Bond Strength of Reinforcing Steel Bars in Self-Consolidating Concrete

Abstract

This paper presents an experimental investigation of the bond strength of reinforcing steel bars in tension in self-consolidating concrete (SCC). The effects of the reinforcing bar’s location, orientation, size, and coating on the bond strength with SCC were studied and compared to those with conventionally vibrated concrete (CVC). Several SCC mixtures were developed to cover a wide range of applications/components and material types. The fresh properties of the SCC mixtures were determined to evaluate their filling ability, passing ability and stability. Two hundred and thirty-four pull-out tests of rebars embedded in cubes, wall panels and slabs were conducted. Almost half of the tests were conducted to evaluate the bond with SCC and the other half with CVC. Load–slippage relationships were measured for each test. Pull-out test results were analyzed, and the bond strength was reported in two values: critical strength, which corresponds to slippage of 0.01 in. *0.25 mm); and ultimate strength, which corresponds to the maximum load. The critical strength of SCC and CVC were compared against the ACI 318-19 provisions and comparisons between the ultimate strength of SCC and CVC were conducted. The comparisons indicated that SCC has lower bond strength with vertical rebars than CVC, and a 1.3 development length modification factor is recommended. A similar conclusion applies to epoxy-coated and large diameter rebars. Also, SCC with high slump flow has shown a less top-bar effect than that of CVC.

1. Introduction

According to the American Concrete Institute (ACI), a bond is the shearing stress at the surface of a reinforcing bar, preventing relative movement between the bar and the surrounding concrete when the bar carries tensile force. The bond strength between the concrete and reinforcement steel plays a main role in the structural design of different reinforced concrete elements. The main assumption to design these elements is the composite behavior of the cross section, which is provided when the yield of the rebar happens before any slippage of it from the concrete. To ensure that no slippage happens, all building codes list certain criteria for the development length of rebars in tension. These criteria depend on many factors, such as the yield strength of steel, compressive strength of concrete, weight of concrete (whether it is light or of normal weight), diameter of the rebar, the confining of the reinforcement, location of the rebar, and type of the rebar coating [1]. These criteria do not include the concrete type: conventionally vibrated concrete (CVC) or self-consolidating concrete (SCC). The significant differences between SCC and CVC with respect to aggregate composition, rheology, and use of viscosity modifying admixtures (VMA) are believed to affect the characteristics of the bond with reinforcing bars. The orientation of the bars, whether parallel or perpendicular to the direction of casting, location of bars, bar diameter and coating could also have an influence on the bond strength with SCC different from that with CVC due to differences in composition and rheology of mixtures. Therefore, the objective of this study is to evaluate the bond strength of reinforcing steel with SCC and compare its behavior with that with CVC when different design parameters are used.

2. Literature Review

Two criteria are used to define the bond strength between concrete and reinforcing steel depending on the purpose of testing. First, the ultimate bond stress, which is the stress at the maximum load. Second, the critical bond stress corresponding to slippage of 0.01 in. (0.25 mm) as proposed by Mathey and Watstein [2], which is more appropriate for design purposes [3]. Pull-out tests are usually used to measure the ultimate and critical bond stresses. Three factors resist the slippage of a rebar in concrete: (1) adhesion between concrete and steel, (2) friction due to rebar confinement, and (3) bearing at bar deformations. Two failure modes are expected: (1) slippage due to shear of the confining concrete [4], (2) splitting in the confining concrete [5].

König et al. [6,7] and Almeida et al. [8] found that CVC performed better in bond tests and achieved 15 to 20% greater bond strength than that of SCC. Almeida et al. [9] evaluated the bond behavior of SCC by varying compressive strength and steel bar diameter in pull-out and beam tests. The comparison between the test results and code equations showed that the same equations adopted for CVC can be used for SCC, which means that bond properties of SCC are similar to those of CVC. Hassan et al. [4] reported that the normalized bond stress was slightly higher in SCC than that in CVC at 3, 7, 14 and 28 days in pull-out tests. Aslani and Nejadi [10] reported that the bond strength of SCC is as high as the bond strength of CVC when large bar diameters are studied. For smaller bar diameters, the bond strength of SCC is slightly higher, with the largest difference occurring for the smallest bar diameters.

Valcuende and Parra [11] conducted pull-out tests and calculated the mean bond strength as the arithmetic mean of the stresses recorded for slips of 0.0004, 0.004 and 0.04 in. (0.01, 0.1, and 1 mm), which was found to be 30% greater in SCC than in 4.5 ksi (31 MPa) CVC, but only 10% greater than in 9 ksi (62 MPa) CVC. The enhanced cohesiveness of SCC ensures a better suspension of solid particles in the fresh state and this, therefore, produces good deformability and filling capability. Bleeding, segregation and surface settlement as a result of a high water-to-cement ratio (w/c) or excessive vibration are generally not factors considered in SCC, which explains the higher bond strength even in deep members [4]. Gibbs and Zhu [12] reported a 4% difference in bond strength between the two types of concrete, Wang and Zheng [13] reported 9%, and Daoud et al. [14] reported 5% higher bond strength for SCC. Zhu et al. [15] reported that the normalized bond strengths of the SCC mixes were found to be about 10–40% higher than those of CVC mixes with the same strength, while the maximum bond strength decreased when the diameter of the steel bar increased from 1/2 to 3/4 in. (13 to 19 mm). Cattaneo and Rosati [5] found SCC exhibits higher bond strength and, compared to CVC, requires a larger concrete cover to attain pull-out failure. Desnerck et al. [16]) conducted beam tests to evaluate the bond of reinforcement in SCC and CVC and found that for the same compressive strength, the bond strength of SCC is as high as that of CVC for large bar diameters, or slightly higher than that of CVC for smaller bar diameters. The bar diameters ranged from 1/2 to 1.5 in. (13 to 38 mm).

Most researchers agreed that SCC still shows the top-bar effect, but the extent is lower than or similar to CVC, and for concretes of more than 7 ksi (50 MPa), the differences between SCC and CVC are not significant. Khayat [17] found VMA helped reduce surface settlement related to bleeding and segregating and significantly reduced the top-bar effect. Attiogbe et al. [18] concluded that highly stable SCC mixtures have a level of top-bar effect for deformed bars that is similar to that of 4 to 6 in. (100–150 mm) slump concrete. Chan et al. [3] conjectured that the plastic settlement during the hardening of SCC may still cause the top-bar effect and reported that fewer top-bar effects were found for SCC in the pull-out tests than for CVC. Castel et al. [19] concluded that the bond strength of SCC is not affected by the orientation of the bars. For the top casting surface, the maximum ultimate bond strengths obtained were approximately 20% higher for SCC than for CVC, regardless of the concrete strength. Hassan et al. [4] reported that the bond stress–slip curve showed similar trends of variation for both SCC and CVC pull-out specimens in the bottom bars. Higher bond stress and stiffness in the top and middle bars were observed in SCC compared to CVC. Trezos et al. [20] found that the top-bar effect seems to be less intense in SCC when stress corresponding to slip of 0.01 in. (0.25 mm) is selected as the bond strength. Esfahani et al. [21] studied the effect of bar position on the bond strengths of reinforcing bars using pull-out tests with top, middle and bottom bars. It was found that the local bond strength of bottom cast bars was almost the same for CVC and SCC, but for the top cast bars, the local bond strength for SCC was about 20% less than that for CVC.

Due to the disagreements in the literature on the bond strength of reinforcing rebars in SCC compared to CVC, an extensive experimental investigation was conducted to study the bond strength when a wide range of SCC materials, proportions, and characteristics are used.

3. Materials and Mixture Proportioning

3.1. Materials

Different types of coarse aggregates and supplementary cementitious materials (SCMs) were used in this study to cover the variance in materials’ availability based on location [22]. Portland Cement type I/II was used in developing the mixtures in this study, which is commonly used in construction. Two types of coarse aggregate (i.e., crushed limestone and natural gravel) and natural sand were used for all mixtures. For each aggregate type, three nominal maximum sizes of aggregates, NMSA, 3/4, 1/2, and 3/8 in. (19, 13, and 10 mm), were used to represent the sizes used in different concrete components. The physical properties of the aggregates and their combinations are presented in

Table 1. Physical properties of aggregates and combination (1 in. = 25.4 mm, 1 pcf = 16 kg/m³).

Property	Limestone			Gravel			Sand
	3/4 in.	1/2 in.	3/8 in.	3/4 in.	1/2 in.	3/8 in.
Specific Gravity	2.66	2.66	2.66	2.74	2.68	2.69	2.62
Absorption	1.3%	1.3%	1.3%	1.1%	1.4%	1.4%	0.5%
Sand-to-Aggregate Ratio	0.45	0.47	0.50	0.45	0.47	0.50	N/A
Combined Aggregate Unit Weight (pcf)	117	118	118	127	124	123	N/A
Percent of Voids	29.0	28.4	28.4	23.7	25.9	27.0	N/A

. Figure 1 shows the particle size distribution of fine and coarse aggregates.

In addition to Portland Cement type I/II, three types of SCMs (i.e., Class C fly ash, Class F fly ash, and GGBFS), and one filler (i.e., limestone powder) were used. The chemical compositions of cement, SCMs, and filler are listed in

Table 2. Chemical compositions of cement, SCMs and filler.

Component	Component Content by Percentage of Weight
	Type I/II Cement	Class C Fly Ash	Class F Fly Ash	GGBFS	Limestone Powder
SiO2	20.10	42.46	50.87	31.63	1.56
Al2O3	4.44	19.46	20.17	11.30	-
Fe2O3	3.09	5.51	5.27	0.34	0.48
SO3	3.18	1.20	0.61	3.30	1.77
CaO	62.94	21.54	15.78	41.31	52.77
MgO	2.88	4.67	3.19	10.77	0.48
Na2O	0.10	1.42	0.69	0.19	0.03
K2O	0.61	0.68	1.09	0.36	0.09
P2O5	0.06	0.84	0.44	0.02	-
TiO2	0.24	1.48	1.29	0.56	-
SrO	0.09	0.32	0.35	0.04	-
BaO	-	0.67	0.35	-	-
LOI	2.22	0.19	0.07	-	42.50

. The particle size distribution of cement, SCMs, and filler are presented in Figure 2. Chemical admixtures were used to control the rheological properties and durability of SCC mixtures, which include a polycarboxylate-type high-range water-reducing admixture (HRWRA) that meets the requirements of the ASTM C494 type F admixture; a viscosity-modifying admixture (VMA) and workability-retaining admixture that meets the requirements of ASTM C494 type S admixture; and an air-entraining admixture (AEA) that meets the requirements of ASTM C260.

3.2. Proportioning

Proportioning SCC mixtures is different from proportioning CVC mixtures as workability targets, in contrast to compressive strength, usually control the proportioning of the mixture. Workability targets were identified based on the geometric characteristics of the component and production and placement conditions. The geometric characteristics of a component include the length, depth, thickness, shape intricacy, formed surface quality, and level of reinforcement (i.e., intensity and spacing). Production and placement conditions include mixing energy, transport time, placement technique, and temperature. For simplicity, each of the geometric characteristics was classified as either “high” or “low” [22].

Table 3. Classes of component geometric characteristics (1 in. = 25.4 mm, 1 ft = 0.305 m).

Component Geometric Characteristics	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 travelling public
	High	Exposed to the travelling public
Level of Reinforcement	Low	Large spacing between bars (≥3 in.)
	High	Small spacing between bars (<3 in.)

shows the value/definition used to describe the classes of each geometric characteristic based on the literature [23,24]. Similarly, two classes were used to describe each of the three key workability properties of SCC: filling ability (FA), segregation resistance (SR), and passing ability (PA).

Table 4. Classes of SCC workability properties (1 in. = 25.4 mm).

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
Passing Ability (PA)	PA1	80% > Filling Capacity ≥ 70% 2 in. < J-Ring ΔD ≤ 4 in. 0.6 in. < J-Ring ΔH ≤ 0.8 in.	Wide spacing between reinforcing bars
	PA2	Filling Capacity ≥ 80% J-Ring ΔD ≤ 2 in. J-Ring ΔH ≤ 0.6 in.	Narrow spacing between reinforcing bars
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

shows the value/range of the parameters used to describe the two classes of each workability property based on the literature [23,24,25,26]. These values/ranges might be adjusted according to the production and placement conditions [27]. To determine which workability target value/range applies to a specific component, the decision tree shown in Figure 3 is used. This decision tree provides guidelines on workability targets based on the geometric characteristics of the concrete component. The three-digit identification shown at the bottom of the tree represents the target workability with respect to filling ability, segregation resistance, and passing ability classes, respectively. For example, 111 means FA1, SR1, and PA1.

Several approaches for proportioning SCC mixtures were reviewed and evaluated [25,27,28,29,30,31,32]. The procedure proposed by Koehler and Fowler [31] was chosen because it considers the effect of aggregate gradation, shape, and angularity, and uses standard workability test methods to identify the necessary parameters [33]. Two steps were added to the procedure to provide guidance on the water content requirements for different NMSA according to ACI 211 [34], and to verify that powder content and aggregate volume are within the recommended ranges of ACI 237 [25]. Forty normal-weight SCC mixtures containing two types of coarse aggregate (i.e., crushed limestone and natural gravel) with three NMSA, three types of SCMs (i.e., Class C fly ash, Class F fly ash, and GGBFS), and one filler (i.e., limestone powder) were designed to be used in the experimental investigation. Six normal-weight CVC mixtures were proportioned according to ACI 211 [34] procedures for the two types of coarse aggregate with three gradations each (No. 67, No. 79, and No. 8) for comparison, as shown in

Table 5. Proportions of SCC and CVC mixtures containing limestone aggregate.

Mixture Type	SCC Mixtures																				CVC Mixtures
SCMs/Fillers	25% Class C Fly Ash					25% Class F Fly Ash					30% GGBFS					20% Class F Fly Ash + 15% LSP					25% Class F Fly Ash
Flowability	Low Slump Flow		High Slump Flow			Low Slump Flow		High Slump Flow			Low Slump Flow		High Slump Flow			Low Slump Flow		High Slump Flow			2–4 in. Slump
NMSA, in.	3/4	1/2	3/4	1/2	3/8	3/4	1/2	3/4	1/2	3/8	3/4	1/2	3/4	1/2	3/8	3/4	1/2	3/4	1/2	3/8	3/4	1/2	3/8
Mixture ID	111	121	211	221	222	111	121	211	221	222	111	121	211	221	222	111	121	211	221	222	No. 67	No. 78	No. 8
Cement Type I/II, lb/cy	531	535	568	572	587	531	535	568	572	587	521	525	539	543	558	456	460	488	491	504	494	553	572
SCM, lb/cy	177	178	189	191	196	177	178	189	191	196	223	225	231	233	239	140	141	150	151	155	165	184	191
Filler, lb/cy	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	105	106	113	113	116	0	0	0
Coarse Agg., lb/cy	1542	1462	1518	1439	1334	1542	1462	1518	1439	1334	1542	1462	1530	1450	1345	1542	1462	1518	1439	1334	1674	1485	1350
Natural Sand, lb/cy	1262	1297	1242	1276	1334	1262	1297	1242	1276	1334	1262	1297	1252	1286	1345	1262	1297	1242	1276	1334	1193	1271	1356
Water, lb/cy	280	295	280	295	305	280	295	280	295	305	280	295	280	295	305	280	295	280	295	305	280	295	305
HRWRA, oz/cwt	12.0	14.0	12.0	16.0	13.0	6.0	4.0	8.0	8.0	13.0	12.0	10.0	18.0	16.0	15.0	11.0	9.0	12.0	12.0	15.0	0.0	0.0	0.0
VMA, oz/cwt	0.0	0.0	6.0	0.0	0.0	3.0	0.0	3.0	6.0	0.0	0.0	0.0	3.0	3.0	0.0	0.0	0.0	3.0	6.0	0.0	0.0	0.0	0.0
AEA, oz/cwt	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5
Total Weight, lb/cy	3792	3767	3797	3772	3756	3792	3767	3797	3772	3756	3828	3803	3832	3807	3792	3786	3761	3790	3765	3749	3806	3788	3774
Total Aggregate, lb/cy	2804	2759	2760	2714	2669	2804	2759	2760	2714	2669	2804	2759	2782	2737	2691	2804	2759	2760	2714	2669	2867	2756	2706
Total Powder, lb/cy	708	713	757	763	783	708	713	757	763	783	744	750	770	776	797	702	707	751	756	776	659	738	763
W/P Ratio	0.40	0.41	0.37	0.39	0.39	0.40	0.41	0.37	0.39	0.39	0.38	0.39	0.36	0.38	0.38	0.40	0.42	0.37	0.39	0.39	0.43	0.40	0.40
S/A Ratio	0.45	0.47	0.45	0.47	0.50	0.45	0.47	0.45	0.47	0.50	0.45	0.47	0.45	0.47	0.50	0.45	0.47	0.45	0.47	0.50	0.42	0.46	0.50
Paste Volume %	37.0%	38.0%	38.0%	39.0%	40.0%	37.0%	38.0%	38.0%	39.0%	40.0%	37.0%	38.0%	37.5%	38.5%	39.5%	37.0%	38.0%	38.0%	39.0%	40.0%	36.0%	38.5%	39.6%
Coarse Agg. Vol. %	34.4%	32.6%	33.9%	32.1%	29.8%	34.4%	32.6%	33.9%	32.1%	29.8%	34.4%	32.6%	34.1%	32.4%	30.0%	34.4%	32.6%	33.9%	32.1%	29.8%	37.4%	33.1%	30.1%

and

Table 6. Proportions of SCC and CVC mixtures containing gravel aggregate.

Mixture Type	SCC Mixtures																				CVC Mixtures
SCMs/Fillers	25% Class C Fly Ash					25% Class F Fly Ash					30% GGBFS					20% Class F Fly Ash + 15% LSP					25% Class F Fly Ash
Flowability	Low Slump Flow		High Slump Flow			Low Slump Flow		High Slump Flow			Low Slump Flow		High Slump Flow			Low Slump Flow		High Slump Flow			2–4 in. Slump
NMSA, in.	3/4	1/2	3/4	1/2	3/8	3/4	1/2	3/4	1/2	3/8	3/4	1/2	3/4	1/2	3/8	3/4	1/2	3/4	1/2	3/8	3/4	1/2	3/8
Mixture ID	111	121	211	221	222	111	121	211	221	222	111	121	211	221	222	111	121	211	221	222	No. 67	No. 78	No. 8
Cement Type I/II, lb/cy	494	498	568	572	587	494	498	568	572	587	485	489	539	543	558	440	444	488	491	504	459	516	534
SCM, lb/cy	165	166	189	191	196	165	166	189	191	196	208	209	231	233	239	135	137	150	151	155	153	172	178
Filler, lb/cy	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	102	102	113	113	116	0	0	0
Coarse Agg., lb/cy	1580	1497	1530	1450	1344	1580	1497	1530	1450	1344	1580	1497	1543	1462	1355	1567	1486	1530	1450	1344	1674	1485	1350
Natural Sand, lb/cy	1292	1328	1252	1286	1344	1292	1328	1252	1286	1344	1292	1328	1262	1296	1355	1282	1317	1252	1286	1344	1277	1358	1455
Water, lb/cy	280	295	280	295	305	280	295	280	295	305	280	295	280	295	305	280	295	280	295	305	260	275	285
HRWRA, oz/cwt	5.0	5.0	9.0	5.0	8.0	7.0	4.0	7.0	5.0	5.5	6.0	5.0	10.0	7.0	7.5	3.0	3.0	6.0	7.5	6.0	0.0	0.0	0.0
VMA, oz/cwt	0.0	0.0	3.0	0.0	3.0	0.0	0.0	2.0	3.0	3.0	0.0	0.0	3.0	0.0	0.0	0.0	0.0	2.0	3.0	0.0	0.0	0.0	0.0
AEA, oz/cwt	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5	1.5
Total Weight, lb/cy	3811	3784	3819	3793	3776	3811	3784	3819	3793	3776	3844	3818	3854	3829	3812	3807	3781	3813	3787	3769	3823	3806	3803
Total Aggregate, lb/cy	2872	2825	2782	2736	2688	2872	2825	2782	2736	2688	2872	2825	2805	2758	2711	2849	2803	2782	2736	2688	2951	2843	2805
Total Powder, lb/cy	659	664	757	763	783	659	664	757	763	783	692	698	770	776	797	677	683	751	756	776	612	688	713
W/P Ratio	0.43	0.44	0.37	0.39	0.39	0.43	0.44	0.37	0.39	0.39	0.40	0.42	0.36	0.38	0.38	0.41	0.43	0.37	0.39	0.39	0.43	0.40	0.40
S/A Ratio	0.45	0.47	0.45	0.47	0.50	0.45	0.47	0.45	0.47	0.50	0.45	0.47	0.45	0.47	0.50	0.45	0.47	0.45	0.47	0.50	0.43	0.48	0.52
Paste Volume %	36.0%	37.0%	38.0%	39.0%	40.0%	36.0%	37.0%	38.0%	39.0%	40.0%	36.0%	37.0%	37.5%	38.5%	39.5%	36.5%	37.5%	38.0%	39.0%	40.0%	33.9%	36.3%	37.4%
Coarse Agg. Vol. %	34.7%	32.9%	33.6%	31.9%	29.5%	34.7%	32.9%	33.6%	31.9%	29.5%	34.7%	32.9%	33.9%	32.1%	29.8%	34.5%	32.7%	33.6%	31.9%	29.5%	37.4%	33.1%	30.1%

[22].

4. Experimental Investigations

The bond strength of SCC was evaluated experimentally in three phases, as listed in

Table 7. Summary of the testing phases.

Testing Phase	Bar Orientation	Bar Size	No. of Tested Bars	Purpose
PHASE I	Vertical	#6 (19M)	36	Compare SCC vs. CVC
PHASE II	Horizontal	#6 (19M)	54	Evaluate top bar and SCC flowability effects
PHASE III	Vertical and Horizontal	#5 (16M) #8 (25M)	144	Evaluate bar location, orientation, coating, and size effects

. In PHASE I, pull-out testing was conducted on six SCC mixtures and six CVC mixtures to evaluate the bond strength of black (uncoated) deformed vertical reinforcing steel bars in tension according to RILEM/CEB/FIB [35]. For each concrete type, three mixtures containing crushed limestone aggregate with ³/₄, ¹/₂, and ³/₈ in. (19, 13, and 10 mm) NMSA, and three mixtures containing gravel aggregate with ³/₄, ¹/₂, and ³/₈ in. (19, 13, and 10 mm) NMSA were tested. All mixtures had the same SCM (25% Class F fly ash) and three specimens were tested from each mixture (total number of specimens was 36). In each specimen, #6 (19M) Grade 60 deformed bar was located vertically (such as in columns) at the center of an 8 in. (200 mm) cube and concrete was placed in the wooden form shown in Figure 4. A rigid plastic sheathing was attached to the top 4.25 in. (108 mm) of the bar, resulting in a bonded length of 3.75 in. (95 mm) (five times bar diameter). The forms were stripped after 24 h and the specimens were then moist-cured until day 28.

A pull-out force was applied at a rate of 0.05 in./min. (1.3 mm/min) and the slip at the other end of the bar was measured using two linear variable differential transformers (LVDTs), as shown in Figure 5. The average compressive strength of the specimen ranged from 4.0 to 8.7 ksi (28 to 60 MPa) at the time of testing. The bond strength of each specimen was calculated at different slippage values: first, the critical bond strength corresponding to slippage of 0.01 in. (0.25 mm); and second, the ultimate bond strength at the maximum load, which is convenient for comparison purposes.

According to CEB-fib [36], the bar direction has a significant effect on the bond strength as horizontal bars have a larger area under which bleed water could accumulate in addition to the surface settlement due to lack of static stability. These effects can significantly lower the bond strength of horizontal bars compared to that of vertical ones. To evaluate the bond strength of horizontal bars (such as in beams) as well as the top-bar effect, in PHASE II, six 48 in. × 48 in. × 8 in. (1.2 m × 1.2 m × 0.2 m) wall specimens, shown in Figure 6, were cast: two walls using high-slump-flow SCC, two using low-slump-flow SCC, and two using CVC. Each wall specimen had 9 #6 (M19) bars located horizontally in three rows, bottom (B), center (C), and top (T), as shown in Figure 7. The specimens were cast from the top with the flow of concrete perpendicular to the bar direction. Figure 8 shows the test setup using chuck’s barrel and wedges of 0.7 in. (17.8 mm) diameter prestressing strand to restrain the movement of the hydraulic jack. Using the same procedures used for the pull-out of vertical bars, a total of 54 #6 (19M) Grade 60 black deformed horizontal steel bars were pulled out. These specimens were made using ready-mixed concrete and had an average concrete strength ranging from 7.1 to 8.3 ksi (49 to 57 MPa) at the time of testing. The test results indicated the effect of bar location and concrete type on the bond strength of horizontal bars.

To provide a more comprehensive evaluation of the bond strength of horizontal and vertical bars, in PHASE III, twelve 48 in. × 48 in. × 8 in. (1.2 m × 1.2 m × 0.2 m) wall specimens and two 96 in. × 48 in. × 8 in. (2.4 m × 1.2 m × 0.2 m) slab were cast. A set of six walls and one slab were cast using high-slump-flow SCC and a similar set using CVC. The bars used in each set are #8 (25M) black (B), #8 (25M) epoxy-coated (C), and #5 (16M) black (b). Each wall specimen had nine bars located horizontally in three rows as shown in Figure 7 and was cast perpendicularly at the bar orientation. Two walls with typical embedded bars were cast for each bar size and type. One slab for each concrete type had 18 different bars located vertically in six rows with the arrangement shown in Figure 9 and was cast in parallel to the bar orientation. Figure 10 shows the test setup for #8 bars. This figure shows the use of a mechanical bar splice to restrain the movement of the hydraulic jack. Using the same procedures used for the former test, a total of 144 Grade 60 steel bars were pulled out as listed in

Table 8. Test matrix showing the number of bond strength tests for each category (PHASE III).

ID	No. of Specimens	Concrete Type		Bar Orientation		Bar Size and Coating
		CVC	SCC	Vertical	Horizontal	#5 Black	#8 Black	#8 Epoxy-Coated
S1	1	18		18		6	6	6
W1	2	18			18		18
W2	2	18			18			18
W3	2	18			18	18
S2	1		18	18		6	6	6
W4	2		18		18		18
W5	2		18		18			18
W6	2		18		18	18
TOTAL		72	72	36	108	48	48	48

. These specimens were made using ready-mixed concrete and had an average concrete strength of 4.4 to 7.4 ksi (30 to 51 MPa) for CVC and SCC, respectively, at the time of testing and slump of 2 in. (50 mm) for CVC and 26.25 in. (667 mm) for SCC. The testing results were used to evaluate the effect of bar location, concrete type, and bar orientation on the bond strength.

5. Bond Test Results

The bond strength is obtained by dividing the load by the surface area of the embedded bar’s length. The bond strength is calculated for two loading stages: load corresponding to 0.01 in. (0.25 mm) slippage; and maximum load. The results of bond strength of these two loading stages are summarized in tables in the appendix section for the three testing phases. The following subsections discuss these results and explain the effect of concrete type, bar location, size, orientation, and coating on the bond strength.

5.1. Bond Strength of SCC and CVC

The results of the PHASE I pull-out tests were used to compare the ultimate bond strength of SCC and CVC. Figure 11 shows the pull-out bond strength of 36 vertical deformed #6 reinforcing bars in tension versus $\sqrt{f_c^{\prime }}$ for SCC and CVC mixtures. The two linear relationships indicate that the pull-out bond strength of SCC was consistently lower than that of CVC, which agrees with earlier studies [6,7,9]. Therefore, a bond strength modification factor of 1.3 is proposed for vertical bars when used with SCC mixtures. Pull-out test results shown in Appendix A also indicate that mixtures containing limestone aggregate exhibited slightly higher bond strength than those containing gravel aggregate.

5.2. Top-Bar Effect

The results of PHASE II pull-out tests were used to evaluate the top-bar effect in SCC and CVC mixtures. These results are shown in Appendix A. Figure 12 shows the normalized bond strength by dividing it by $\sqrt{f_c^{\prime }}$ to compare the change in bond strength of 54 horizontal deformed reinforcing bars with height. This figure indicates that there was no significant difference in the bond strength of horizontal bars between low-slump-flow SCC and CVC mixtures, but there was a slight difference between low-slump-flow SCC and high-slump-flow SCC. The one-way analysis of variance (ANOVA) of pull-out test data for the three groups of mixtures confirmed this conclusion at a 95% confidence level (p = 0.38). The figure also shows a reduction in the bond strength as the distance from the bottom of the form increases (top-bar effect). This effect was more evident in CVC and low-slump-flow SCC mixtures than it was in high-slump-flow SCC mixtures, indicating that the top-bar effect was dependent on the rheological properties of SCC.

Due to the high scatter of the previous results, additional pull-out tests were conducted in PHASE III to confirm the top-bar effect on the bond strength in CVC and SCC. Figure 13 shows the variation in bond strength with the height of different bar configurations in CVC and SCC. Each point represents the average of the results of three bars. These figures show that the upper bars have less bond strength than the bottom ones for both CVC and SCC. To provide a rational evaluation of the effect of concrete type on the top-bar effect, a statistical analysis was conducted and indicated that there is no statistically significant difference between CVC and SCC with respect to top-bar effect. Therefore, the development length modification factor of 1.3 for top horizontal bars with more than 12 in. (300 mm) of fresh concrete cast below them could conservatively be applied to SCC regardless of the slump flow.

5.3. Bond of Horizontal Bars

Figure 14 and Figure 15 present comparisons between the bond strength of horizontal bars in CVC and SCC at 0.01 in. (0.25 mm) slippage (critical load) and ultimate load, respectively. At the critical loading stage, #8 (25M) and #5 (16M) black bars show similar bond strength in CVC and SCC. Only #8 (25M) epoxy-coated bars presented higher critical bond strength in SCC than CVC. At the ultimate loading stage, all bars show lower bond strength in SCC than CVC by approximately 17%. Figure 16 and Figure 17 show the variance between the critical and ultimate bond strength of horizontal bars in CVC and SCC, respectively. They show that the bars in CVC develop significant bond strength after 0.01 in. (0.25 mm) slippage, especially for #8 (25M) bars, which is dissimilar to those in SCC. This explains the lower ultimate bond strength of SCC at the ultimate loading stage after being similar to CVC or lower at the critical loading stage.

5.4. Bond of Vertical Bars

Figure 18 and Figure 19 present comparisons between the bond strength of vertical bars in CVC and SCC at 0.01 in. (0.25 mm) slippage and ultimate load, respectively. At the critical loading stage, #8 (25M) black bars show similar bond strength in CVC and SCC. However, #5 (16M) black and #8 (25M) epoxy-coated bars presented lower and higher, respectively, critical bond strength in SCC than in CVC. At the ultimate loading stage, #5 (16M) black bars show lower bond strength in SCC than CVC by 11%. ANOVA confirmed that no significant difference at a 95% confidence level existed between the ultimate bond strength of the #8 (25M) black and epoxy-coated vertical bars in CVC (p = 0.06) and SCC (p = 0.91). Figure 20 and Figure 21 show the critical and ultimate bond strength of vertical bars in CVC and SCC, respectively. They show that vertical epoxy-coated bars in CVC and SCC develop significant bond strength after 0.01 in. (0.25 mm) slippage. This explains the lower ultimate bond strength of SCC after being similar to CVC or lower at critical stage. ANOVA confirmed that there is no significant difference between the critical and the ultimate bond strength of #5 (16M) black bars in SCC mixtures (p = 0.13).

5.5. Effect of Bar Orientation

Figure 22 and Figure 23 show the comparison between the bond strength of the vertical and horizontal bars at the critical and ultimate loading stages in CVC and SCC, respectively. The horizontal bars represent those located at the bottom 8 in. (200 mm) of the wall’s height. These figures show that 91% and 83% of the CVC and SCC data points, respectively, are ±20% away from the line of equality. This indicates that bar orientation has a slightly more pronounced effect on the bond strength in SCC than in CVC. ANOVA was performed to evaluate the significance of the difference between the two orientations at a 95% confidence level. The analysis results indicated that there is a significant difference between the bond strength of vertical and horizontal black bars in CVC mixtures, while in SCC mixtures, the significant difference is between the bond strength of vertical and horizontal epoxy-coated bars.

5.6. Bond Strength Prediction

The design provisions in ACI 318 [1] for the development and splice length of straight reinforcement in tension are based on the expressions developed by Orangun, Jirsa, and Breen in the ACI 408R report [37]. This model predicts the bond strength of bars without confining transverse reinforcement by testing beams under flexure. This model expresses the average bond strength at failure ($u_c)$ and is represented by the following expression:

$$\frac{u_c}{\sqrt{f_c^{\prime }}}=1.22+3.23\frac{c_{min}}{d_b}+53 \frac{d_b}{l_d}$$

where c_min = smaller value of the minimum concrete cover or half of the clear spacing between bars; l_d = development or splice length; d_b = bar diameter; and $f_c^{\prime }$ = concrete compressive strength.

This expression was used to evaluate the critical bond strength from the test results of this study. Figure 24, Figure 25, Figure 26 and Figure 27 show the measured and predicted values of the normalized bond strength for different bar configurations. These figures indicate that this expression overestimates the bond strength of all tested bars, except #8 (25M) black bars in both CVC and SCC mixtures (Figure 25) and #8 (25M) epoxy-coated horizontal bars in SCC (Figure 27).

6. Conclusions

Based on the results of the experimental investigation presented in this study, the following conclusions can be drawn:

• The pull-out bond strength of horizontal reinforcing steel bars cast in high-slump-flow SCC was similar to that of bars cast in CVC, but the pull-out bond strength of horizontal reinforcing steel bars cast in medium- and low-slump-flow SCC was lower than that of bars cast in CVC.
• A development length modification factor of 1.3 was recommended to account for the difference between the bond strength of SCC and CVC.
• The top-bar effect was lower in high-slump-flow SCC than that in low-slump-flow SCC and CVC. Therefore, the development length modification factor of 1.3 (ACI 318-19) for top horizontal bars with more than 12 in. (300 mm) of fresh concrete cast below them could conservatively be applied to SCC regardless of the slump flow.
• The orientation of the bar could have a significant effect on the bond strength, in both CVC and SCC, depending on bar’s size and coating.
• The bond strength prediction using ACI 408R–03 overestimates the bond strength of # 6 (19M) black bars in SCC, #8 (25M) epoxy-coated in CVC and #5 (16M) black bars in both concrete types.