DSI is less than 20%. Other test methods include the tilting box, in which the penetration depth in SCC is compared before and after traveling several cycles in a tilting box (Esmaeilkhanian et al., 2014). B.3.7 Heat of Hydration Any action that promotes the hydration process would increase heat liberation, such as increasing the portland cement quantity or using finer cement. There are no unique effects of using SCC with respect to heat liberation, especially when common ranges of paste content are used in mix design and a significant portion of the cement is replaced using SCMs and/or fillers. Calorimeters used to assess the heat of hydration of concrete mixtures are semi-adiabatic calorimeters (RILEM 119-TCE) and isothermal calorimeters (ASTM C 1749). Using both semi-adiabatic and isothermal calorimeters to measure the heat generation indicated that there was no significant dif- ference in temperature rise/heat generation between SCC and CVC mixtures. However, there was a significant delay in reach- ing the peak temperature in SCC mixtures. It should be noted that the rate of reaching the peak temperature is also depen- dent on the type of SCM/filler used. In mass concrete, special attention should be given to the release of heat of hydration to avoid cracking. B.3.8 Pumpability Pumping is the most common method of placing SCC because it provides the highest placing rate (EFNARC, 2002). Pump lines should be lubricated with cement mortar for the first part of the load (25–40 gallons) before pumping SCC. When SCC is pumped from the top, it is recommended that it be placed with a submerged hose in order to minimize the trapped air and segregation that could occur from free fall. Pumping SCC from the bottom minimizes the entrapped air and risk of segregation, which results in higher quality formed surfaces.

B-11 Despite the fact that SCC is more fluid than CVC, the higher pressure loss of SCC is attributed to the difference in the rheological properties of SCC and CVC (lower yield stress and higher viscosity and thixotropy) resulting in a different shear rate distribution and, consequently, a dif- ferent velocity profile inside pipes (Feys, Verhoeven, and De Schutter, 2008). Using a slightly larger hose diameter for SCC than the corresponding hose diameter for high slump CVC with similar aggregate type and size can significantly reduce the pressure loss and reduce the risk of blockages in the pump line. B.3.9 Time of Setting Time of initial setting is the elapsed time after initial con- tact of cement and water required for sieved mortar to reach a penetration resistance of 500 psi according to AASHTO T 197. In general, there was no evidence that the time of initial set of SCC mixtures with low slump flow was different from that of CVC mixtures (4 to 6 hr). However, the time of setting is highly dependent on the type of SCM/filler, dosage of HRWRA, and temperature. Mixtures with Class C fly ash have the longest time of setting, while mixtures with Class F fly ash have the shortest time of setting. Also, mixtures with high slump flow are expected to have a longer time of setting (6 to 10 hr) than those with low slump flow due to the retarding effects of HRWRA. The ambient temperature also has a significant effect on the time of setting. The higher the temperature, the shorter the time of setting. B.3.10 Workability Retention In some cast-in-place applications, retaining concrete work- ability for extended periods (90 minutes or more) is vital. Workability retention of SCC mixtures is dependent on the temperature, initial slump flow, type and dosage of admixtures used, and type and replacement percentage of SCM/filler used. This property was investigated by conducting the slump flow test (AASHTO T 347) at different times (15, 30, 60, and 90 minutes) to evaluate the loss of slump flow with time. Results indicated that the type of SCM/filler had a slight effect on workability retention; however, initial slump flow had a sig- nificant effect on the rate of losing workability. For example, SCC mixtures with an initial slump flow of 30 in. lose flow- ability at an average rate of 7 in. per hr, while mixtures with an initial slump flow of 24 in. lose flowability at an average rate of 4 in. per hr. The use of workability retaining admixtures early in the mixing phase has shown satisfactory results in reduc- ing the rate of workability loss. Also, adding limited dosages of HRWRA later on, after observing slump flow loss at the job site, can be effective in recovering the initial slump flow. However, the additional dosages of HRWRA may have a negative effect on the entrained air content and, therefore, should be carefully observed. B.3.11 Formwork Pressure For cast-in-place bridge components, formwork pressure plays a significant role in the construction cost and duration. Formwork pressure development is significantly influenced by casting rate, casting method, ambient environmental con- ditions, and mixture composition. Comparing the formwork pressure of several SCC and CVC mixtures has shown that the ratio of maximum exerted lateral pressure to hydrostatic pressure (Pmaximum/Phydrostatic) was higher in SCC than CVC. This comparison has also shown that low slump flow SCC mixtures (22–26 in.) exert less lateral pressure than high slump flow mixtures (26–30 in.). The correlations between Pmaximum/Phydrostatic and the rheological properties (i.e., thix- otropy and yield stress) of SCC mixtures indicated that Pmaximum/Phydrostatic exhibited a linear relationship with thix- otropy and yield stress as reported by Assaad et al. (2003) and Khayat and Assaad (2012). The higher the thixotropy and yield stress, the lower the lateral pressure. Therefore, the rheological properties of SCC need to be evaluated in order to allow for lower formwork pressure than the hydrostatic pressure in formwork design when SCC is used. Alternatively, a pressure test (according to the AASHTO TP 094) can be performed to determine the pressure dis- tributed in a mockup form. B.4 Guidelines for Hardened Properties B.4.1 General The hardened properties of SCC need to be accurately pre- dicted to allow bridge engineers to properly design cast-in- place bridge components using current design specifications and ensure their satisfactory performance and durability. B.4.2 Compressive Strength For cast-in-place bridge construction, the minimum specified 28-day compressive strength commonly ranges from 4.0 ksi to 6.0 ksi. In several situations, 7-day, 14-day, and 56-day compressive strength are specified for construc- tion and structural purposes. Studying the relationships between the average 28-day compressive strength and that at 7, 14, and 56 days for SCC mixtures with different SCMs/ fillers and aggregate types indicated that there was no significant difference between SCC and CVC with respect to compressive strength development. The ACI 209R-92 model can be used to accurately predict the ratio of compressive strength in ksi

B-12 at a given time (t) in days (fc)t to the 28-day compressive strength as follows: 28 f f t t c t c d ( ) ( ) = α + β The values of the constants a and b for cement type I and moist cured concrete are 4.0 and 0.85, respectively, accord- ing to Table 2.2.1 of ACI 209R-92. The average ratio of 7-day, 14-day, and 56-day compressive strength to 28-day compres- sive strength were found to be 0.77, 0.88, and 1.12, respectively, for SCC mixtures. Also, mixtures with limestone aggregate have shown higher compressive strength than those with gravel aggregate, which is attributed to the effect of the interfacial tran- sition zone (ITZ) between aggregate particles and paste. Addi- tionally, all mixtures with limestone powder have experienced low compressive strength, which is attributed to the powder particle size. Limestone powder with finer particle size results in higher compressive strength. B.4.3 Modulus of Elasticity Modulus of elasticity (MOE) is an important design param- eter for deflection, deformation, and prestress loss calculations. In the design phase, MOE is predicted, according to AASHTO LRFD Equation 5.4.2.4-1, as a function of specified compressive strength and the unit weight of concrete. The measured MOE of SCC mixtures, according to ASTM C469, was found to be slightly lower than predicted by AASHTO LRFD, as reported by Pineaud et al. (2005). Therefore, it is recommended to introduce a modification factor to the AASHTO LRFD equation as follows: 33,000 5.4.2.4-11 2 1.5E K K w fc c c ( )= ′ where K2 = modification factor to be taken as 0.96 for SCC and 1.0 for CVC. Comparing the predicted MOE using the revised equa- tion with the measured MOE for SCC mixtures with different aggregate types indicated that the aggregate source factor (K1) should be taken as 1.0 for crushed limestone and 0.95 for natu- ral gravel. This is because SCC mixtures with gravel aggregate demonstrated slightly lower MOE than those with limestone aggregate as reported by Mokhtarzadeh and French (2000). B.4.4 Splitting Tensile Strength The splitting (direct) tensile strength of concrete is used in the design of bridge components subjected to tension force caused by means other than flexure, such as in anchorage zones. The direct tensile strength of SCC obtained from test- ing, according to AASHTO T198, was found to be 20% less than that predicted by AASHTO LRFD 5.4.2.7 as 0.23 f c′. Therefore, a modification factor of 0.8 is recommended for all SCC mixtures with compressive strength less than 8 ksi and regardless of the aggregate type. B.4.5 Modulus of Rupture The modulus of rupture (MOR) of concrete is used primar- ily in calculating the cracking moment of bridge components for serviceability limit states. AASHTO LRFD Section 5.4.2.6 provides a range for the MOR of normal weight concrete from 0.24 f c′ to 0.37 f c′. The measured MOR of SCC mix- tures according to AASHTO T97 was found to be within the predicted range (mostly closer to the upper limit) and very comparable to that of CVC, a finding that is in agreement with findings reported in NCHRP Report 628 (Khayat and Mitchell, 2009). Therefore, no changes are recommended to the current AASHTO LRFD with respect to MOR. B.4.6 Bond to Deformed Reinforcing Steel Bars The bond strength of concrete to reinforcing steel is an important parameter for determining the development length of bars. Results of pull-out testing of #6 deformed bars embed- ded vertically in concrete blocks made of SCC and CVC mix- tures, according to Moustafa (1974), indicated that the bond strength of SCC was significantly lower than that of CVC. Therefore, a development length modification factor of 1.3 is recommended to AASHTO LRFD 5.11.2.1.2 for bars in tension placed vertically in SCC. This factor is close to the 1.4 factor recommended in NCHRP Report 628 for bond with prestressing strands (Khayat and Mitchell, 2009). Results of pull-out testing of #6 deformed bars embedded horizontally at different elevations in concrete walls made of SCC and CVC mixtures indicated that the bond strength of SCC is not significantly different from that of CVC. This testing also indicated that the top-bar effect in SCC mixtures with high slump flow is lower than that in CVC mixtures and SCC mixtures with low slump flow. However, no devel- opment length modification factor is recommended for top horizontal tension bars in high slump flow SCC, and the 1.4 factor used for top horizontal tension bars in CVC according to AASHTO LRFD 5.11.2.1.2 will be used for SCC regardless of the slump flow as the slump flow may change with time in the same component. B.4.7 Shear Resistance The interface shear resistance of SCC and CVC mixtures was evaluated using push-off testing for a wide range of concrete

B-13 strengths (4.0–8.0 ksi). The results indicated that the average interface shear resistance of SCC was similar to that of CVC. Comparing the measured interface shear resistance versus that predicted according to AASHTO LRFD 5.8.4.1 indicated that the AASHTO LRFD over estimates the interface shear resis- tance for mixtures with compressive strength less than 6 ksi. Therefore, it is recommended to take the cohesion factor, c, as 0.0 for predicting the interface shear resistance when concrete compressive strength is less than 6 ksi. It should be noted that AASHTO LRFD doesn’t consider the concrete compressive strength in estimating interface shear resistance, but does con- sider both cohesion and friction factors. The nominal shear resistance of SCC and CVC mixtures was evaluated using beams with different levels of transverse reinforcement tested under point loads. Results indicated that there was no significant difference in the shear resistance of SCC and CVC regardless of the level of shear reinforcement. Results also showed that the nominal shear resistance of SCC can be conservatively predicted using AASHTO LRFD Sec- tion 5.8.3.3 (sectional design method) when different levels of transverse reinforcement are used. B.4.8 Drying Shrinkage Predicting drying shrinkage of concrete is important to minimizing cracking and estimating long-term losses in post- tensioned components. The drying shrinkage of SCC and CVC mixtures was measured using AASHTO T160 for 56 days and compared against predicted shrinkage according to AASHTO LRFD 5.4.2.3. Results indicated that measured shrinkage was significantly higher than predicted, as reported in the findings of NCHRP Report 628 (Khayat and Mitchell, 2009). Results also indicated that the type of SCM has a significant effect on the drying shrinkage. Therefore, in the absence of a physical test, the following modification factors are proposed to better estimate drying shrinkage for each type of SCM/filler: 1.6 for Class C fly ash, 1.4 for GGBFS, and 1.3 for Class F fly ash with/ without limestone powder. B.4.9 Restrained Shrinkage Most cast-in-place concrete components experience shrink- age while they are restrained, which results in tensile stresses that cause cracking. Therefore, measuring the restrained shrinkage of SCC and CVC mixtures according to ASTM C1581 is impor- tant in order to compare their cracking potential in conditions similar to the site conditions. Under NCHRP Project 18-16, the time to cracking and average stress rate of SCC and CVC mix- tures were measured for up to 28 days. Results indicated that there was no significant difference in the cracking potential of SCC and CVC mixtures containing the same SCM/filler and aggregate size. Mixtures with Class C fly ash and/or 3⁄8 in. NMSA had a higher cracking potential than those with Class F fly ash and ¾ in. NMSA. B.4.10 Creep Predicting creep of concrete is important for determining long-term deformation and prestress losses in bridge compo- nents. Creep of SCC and CVC mixtures was measured accord- ing to ASTM C512 over a 1-year period and used to calculate the creep coefficient. Results indicated no significant differ- ence between the creep of SCC and CVC mixtures containing the same type of SCM/filler. Also, comparing measured and predicted creep coefficients using AASHTO LRFD 5.4.2.3.2-1 indicated that the creep coefficient can be accurately predicted using AASHTO LRFD for all SCC mixtures except those with limestone powder that exhibit higher creep strains. Therefore, a modification factor of 1.2 is proposed only for predicting the creep of SCC mixtures containing limestone powder as reported by Heirman et al. (2008). B.4.11 Durability Properties The durability of a concrete element is highly dependent on its permeability. As an alternative to the rapid chloride ion penetrability test of concrete mixtures, a surface resistivity test was conducted (according to AASHTO TP 95) on SCC and CVC mixtures to evaluate their penetrability. Results indicated that the surface resistivity of SCC and CVC was not significantly different. Results also indicated that the surface resistivity was highly dependent on the type of SCM/filler: mixtures with Class C fly ash had the lowest surface resistivity, while mixtures with GGBFS had the highest surface resistivity. Comparing the surface resistivity results to the chloride ion penetration classes (according to ASTM C1202) indicated that all SCC mixtures developed for cast-in-place bridge compo- nents had low-moderate penetrability. For bridge components subjected to freezing and thawing, the air void system is vital to their durability. Air void system parameters (i.e., air content, space factor, and specific surface) are measured for SCC and CVC mixtures according to ASTM C 457. Results indicated that there was no significant difference between SCC and CVC with respect to air void system param- eters. All mixtures had a space factor less than 0.2 mm and specific surface higher than 24 mm2/mm3, which are the rec- ommended thresholds for good freeze-thaw resistance (PCA, 2009). The air content in hardened concrete varied from 3% to 8%, which might be accepted by several transportation agen- cies depending on the application and environment. However, comparing the air content in fresh SCC at the plant to that at the job site showed significant differences, which may be due to the effect of the additional dosages of HRWRA (to increase slump flow) on the entrained air content.

B-14 B.5 Guidelines for Production and Construction B.5.1 General Successful production and construction of cast-in-place bridge components using SCC requires more attention to the selection of materials, mixing and testing procedures, and placement and finishing methods than using conventional vibrated concrete. Trial batches are necessary to verify that SCC properties meet the target values/ranges and to make neces- sary adjustments to mixture proportions and/or production and construction procedures. B.5.2 Quality Control of Constituent Materials SCC is more sensitive to variations in constituent materi- als than CVC; therefore, proper control of the properties and quantity of all constituent materials should be ensured. Special consideration should be given to water content as it is a key factor in the stability of the mixture. RILEM (2006) recom- mends the use of at least two different methods for measur- ing the moisture content of aggregates. Also, only production equipment that has a tolerance of 1 to 2 gallons per cubic yard in water content can be used. It is highly recommended that best practices on maintaining material stockpiles be used, such as moisture control, free drainage, cleanliness, and prevention of segregation. The use of overhead bins for material storage is also recommended. Concrete plants should have additional silo capacity to store various filler materials and extra high vol- ume tanks and dispersing systems for liquid admixtures (i.e., HRWRA and VMA). Combining admixtures is not recom- mended due to the different dosage rate requirements of each admixture. Since small variations in the physical properties of the aggregates (i.e., gradation, particle shape, absorption, mois- ture content, and percentage of fines) can have a significant effect on workability, frequent inspections of the aggregate stor- age places are necessary. B.5.3 Mixing Procedures According to AFGC (2002), a concrete plant with the fol- lowing characteristics is recommended for the production of SCC: a mixer with a high shear rate, entirely automatic pro- duction control, wattmeter or equivalent; moisture probes for sand, and storage of aggregates in a dry place and/or use of a reliable moisture content evaluation system for each aggregate size. The EFNARC (2002) does not recommend any specific mixer type. Forced action mixers (e.g., paddle mixers), and free-fall mixers (e.g., truck mixer) can be used. RILEM (2006) indicates that force type mixers are more efficient in mixing SCC and large mixers are recommended because small mixers tend to require a longer mixing time. Generally, the mixing time of SCC is expected to be longer than that of CVC (an additional 30 to 90 sec) (ACI Committee 237). Below are con- crete mixing guidelines according to AFGC (2002) • Use stationary equipment during the time required to obtain complete stabilization of the wattmeter, or set up a reliable procedure to measure mixing efficiency. • Whatever the case is, the mixing time must not be fewer than 35 sec for strengths less than or equal to 4.5 ksi and 55 sec for other strengths. • In the case of on-site production of concrete that is not to be kept at least 5 minutes in a receptacle that keeps the concrete moving (truck mixer or receiving bin), the mixing time in a concrete plant must be at least 55 sec. B.5.4 Plant Quality Assurance EFNARC (2002) suggests that additional resources may be needed for supervision of all aspects of the initial produc- tion of SCC. According to ACI Committee 237 (2007), the employees associated with the production, testing, or use of SCC should be trained and qualified appropriately for day-to- day quality control, have appropriate certification, understand the engineering properties and placement techniques of fresh SCC, and learn the proper corrective actions when perfor- mance requirements are not met. The slump flow test and VSI should be checked for each truck load in addition to air content and unit weight of fresh SCC. The rapid penetration and J-ring tests could be conducted if specified by the material engineer. Records should be kept for future adjustments if necessary. B.5.5 Transportation Procedures According to ACI Committee 237 (2007), SCC can be transported using all of the conventional concrete devices, but some precautions should be considered as follows: 1. Transit mixer: Deliver SCC to a job site by a concrete truck. – Place the volume of SCC into a truck without exceeding 80% capacity of the drum to ensure that SCC does not spill out of a concrete truck whenever the truck goes up or down a steep incline. – Keep the revolving drum turning in the mixing mode direction while in transport. Alternatively, deliver the mixtures to the project at a lower slump flow than required and add an HRWRA to bring the mixture to the required slump flow. – When SCC is delivered and placed by a concrete truck where the speed of discharge and volume of concrete deliv- ered is high and continuous, the mixture may experience further flowing distances and improved filling capacities.

B-15 – A concrete truck is an effective method of placing SCC mixtures with all slump flow levels. 2. Hopper or bucket: SCC can also be transported to forms by hopper or bucket transporters or other specialized devices. – When hopper-type vehicles are used, SCC mixtures should be very stable and able to resist segregation from vibratory forces without receiving any addi- tional mixing. – SCC transported by bucket from an overhead lift receives minimal or no vibration and does not require the same level of stability as SCC transported in a hopper-type vehicle. – Bolting rubber strips or pads to the clam shell discharge point of buckets is an effective method to prevent leakage of low viscosity SCC mixtures during transport to forms. – A chute attached below the bucket opening can direct the flow of SCC toward specific areas of the form to be filled. – When a hopper or bucket placement method is used, a limited volume of SCC can be placed at any one time, thus reducing the rate of concrete placement and result- ing in a discontinuous discharge rate of flow of concrete compared with placement by truck chute, where a large volume of concrete can be continuously cast. – The use of larger volume transport vehicles, such as concrete trucks, rather than hopper or bucket trans- porters, is advantageous in rapidly filling forms and avoids the production of multiple batches of concrete in the case of a relatively large section. – SCC placement by bucket has a high discharge rate and is discontinuous. SCC transported to a form by bucket should have a slump flow of 24 to 28 in. to help facilitate placement by increasing flow distance and permitting consolidation with consecutive loads. B.5.6 Site Quality Assurance Inspection of SCC should be conducted on site to con- firm that workability requirements are satisfied. According to JSCE (1999), for on-site quality control: • SCC should be tested at the time of concrete placing/ unloading, and the slump flow test and VSI shall be carried out for each batch unless additional workability require- ments are specified. • When SCC is rejected due to low slump flow, HRWRA may be added at a predetermined dosage. But, when SCC is rejected due to segregation, SCC must not be used. Causes of rejection should be identified and documented to improve subsequent production and transportation practices. Cau- tion should be given to the effect of added HRWRA at the job site on the entrained air content. RILEM (2006) recommends tighter quality assurance in the start-up phase of casting SCC. This is because of large work- ability fluctuations caused by the starting up of mixer, truck, pumps, etc. Sampling of every batch or truckload is recom- mended until the stability and consistent quality of the deliv- ered concrete is achieved. AFGC (2002) recommends that on-site acceptance of concrete involves checking whether SCC is suitable for placement without consolidation. It is recom- mended to carry out an acceptance test on at least the first batch of the day and systematically whenever there is any doubt. The acceptance procedure includes sampling of a rep- resentative specimen of concrete (if the concrete is delivered by truck, it should be mixed at high speed for at least one minute); conducting a slump flow test using the traditional slump cone; and checking that the results lie within the specified range. B.5.7 Placement Methods Before placing SCC, reinforcement and formwork should be inspected as it is for vibrated concrete to ensure that they are arranged as planned, and the formwork is in good condition (EFNARC, 2002). When placing SCC in the forms, the free-fall heights and horizontal flow distance defined for the different SCC classes must be respected unless the SCC has been tested and no segregation was found (AFGC, 2002). The following rules are advised by EFNARC (2002) to minimize the risk of segregation: • Limit the vertical free-fall distance to 16.5 ft. • Limit the permissible distance of horizontal flow from point of discharge to 33 ft. For low viscosity SCC (T50 < 2 sec), the maximum period of time between layers is about 90 minutes. However, for high viscosity SCC, the maximum period of time between layers should be studied for specific viscosity and layer thickness (AFGC, 2002). For horizontal applications, SCC can be placed by pouring it directly from the chute of the truck mixer or the concrete skip or by pumping. For vertical applications, AFGC (2002) recommends several SCC placement methods includ- ing the following: 1. Concrete skip with flexible pipe: – Unsatisfactory results in terms of formed surface qual- ity (bugholes) may be obtained in placing SCC by pouring it into the forms from above even if the free- fall height is respected. – Use a concrete skip with a flexible pipe and limit the free- fall height of the concrete into the forms, along with pos- sibly reducing the pipe diameter (3 to 4 in. maximum). – Attention should be paid to the closure of the gate when concrete is placed by skip.

B-16 2. Skip with tremie pipe: – Insert a tremie pipe into the concrete in order to avoid the fall of fresh concrete into the forms. – The diameter of the tremie pipe must be adjusted to suit both the geometry of the forms (height and thick- ness) and the density of the reinforcement (passages must be left for the pipes to go through). – The diameter of the tremie pipe must reduce the risk of plugging the tube (usual pumping rules). – A funnel should be placed on top of the tremie pipe to make it easier to pour the concrete. – The advantages of this method are that all the place- ment precautions are systematically respected, and keeping the pipe in the concrete during pouring pre- vents entrapped air during placement. 3. Pump (with tremie pipe): – Using a pump allows high placement rates of SCC as no interruptions are needed to consolidate concrete. The rate of discharge is dependent on the availability of concrete and formwork design (no upper/lower bounds for the rate of discharge). – Pumping pressure depends on the rheological prop- erties of the SCC mixture in addition to the external factors. Also, the rheology of SCC may be changed due to pumping. – Pumping SCC follows the same procedures as pump- ing CVC. 4. Pumping from the bottom of the form (with injection pipe): – Pumping SCC from the bottom of the forms via injec- tion pipes prevents the concrete free falling and reduces the number of site workers. – Pumping SCC from the bottom of the form reduces the entrapped air and, consequently, the formation of bugholes, which results in a better concrete surface quality. – The concrete injection system at the bottom of the form must be designed to prevent the concrete from bouncing off the opposite side of the form and facili- tate closing of the box-out at the end of placement (sliding hatch). According to the JSCE (1999), pumping SCC tends to reduce the slump flow, and increasing the pumping rates leads to greater pumping pressure loss than with CVC. Therefore, the pump type and the diameter and length of the pipeline should be examined before consideration. In gen- eral, pumping through 4 to 5 in. diameter pipes that are not longer than 1,000 ft is common with SCC. High placement rates of SCC can entrap air if the mixture is not propor- tioned adequately for the given geometry and reinforcement condition. B.5.8 Formwork Considerations According to AFGC (2002) and ACI Committee 237 (2007), the following guidelines are provided to attain a good formed surface quality of the SCC component: • Special attention must be paid to the condition of the forms, which should be free of grease, grout, and rust when metal forms are used. • High quality release oil in spray form should be used to produce a uniform film with no drips. • Any excess oil, which results in bugholes and concrete build- up on the surface, can be removed. • SCC’s very fluid consistency (especially with low viscosity SCC) requires that the formwork used for SCC be designed with more attention to water tightness and grout tightness, particularly at the bottom, than conventional formwork in order to avoid honeycombs, surface defects, and leakage. • Due to the good cohesion of SCC, the formed surface quality is not altered by slight tightness defects (typically less than 0.1 in.). • When placing SCC in closed spaces, vent holes shall be pro- vided in an appropriate position in the top forms to allow entrapped air to escape. Despite the fact that SCC has thixotropic properties resulting in lower formwork pressure than hydrostatic pressure, the high fluidity of SCC promotes a high placement rate, which offsets the benefit from SCC thixotropic effects and leads to higher formwork pressure than CVC (AFGC, 2002). Therefore, it is highly recommended to dimension the forms to withstand the full hydrostatic pressure, especially for high form filling speed (greater than or equal to 40 ft/hr), unless a mockup form has been tested to prove otherwise (AFGC, 2002; JSCE, 1999; ACI Committee 237, 2007). Below are other recommendations from AFGC (2002) regarding the formwork pressure of SCC: • It is essential to design forms, falsework, and bracing to withstand the pressure at the bottom of the formwork (i.e., where it is highest). • When SCC is pumped or injected from the bottom, the local dynamic effects due to injection must be considered in addition to the pressure exerted by the concrete. • Thixotropic properties of SCC depend on the temperature of the fresh concrete and can be altered by vibration after placement (e.g., site traffic). ACI Committee 237 (2007) states that the maximum initial formwork pressure and its rate of drop with time are affected by the rheology, thixotropy, initial consistency of the concrete, casting rate, and ambient temperatures. Mixture proportions affecting formwork pressure of SCC include coarse aggregate volume, binder type and content, and the type and dosage of

B-17 HRWRA (Assaad and Khayat, 2005). Formwork designs that accommodate the expected liquid head formwork pressures can allow unrestricted placement rates and permit the contractor to take full advantage of the fast casting rate of the SCC. B.5.9 Finishing Techniques Due to the absence of bleeding water and the possible thixotropic stiffening of SCC, the surface finish of hori- zontal concrete surfaces could be problematic. To obtain an acceptable finish on a horizontal surface, a float should be used immediately after placing concrete (AFGC 2002; JSCE 1999). Otherwise, measures to prevent surface drying until the time of finishing should be considered. EFNARC (2002) advises that surfaces of SCC should be roughly lev- eled to the specified dimensions, and the finishing should then be applied at an appropriate time before the concrete stiffens. The small amount of bleeding water in SCC forms less laitance on the surface of the joint, which improves the performance of the joint surface even with little surface roughening (JSCE, 1999). According to ACI Committee 237 (2007), applying a roughened finish too soon may result in the SCC mixture flowing back to a smooth, level surface. Performing a setting-time test on the SCC mixture before placement can provide the information necessary to estab- lish the correct timing of the final finish operation (ASTM C 403/C 403M). B.5.10 Curing Methods According to ACI Committee 237 (2007), SCC is no different than CVC in terms of outside factors that affect performance. Other factors, such as cement type, aggregate gradations, water content, mixture proportions, and air content, can affect SCC in a manner similar to CVC. Therefore, the established guide- lines for curing in ACI 308R and AASHTO LRFD 8.11 should be followed with SCC. EFNARC (2002) advises that initial curing should be com- menced as soon as practicable after placing and finishing SCC in order to minimize the risk of shrinkage cracking. AFGC (2002) recommends that particular care should be taken in choosing the curing methods to be used after placement in order to pre- vent too much evaporation during the first hours of hardening depending on the type and amount of SCM and filler used. For horizontal applications, curing should be applied immediately after concrete placement in order to prevent too much evapora- tion, which causes early cracking and loss of durability in con- crete cover. The used curing agent should be compatible with the subsequent addition of a sealing coat. Membrane curing or similar methods for curing cast-in-place components should stay for at least 4 days (Swedish Concrete Association, 2002).

Abbreviations and acronyms used without definitions in TRB publications: A4A Airlines for America AAAE American Association of Airport Executives AASHO American Association of State Highway Officials AASHTO American Association of State Highway and Transportation Officials ACI–NA Airports Council International–North America ACRP Airport Cooperative Research Program ADA Americans with Disabilities Act APTA American Public Transportation Association ASCE American Society of Civil Engineers ASME American Society of Mechanical Engineers ASTM American Society for Testing and Materials ATA American Trucking Associations CTAA Community Transportation Association of America CTBSSP Commercial Truck and Bus Safety Synthesis Program DHS Department of Homeland Security DOE Department of Energy EPA Environmental Protection Agency FAA Federal Aviation Administration FAST Fixing America’s Surface Transportation Act (2015) FHWA Federal Highway Administration FMCSA Federal Motor Carrier Safety Administration FRA Federal Railroad Administration FTA Federal Transit Administration HMCRP Hazardous Materials Cooperative Research Program IEEE Institute of Electrical and Electronics Engineers ISTEA Intermodal Surface Transportation Efficiency Act of 1991 ITE Institute of Transportation Engineers MAP-21 Moving Ahead for Progress in the 21st Century Act (2012) NASA National Aeronautics and Space Administration NASAO National Association of State Aviation Officials NCFRP National Cooperative Freight Research Program NCHRP National Cooperative Highway Research Program NHTSA National Highway Traffic Safety Administration NTSB National Transportation Safety Board PHMSA Pipeline and Hazardous Materials Safety Administration RITA Research and Innovative Technology Administration SAE Society of Automotive Engineers SAFETEA-LU Safe, Accountable, Flexible, Efficient Transportation Equity Act: A Legacy for Users (2005) TCRP Transit Cooperative Research Program TDC Transit Development Corporation TEA-21 Transportation Equity Act for the 21st Century (1998) TRB Transportation Research Board TSA Transportation Security Administration U.S.DOT United States Department of Transportation

TRA N SPO RTATIO N RESEA RCH BO A RD 500 Fifth Street, N W W ashington, D C 20001 A D D RESS SERV ICE REQ U ESTED ISBN 978-0-309-37562-7 9 7 8 0 3 0 9 3 7 5 6 2 7 9 0 0 0 0 N O N -PR O FIT O R G . U .S. PO STA G E PA ID C O LU M B IA , M D PER M IT N O . 88 Self-Consolidating Concrete for Cast-in-Place Bridge Com ponents N CH RP Report 819 TRB