ACI Structural Journal/November-December 2011 1

Title no. 108-S65

ACI STRUCTURAL JOURNAL TECHNICAL PAPER

ACI Structural Journal, V. 108, No. 6, November-December 2011.MS No. S-2010-163.R1 received September 23, 2010, and reviewed under Institute

publication policies. Copyright © 2011, American Concrete Institute. All rights reserved, including the making of copies unless permission is obtained from the copyright proprietors. Pertinent discussion including author’s closure, if any, will be published in the September-October 2012 ACI Structural Journal if the discussion is received by May 1, 2012.

Bond and Shear Behavior of Concrete Beams Containing Lightweight Synthetic Particlesby Matthew J. Heiser, Amr Hosny, Sami H. Rizkalla, and Paul Zia

This paper summarizes a comprehensive experimental program that investigated the bond and shear behavior of concrete beams containing lightweight synthetic particles (LSP). LSP is a new concrete additive that, when used, leads to reduced unit weight of concrete, enhances flowability of the fresh concrete for pumping purposes, and produces durable concrete for freezing and thawing and deicing exposed conditions. It also reduces the thermal conductivity (increases R-value), thus reducing the energy required for heating and cooling. The use of these specially formulated particles, in combination with normalweight aggregates, could reduce the unit weight of concrete by 10 to 20%, ranging from 120 to 130 lb/ft3 (1920 to 2080 kg/m3), depending on the amount of LSP used in the concrete mixture. The experimental program included 27 large-scale specimens. Research findings indicate that the bond and shear behavior of beams with LSP additive is similar to the behavior of beams made with normalweight concrete. Test results confirm that ACI 318-08 can be used for the design of LSP concrete members for shear and the development length of steel reinforcement without the use of the reduction factor l required for lightweight concrete.

Keywords: additive; beams; lightweight synthetic particles; reduced unit weight concrete; shear.

INTRODUCTIONACI 318-081 defines normalweight concrete as concrete

containing aggregates that conform to ASTM C33/C33M-082 with a unit weight between 135 and 160 lb/ft3 (2160 and 2560 kg/m3). ACI 318-081 also defines lightweight concrete as concrete containing only lightweight aggregates conforming to ASTM C330-053 with a unit weight between 90 and 115 lb/ft3 (1440 and 1840 kg/m3). Lightweight synthetic particles4 (LSP) are polymer spheres with a closed cell inner structure containing air. They are inert and hydro-phobic. They have a maximum sphere diameter of 0.25 in. (6.4 mm) and a specific gravity of 0.042. They are specially formulated for use with concrete with the ability to disperse uniformly in the concrete, resulting in a reduced concrete density. In general, they are added to concrete as a partial replacement for conventional fine or coarse aggregate. These specially formulated particles are considered as an additive because they do not conform to either ASTM C33/C33M-08 or ASTM C330-05. Therefore, concrete made with normal-weight aggregates and LSP as an additive to reduce the unit weight is regarded as a normalweight concrete with “reduced” density.

This study summarizes the findings of an extensive research program conducted to examine compliance with ACI 318-08 for the bond and shear behavior of concrete members containing LSP. A parallel study by the authors to evaluate the material characteristics of LSP concrete indicates that, for a given mixture design, as the amount of LSP is increased in the concrete mixture to reduce the concrete density, the compressive strength of concrete is

reduced. For example, when the concrete density was reduced from 145 lb/ft3 (2333 kg/m3) without LSP to 130 lb/ft3 (2080 kg/m3) with LSP, the compressive strength of concrete was reduced by 30% from the original compressive strength of 5740 psi (39.6 MPa). When the concrete density was further reduced to 120 lb/ft3 (1920 kg/m3), the compressive strength was reduced by 50%. In all cases, the tensile strength and the modulus of elasticity of the LSP concrete correlated well with the predictions by the ACI 318-08 equations using the corresponding compressive strength and density. It should be noted that by adjusting the basic mixture design, one could achieve the desired structural strength after LSP is added. The product is being used commercially for structural concrete ranging from 130 lb/ft3 to 105 lb/ft3 unit weight with 28-day compressive strengths ranging from 3000 psi to 6000 psi.

The experimental program presented in this study consisted of 27 large-scale specimens, tested under static loading up to failure. Research findings indicate that the design of concrete members containing LSP for bond and shear can use the provisions of ACI 318-08 for normalweight concrete; and the modification factor l in the code, normally associated with lightweight concrete, is not applicable in this case for concrete with LSP additive.

The first phase of this study investigated the bond behavior of 18 large-scale specimens consisting of both slabs and beams. The second phase investigated the shear behavior of nine large-scale beams—each end of a beam being tested to replicate the results—thus providing a total of 18 tested specimens. The test results showed that the structural behavior of concrete containing LSP additive with a unit weight between 120 and 130 lb/ft3 (1920 and 2080 kg/m3) was similar to that of normalweight concrete.

RESEARCH SIGNIFICANCEReduced unit weight concrete using LSP additive has

several structural and economical advantages in comparison to normal- and lightweight concrete. The addition of LSP to a concrete mixture has been shown to reduce the pumping pressures of fresh concrete, improve the thermal resistance, and reduce the unit weight for the hardened concrete. The research presented in this paper provides the data and demonstrates that the ACI 318-08 code provisions can be used for the design of structural concrete members containing LSP additive and normalweight aggregates without the modification factor l associated with lightweight

2 ACI Structural Journal/November-December 2011

(2080 kg/m3) and were classified as Group 3. The targeted concrete compressive strength for the specimens within Group 1 was 2500 psi (20 MPa) and 4000 psi (30 MPa) for the specimens in Groups 2 and 3. All tested beams had cross-sectional dimensions of 12 x 18 in. (305 x 457 mm) and a total length of 16 ft (4877 mm), as shown in Fig. 1. All tested slabs had cross-sectional dimensions of 18 x 8 in. (457 x 203 mm) with a total length of 16 ft (4877 mm). Within each group, the tension splice length of the reinforcement was designed according to ACI 318-08 based on the targeted concrete strengths of 2500 or 4000 psi (20 or 30 MPa). The specimens were constructed using splice lengths equal to 0.75Ld, 1.0Ld, and 1.25Ld, where Ld is the splice length determined by ACI 318-08 requirements. For the beams, all the transverse reinforcement consisted of No. 3 closed stirrups designed according to ACI 318-08 requirements, with an extension of six times the bar diameter past the 90-degree bend. Two specimens were fabricated and tested for each of the selected splice lengths for the repeatability of the test results.

The bond specimens given in Table 1 are identified using three parameters: the first letter “B” stands for bond specimens; the following number identifies the targeted unit weight of concrete in pounds per cubic feet (120 or 130). The second number identifies the specimen within each group and ranges from 1 to 3. The letter “R” indicates the replicate specimens. The letter “S” was used to designate the slab specimens.

The test setup and cross-sectional dimensions for the beams and slabs are shown in Fig. 1. All beams and slabs were tested using a four-point bending setup to develop a constant moment region at the location of the spliced bars. The length of the beams and slabs was kept constant at 16 ft (4877 mm). The test setup allowed a constant moment region of 6 ft (1829 mm) with two outer shear spans of 4.5 ft (1372 mm) each. The beams were supported at both ends using composite steel sections restrained to the floor. A 150 kips (670 kN) load cell was placed at one end of the beams to measure the reaction at the support, whereas the loads were applied using four hydraulic jacks—two at each location—with a capacity of 120 kips (535 kN) each. Vertical deflections were measured using string potentiometers located at the midspan of the test specimen.

concrete when the concrete unit weight is 120 lb/ft3 (1920 kg/m3) or greater.

EXPERIMENTAL INVESTIGATIONThe overall experimental program consisted of 27 large-

scale reinforced concrete specimens tested under static loading up to failure. The first phase of the program included 12 beams and six slabs tested to evaluate the bond characteristics of the LSP concrete. The second phase included nine beams to evaluate the shear behavior of the concrete, with each beam being tested once at each end to replicate the test results. The steel reinforcement used had a specified minimum yield strength of 60 ksi (415 MPa), according to ASTM A615/A615M-09b5 specifications.

Phase 1: Bond behavior of LSP concrete membersThe bond behavior of concrete members with LSP

additive was evaluated using 12 beams and six slabs. The beam specimens were divided into three main groups using two targeted unit weights: 120 lb/ft3 (1920 kg/m3) for Group 1 and 130 lb/ft3 (2080 kg/m3) for Group 2. The slab specimens had a targeted unit weight of 130 lb/ft3

ACI member Matthew J. Heiser is a Graduate Research Assistant in the Civil, Construction, and Environmental Engineering Department at North Carolina State University, Raleigh, NC. He received his BSc in 2008 and his MSc in 2010 from North Carolina State University.

ACI member Amr Hosny is a PhD Candidate in the Civil, Construction, and Environmental Engineering Department at North Carolina State University. He received his BSc from Ain Shams University, Cairo, Egypt, and his MSc from North Carolina State University in 2004 and 2007, respectively.

Sami H. Rizkalla, FACI, is a Distinguished Professor of Civil and Construction Engineering in the Department of Civil, Construction, and Environmental Engineering at North Carolina State University, where he also serves as the Director of the Constructed Facilities Laboratory and NSF I/UCRC in Repair of Structures and Bridges with Composite (RB2C).

ACI Honorary Member Paul Zia is a Distinguished University Professor Emeritus at North Carolina State University. He served as ACI President in 1989 and is a member of several ACI committees, including ACI Committee 363, High-Strength Concrete; Joint ACI-ASCE Committees 423, Prestressed Concrete, and 445, Shear and Torsion; the Concrete Research Council; and the TAC Technology Transfer Committee as Chair of ITG-6.

Fig. 1—Test setup used for bond beams and slabs.

ACI Structural Journal/November-December 2011 3

The ready mixed concrete supplied by a local concrete producer consisted of Type I cement, Class F fly ash, ASTM C33/C33M-08 natural sand, No. 78 granite, LSP concrete additive,4 and standard high-range water reducer (HRWR). The compressive strength of the concrete was determined by using 4 x 8 in. (102 x 204 mm) cylinders cured in the same environment as the test specimens. Table 2 provides the mixture design of the two concrete mixtures.

Test resultsMaterial properties—Table 3 summarizes the measured

compressive strength of the concrete at 28 days and on the day of testing for both the bond and shear studies. It should be noted that the concrete produced for the first cast according to the given mixture design resulted in both a higher unit weight and higher compressive strength than the targeted values. It is believed that the moisture content of the

Table 1—Test results of bond experimental program

Beam IDfc, psi (MPa)

fy, ksi (MPa)

ld,required,in. (mm)

ld,provided, in. (mm)

Ratio ofld,provided to

ld,required

Predicted failure load,

kip (kN)

Maximum measured

load, kip (kN)Measured-to-predicted load

Mode of failure

Yield strain,in./in.

Maximum measured strain, in./in.

B-120-1

6000 (41.4) 74 (510) 32.3 (818)

31.0 (787) 0.96

40.1 (178)

37.8 (168) 0.94 Splitting

0.0026

0.0030

B-120-1-R 31.5 (140) 0.79 Splitting 0.0024

B-120-241.0 (1041) 1.27

40.3 (179) 1.00 Flexural 0.0060

B-120-2-R 40.6 (181) 1.01 Flexural 0.0060

B-120-351.0 (1295) 1.58

39.2 (174) 0.98 Flexural 0.0060

B-120-3-R 39.5 (176) 0.99 Flexural 0.0060

B-130-1

5500 (37.9) 74 (510) 33.8 (859)

26.0 (660) 0.77

39.9 (177)

31.8 (141) 0.80 Splitting

0.0026

0.0022

B-130-1-R 34.8 (155) 0.87 Splitting 0.0023

B-130-233.0 (838) 0.98

38.2 (170) 0.96 Splitting 0.0027

B-130-2-R 38.4 (171) 0.96 Splitting 0.0026

B-130-341.0 (1041) 1.21

39.8 (177) 1.00 Flexural 0.0060

B-130-3-R 40.2 (179) 1.01 Flexural 0.0060

B-130-1-S

6000 (41.4) 61 (421) 11.8 (300)

11.0 (279) 0.93

6.1 (27)

5.8 (26) 0.95 Splitting

0.0021

0.0021

B-130-1-R-S 6.4 (28) 1.05 Splitting 0.0060

B-130-2-S15.0 (381) 1.27

7.1 (32) 1.16 Splitting 0.0026

B-130-2-R-S 6.7 (30) 1.10 Splitting 0.0060

B-130-3-S18.0 (457) 1.52

7.3 (32) 1.20 Flexural —

B-130-3-R-S 7.8 (35) 1.28 Flexural 0.0060

Table 2—Typical concrete mixture designTarget unit weight,

lb/ft3 (kg/m3) w/cmCement,

lb/yd3 (kg/m3)Class F fly ash, lb/yd3 (kg/m3)

Natural sand, lb/yd3 (kg/m3)

No. 78 granite, lb/yd3 (kg/m3)

Concrete additive, lb/yd3 (kg/m3)

Water reducer,oz/cwt (mL/100 kg)

HRWR,oz/cwt (mL/100 kg)

By weight 120 (1922) 0.4 752 (446) 188 (112) 1380 (819) 722 (428) 9 (5.34) 4 (261) 3 (196)

By weight 130 (2082) 0.4 677 (402) 200 (119) 1675 (994) 914 (542) 4 (2.37) 4 (261) 5 (326)

Table 3—Concrete strength for test specimens

Specimens cast Measured unit weight, lb/ft3 (kg/m3)

Compressive strength, psi (MPa)

Target fc′ fc Day of Testing

Bond beams Target U.W. = 120 lb/ft3 126.6 (2030) 2500 (20.0) 5660 (39.0) 6000 (41.4)

Bond beams Target U.W. = 130 lb/ft3 132.6 (2125) 4000 (30.0) 5000 (34.5) 5500 (37.9)

Bond slabs Target U.W. = 130 lb/ft3 131.2 (2105) 4000 (30.0) 5470 (37.7) 6000 (41.4)

Shear beams Target U.W. = 120 lb/ft3 120.2 (1925) 2500 (20.0) 3620 (25.0) 4120 (28.4)

Shear beamsTarget U.W. = 130 lb/ft3 131.1 (2100) 4000 (30.0) 5930 (40.9) 6890 (47.5)

4 ACI Structural Journal/November-December 2011

(610 mm), taken from the same batches of the longitudinal reinforcement used in the beams. The No. 9 (No. 29) bars used as the longitudinal reinforcement and the No. 3 (No. 16) bars used as the transverse reinforcement of the bond specimens had yield strengths of 74 and 69 ksi (510 and 476 MPa), respectively. The yield strength of the No. 5 bars used as the longitudinal reinforcement in the slabs for the bond tests was 61 ksi (421 MPa).

Load-deflection behavior—The load-deflection relationship of the tested beams in the second group is shown in Fig. 2. The test results indicate that the beams with a splice length shorter than the required splice length failed before achieving their full design capacity. When the full splice length was provided, the beams were able to achieve their design capacity but failed due to splitting without providing much ductility. When the provided splice length exceeded the required splice length up to 1.58Ld, the beams achieved the flexural strength and a sufficient level of ductility. Similar behavior was observed for the beams tested in Group 1 and the slabs in Group 3.

Crack width—Crack width was measured by using PI gauges located at the ends of the splice and at the midspan of the specimens. A crack comparator was also used to measure the crack width at different load levels. It was observed that in all cases, the first flexural cracks developed at the two ends of the splice zone—at the maximum moment and shear locations. As the load was increased, the flexural cracks propagated toward the compression zone and increased in number and width. A further increase in the applied load led to the formation of splitting cracks parallel to the longitudinal bars (initially on the tension surface of the beam), followed by splitting cracks on the side of the beam. Measurement using the crack comparator was discontinued after the yielding of the longitudinal reinforcement; however, PI gauge measurements showed that with adequate development length according to ACI 318-08, the yield strength of the longitudinal bar was fully developed and the beam achieved considerable ductility. Figure 3 shows the width of the splitting crack and the applied loads for the first group of beams reinforced with No. 9 (No. 29) bars. The test results indicate that the splitting cracks occurred at the same range of load level for all tested beams of the same group. The initiation of the splitting cracks occurs when the stresses in the spliced bars induce forces in the concrete cover equal to the tensile strength of the concrete. The figure also shows that as the load was increased, the widths of the splitting cracks were larger for the shorter splice lengths. This was due to the higher stresses induced in the longitudinal bars with shorter splices in comparison to the bars with longer splice length at the same load levels and consequently higher induced strains in the steel and the surrounding concrete, causing wider cracks.

Splice strength—Two different modes of failure were observed for the bond specimens. The first mode was splitting failure, characterized by the formation and propagation of splitting cracks parallel to the longitudinal bars along the splice length, as shown in Fig. 4. Such failure is generally sudden and brittle. The second mode of failure was typical flexural failure initiated by the yielding of the reinforcement, followed by the final crushing of the concrete in the compression zone. A summary of the test results for Phase I of the experimental program is presented in Table 1.

The test results presented in Table 1 indicate that when the provided development length was equal to or more than

sand measured at the ready mixed concrete plant was higher than the actual moisture content, making the total water less than designed, which led to a lower water-cement ratio (w/c) and, consequently, a higher concrete compressive strength.

Tension coupons from the Grade 60 steel reinforcement were tested according to ASTM A370-096 to determine the material characteristics. The typical coupon length was 2 ft

Fig. 2—Load-deflection relationship for Group 2 bond beams.

Fig. 3—Splitting crack width for Group 1 beams.

Fig. 4—Splitting failure of Beam B-130-2-R.

ACI Structural Journal/November-December 2011 5

mode. Some of the beams were designed without transverse reinforcement to evaluate the concrete contribution to shear strength, Vc, whereas others were designed with minimum and maximum allowable transverse reinforcement ratios as specified by ACI 318-08 to cover the two levels of shear capacities. Table 4 provides details of the beams tested in shear. Each beam specimen was identified by four parameters: the first letter “S” stands for shear specimens; the following number identifies the targeted unit weight in pounds per cubic foot (120 or 130); and the third parameter defines the transverse reinforcement ratio (0%, 0.25%, and 0.5%) for the beams without transverse reinforcement, with minimum and maximum stirrups, respectively. The number “1.5” following the transverse reinforcement ratio refers to the a/d of 1.5, which is different from the beams tested with an a/d of 3.0 in the first two groups.

The test setup and cross-sectional dimensions for all shear tests are shown in Fig. 5. Two test setups were used—one for each of the selected a/d. The test setup was designed to allow each end of a beam to be tested so that the test could be replicated. The first setup was used to test the first and second groups of beams with an a/d of 3.0 by locating the applied point load at a distance “a” of 43 in. (1092 mm) from the left support, as shown in Fig. 5. The second setup was used to test the third group of beams with an a/d of 1.5 and the single applied load at a distance “a” of 22 in. (559 mm) from the left support, as shown in Fig. 5. The untested portion of each beam was cantilevered over the right support. Load was applied by using a 440 kips (1960 kN) capacity hydraulic actuator supported by a steel frame that was securely anchored to the laboratory strong floor. The load provided by the actuator was transferred by a single steel loading plate that measured 1 in. (25 mm) thick by 8 in. (203 mm) wide.

that required by ACI 318-08 using the measured material properties, the specimens were able to achieve their respective nominal flexural capacity after the yielding of the longitudinal reinforcement, as indicated by the strain measurements. When the provided development length was less than the required value, the specimens showed bond failures with slightly less capacity in comparison to the predicted nominal flexural strengths. The test results clearly confirm compliance of the LSP concrete to the ACI 318-08 provisions for the development length of the flexural reinforcement. It should be noted that an increase in the splice length does not necessarily increase the load-carrying capacity of the beam due to the nonlinear stress distribution along the splice length7 and increase only the ductility, as shown in Fig. 2. For example, in Table 3, although the development length increased from 33 in. (838 mm) for Beam B-130-2-R to 41 in. (1041 mm) for Beam B-130-3-R, the load-carrying capacity increased by only 4%.

Phase II: Shear behavior of LSP concrete membersThe shear behavior of reinforced concrete beams

containing LSP was evaluated using nine beams—each tested twice—for a total of 18 tests. The specimens were divided into three main groups with two targeted unit weights of 120 and 130 lb/ft3 (1920 and 2080 kg/m3). Similar to the bond investigation, target compressive strengths of 2500 and 4000 psi (20 and 30 MPa) were selected for the 120 and 130 lb/ft3 (1920 and 2080 kg/m3) groups, respectively. Two additional parameters considered in the experimental program were the shear span-depth ratio (a/d) and the transverse reinforcement ratio rt. The beams were designed according to ACI 318-08 provisions for shear and flexure, and the design ensured a shear-controlled failure

Table 4—Test results of shear experimental program

BeamMaximum measured

shear, kip (kN)

fc, psi

(MPa)

VC VS Vn

Measured,kip (kN)

Predicted,kip (kN)

Measured versus

predictedMeasured,kip (kN)

Predicted,kip (kN)

Measured versus

predictedPredicted,kip (kN)

Measured versus

predicted

a/d = 3.0

Group 1

S-120-0% 35 (156)

4150(28.6)

39(173)

36(160)

1.08

— — —36

(160)

0.97

S-120-0%-R 43 (191) 1.18

S-120-0.25% 93 (414) 54 (240) 31(138)

1.73 67(298)

1.38

S-120-0.25%-R 83 (369) 44 (196) 1.42 1.24

S-120-5% 126 (560) 87 (387) 62(276)

1.39 99(440)

1.28

S-120-0.5%-R 124 (552) 85 (378) 1.36 1.26

Group 2

S-130-0% 41 (182)

6750(46.5)

40(178)

42(187)

0.95

— — —42

(187)

0.97

S-130-0%-R 38 (169) 0.92

S-130-0.25% 97 (431) 57 (254) 31(138)

1.82 73(325)

1.32

S-130-0.25%-R 104 (463) 64 (285) 2.05 1.42

S-130-0.5% 130 (578) 90 (400) 62(276)

1.43 104(463)

1.24

S-130-0.5%-R 132 (587) 92 (409) 1.47 1.26

a/d = 1.5

Group 3

S-130-0%-1.5 153 (681)

6970(48.1)

56(249)

42(187)

1.32

— — —42

(187)

3.61

S-130-0%-1.5-R 206 (916) 4.87

S-130-0.25%-1.5 194 (863) 138 (614) 31(138)

4.40 74(329)

2.63

S-130-0.25%-1.5-R 159 (707) 103 (458) 3.29 2.16

S-130-0.5%-1.5 214 (952) 158 (703) 62(276)

2.53 105(467)

2.04

S-130-0.5%-1.5-R 215 (956) 159 (707) 2.54 2.05

6 ACI Structural Journal/November-December 2011

The beams were simply supported by a steel pin and a 1 in. (25 mm) thick bearing plate at the left end support and a steel roller and a 1 in. (25 mm) thick bearing plate at the right support. For each test setup, two 200 kips (890 kN) capacity load cells were placed under the left support to measure the reaction, which represented the maximum shear forces within the short shear span under consideration. One linear variable differential transducer (LVDT) was used at the top surface of the beam at both supports to measure any possible deflection at the support. Vertical beam deflections were measured by two string potentiometers placed directly under the beam at the location of the actuator and 1.5 in. (38 mm) from the edge. The crack widths and strains in selected stirrups were measured by using PI gauges and

weldable strain gauges, respectively. PI gauges were arranged in a rosette configuration to capture most of the cracks within the shear span region. For the test specimens in Groups 1 and 2, three PI gauge rosettes were arranged diagonally between the applied load and the left support reaction, as shown in Fig. 6(a), whereas only two PI rosettes were used for the shorter beams of Group 3. On the opposite face of the test specimen, PI gauges were arranged vertically at the location of the selected stirrup located within the shear span region, as shown in Fig. 6(b). The stirrup strains were measured using weldable strain gauges placed on some selected stirrups within each beam.

The same concrete mixture designs given in Table 2 were used for the shear specimens. Table 3 also summarizes the measured concrete compressive strength at 28 days and at the day of testing for the shear specimens. Tension coupons for the Grade 60 reinforcing steel were tested according to ASTM A370-09 using 2 ft (610 mm) coupons that were taken from the same batches of the transverse reinforcement used in the beams and had a yield strength of 69 ksi (476 MPa).

Test resultsLoad-deflection behavior—The load-deflection behavior

of beams with LSP concrete was similar to that of beams of normal concrete reported in the literature.8 The beams without transverse reinforcement failed shortly after the initiation of the first diagonal cracks. Test beams with transverse reinforcement were able to sustain higher load levels, as shown in Fig. 7, for the beams of Group 2. The load at the formation of the first critical diagonal crack for members without shear reinforcement was taken as the concrete contribution to shear resistance Vc for all beams within the same group. It can be observed from Fig. 7 that by increasing the transverse reinforcement ratio from 0.25 to

Fig. 5—Typical shear beam cross section and test setup.

Fig. 6—Location of PI gauge rosettes and PI gauges on back of shear beams.

Fig. 7—Shear load-deflection behavior for Group 2 shear beams.

ACI Structural Journal/November-December 2011 7

0.5%, the shear resistance of the beams was increased. The shear at failure for the specimens with shear reinforcement was taken as the nominal shear strength Vn. The steel contribution to the shear strength Vs was simply determined as the difference of (Vn – Vc). The load-deflection behavior of the beams in Group 3 is shown in Fig. 8. From this curve, it can be observed that the behavior of the specimens within Group 3 is quite different from those within Groups 1 and 2, as shown in Fig. 7. This is due primarily to the small a/d of this category of beams and the formation of a compression strut mechanism in resisting the applied shear. This arching action mechanism is the cause of the significant increase in the shear capacity of the beams within Group 3, especially for those without transverse reinforcement.

Cracking behavior—Cracking on the side of the beam was measured using PI gauge rosettes within the shear span a, whereas cracking on the opposite side of the beam was measured using the vertical PI gauges, as shown in Fig. 6. Crack width comparators were also used to measure the crack width at different load levels. It was observed that the initiation of the first flexural cracks occurred at the location of the maximum bending moment directly under the applied load. Flexural cracks propagated upwards and increased in number and width as the load was increased.

The formation of shear cracks was dependent on the presence of shear reinforcement and a/d. For beams with shear reinforcement and an a/d of 3.0, the first shear cracks appeared as an extension of the flexural shear cracks in the diagonal direction. When the applied load was increased, new diagonal shear cracks formed and were observed on both sides of the tested beam. In this study, failure of the beams was determined when the beam was no longer able to sustain any increase in the applied load. For the beams without transverse reinforcement, once the first shear crack appeared, it propagated diagonally through the shear span and suddenly increased in width by a slight increase in applied load before failure. For beams with shear reinforcement, once the first diagonal tension crack occurred, measurements of the strain in the stirrups, using the weldable strain gauges, clearly indicated the participation of the stirrups in resisting the applied load. The presence of the stirrups controlled the increase in the width of the diagonal tension crack, which led to the formation of multiple diagonal cracks within the shear span. The cracking behavior for specimens with no stirrups, minimum stirrups, and maximum stirrups is shown in Fig. 9. The figure clearly indicates a single diagonal crack, spanning from the location of the applied load to the support for the beams without stirrups and multiple diagonal cracks for the beams with stirrups. Figure 10 shows the measured width of the diagonal shear crack using the crack comparators for the second group of beams with an a/d of 3.0. It should be noted that the graph does not include the beams without transverse reinforcement, as these beams failed suddenly after the initiation of the diagonal crack. The graph also shows the effect of the transverse reinforcement ratio in controlling the crack width. It can be seen that for the same load level, the beams with the minimum transverse reinforcement ratio had a wider crack width compared to the beams with the maximum transverse reinforcement ratio. An increase of the crack width is expected due to the high stresses induced in the stirrups of the beams with the minimum transverse reinforcement ratio in comparison to the beams with the higher transverse reinforcement ratio at any given load. The

higher stress levels correspond to higher strains in the stirrups and thus wider cracks in the concrete.

Failure mode—The failure mode of each group of test beams was highly dependent on the a/d. For Groups 1 and 2 (specimens with an a/d of 3.0), the failure mode was also affected by the amount of transverse reinforcement used. For beams without

Fig. 8—Shear load-deflection behavior for Group 3 shear beams.

Fig. 9—Typical shear cracking behavior.

Fig. 10—Shear crack width for Group 2 beams.

8 ACI Structural Journal/November-December 2011

transverse reinforcement, the failure mechanism occurred as a single shear crack that extended from the support to the location of the applied load. The typical failure observed for beams with shear reinforcement was due to crushing of the concrete at the nodal region of the diagonal compression strut well after the formation of many diagonal cracks within the shear span.

The behavior of Group 3 specimens with an a/d of 1.5 was controlled, as expected, by stirrup spacing within the member. For the beams without transverse reinforcement, failure occurred at much higher-than-anticipated loads due to arching action within the short span. For the beams with the minimum amount of transverse reinforcement, although arching action controlled the short shear span, flexural shear failure was observed in the longer shear span in the uninstrumented side of the beam when the load-carrying capacity of that side was reached. When the transverse reinforcement ratio was increased, the load-carrying capacity of the longer shear span was increased and failure was due to arching action within the short shear span of the beam. The measured ultimate loads are significantly higher than the predicted values according to the ACI 318-08 design equations. When the strut-and-tie method is used as recommended by ACI 318-08 for low a/d, however, the behavior is accurately predicted.

The results from the 18 shear tests are summarized in Table 4, which shows the maximum measured shear force in comparison to the ACI 318-08 predicted values. Because the beams were designed to have a shear failure, the longitudinal reinforcement was increased to avoid premature flexural failure. To account for the increase of the flexural reinforcement, the concrete contribution to the shear strength, Vc, was predicted by using the following ACI 318-08 Eq. (11-5), which accounts for the effect of the longitudinal reinforcement ratio rw.

= l ′ + r

1.9 2500 uc c w w

u

V dV f b d

M

where l is a reduction factor used with the lightweight concrete and was assumed unity for the LSP concrete; Vu and Mu are the ultimate shear force and moment, respectively, at the section considered; bw is the width of the member; and d is the depth from the centroid of the longitudinal reinforcement to the extreme compression fiber. For the beams with an a/d of 3.0, Table 4 shows a very close correlation between the measured and predicted shear force for the case without transverse reinforcement. With transverse reinforcement, the average measured-to-predicted ratio becomes 1.20 with a coefficient of variation of 0.14, which shows that the ACI 318-08 provisions for shear can be safely used for the design of structural members with LSP additive. For the beams with an a/d of 1.5, the measured-to-predicted shear strength according to ACI 318-08 was significantly higher due to the arching action mechanism developed for this category of beams.

CONCLUSIONS AND RECOMMENDATIONSBased on the test results of the experimental program and

the comparisons to the ACI 318-08 provisions for shear and development of reinforcement, the following conclusions may be drawn:

1. The concrete containing LSP additive used in this study, with unit weights ranging between 120 and 130 lb/ft3 (1920 and 2080 kg/m3), achieved comparable compressive strength to what is commonly used for structural applications.

2. Concrete containing LSP additive was flowable and had good workability.

3. The load-deflection characteristics and the crack pattern of the beams and slabs tested in this experimental program were similar to the expected behavior of beams and slabs using normalweight concrete.

4. The cracked section analysis used to predict the behavior of regular concrete could be used to predict the behavior of concrete members produced with LSP additive.

5. ACI 318-08 equations could be applied to the design of the tension splice required to develop the design strength of the reinforcement at a critical section.

6. The analysis of the bond specimens accurately predicted the behavior of all tested beams. With sufficient splice length, the beams were capable of achieving their load-carrying capacity. The results confirmed that the bond strength of beams containing LSP concrete additive met the bond requirements specified by ACI 318-08.

7. The beams with LSP concrete additive satisfy the shear design requirements of ACI 318-08.

8. The test results also confirmed that, for the structural design of LSP concrete containing normalweight aggregates, the reduction factor l normally used for lightweight concrete do not need to be applied for LSP concrete with a unit weight equal to or higher than 120 lb/ft3 (1920 kg/m3).

Due to the limited number of bond and shear specimens used in this study, it is recommended that additional tests should be conducted to include a wider range of concrete densities. In addition, whether the use of LSP in a concrete mixture will affect the long-term bond and shear behavior of concrete members should also be investigated.

ACKNOWLEDGMENTSThe authors would like to thank NOVA Chemicals Inc. for their financial

assistance and for providing the concrete additive for the construction of the test specimens. The authors would also like to thank the Argos USA for supplying the ready mixed concrete used in the research program. Special thanks go to the staff at the Constructed Facilities Laboratory, including J. Atkinson, J. McEntire, G. Lucier, and M. Dawood for their invaluable help.

REFERENCES1. ACI Committee 318, “Building Code Requirements for Structural

Concrete (ACI 318-08) and Commentary,” American Concrete Institute, Farmington Hills, MI, 2008, 473 pp.

2. ASTM C33/C33M-08, “Standard Specification for Concrete Aggregates,” ASTM International, West Conshohocken, PA, 2008, 11 pp.

3. ASTM C330-05, “Standard Specification for Lightweight Aggregates for Structural Concrete,” ASTM International, West Conshohocken, PA, 2005, 4 pp.

4. NOVA Chemicals Inc., “Elemix Concrete Additive,” 2008, http://www.elemix.com/. (last accessed Feb. 19, 2010)

5. ASTM A615/A615M-09b, “Standard Specification for Deformed and Plain Carbon-Steel Bars for Concrete Reinforcement,” ASTM International, West Conshohocken, PA, 2009, 6 pp.

6. ASTM A370-09, “Standard Test Methods and Definitions for Mechanical Testing of Steel Products,” ASTM International, West Conshohocken, PA, 2009, 47 pp.

7. Canbay, E., and Frosch, R. J., “Bond Strength of Lap-Spliced Bars,” ACI Structural Journal, V. 102, No. 4, July-Aug. 2005, pp. 605-614.

8. Wight, J. K., and MacGregor, J. G., Reinforced Concrete: Mechanics and Design, fifth edition, Prentice Hall, Upper Saddle River, New Jersey, 2009, 1126 pp.