Tests of high-performance fiber-reinforced concrete beams with different shear span-to-depth ratios and main longitudinal reinforcement

Abstract

Test results of 12 reinforced concrete (RC) beams having three shear span to effective depth ratios and two main longitudinal reinforcement amounts are reported. Six of the test specimens were produced with engineered cementitious composites (ECCs) and another corresponding six with ultra high-strength fiber-reinforced cementitious composite (RMC). The shear span to effective depth ratios of test specimens were selected low to investigate the shear performance of ECCs and RMC characterized by ultra high ductility-moderate strength and ultra high strength-moderate ductility, respectively. Shear-carrying capacity and ductility ratios of large-scale reinforced ECC and RMC beams were compared and interpreted for different shear spans, energy absorption capacities, and stiffnesses. The experimental results showed that for the three different shear spans, the RMC beams had higher shear capacity results than the ECC beams. Using both concrete types markedly restricted shear cracking and led to flexural failure. Overall findings suggest that RMC beams have higher shear capacity and yield stiffness than ECC beams, while ECC beams have a higher deflection ductility ratio and energy absorption capacity.

Keywords

Engineered cementitious composites ultra high-strength fiber-reinforced cementitious composite shear ductility

Introduction

The shear behavior of reinforced concrete (RC) elements has been extensively investigated by researchers for more than 100 years.^1,2 However, no rational method has been developed that can accurately assess the shear capacity of RC structural elements; too many parameters such as type of applied loading, section shape, and ratio of shear span to effective depth are related to shear behavior. Moreover, current codes for calculating shear strength of RC elements can vary by a factor of more than two.^1,2 However, variations in flexural strength calculated by the same codes are not different by 10%.¹ Unlike pure flexure, experiments cannot be performed on reinforced concrete beams subjected to pure shear. Therefore, it is not easy to predict shear behavior in general, since responses vary from section to section along the beam shear span.²

High-strength concretes have become increasingly popular since they allow reduced cross sections, longer spans, and improved durability for many infrastructures. However, shear capacity of conventional concretes does not increase with higher strength grades. Furthermore, shear failure of concrete structures is generally brittle, and the brittleness gets significantly higher with increased concrete compressive strength. Brittleness of high-strength concretes is also observable from the post-peak portion of stress–strain diagrams, which descends very sharply.^3,4 However, the inverse relationship between strength and ductility appears to be counteracted when fibers are included in cementitious systems. Randomly and/or regularly oriented fibers affect excessive loading by arresting microcracking and turning a brittle and sudden failure into a ductile and highly energy-absorbent one. As in brittle behavior, this improvement in ductility is also very noticeable from the significantly increased post-peak strain and descending portion of stress–strain curves of such materials. Along these lines, high-performance fiber-reinforced cementitious composites (HPFRCC) have emerged⁵ as a result of many attempts to tailor the brittleness of conventional concrete mixtures.

Engineered cementitious composites (ECCs) are special types of HPFRCCs, designed with micromechanical constraints to achieve strain-hardening response instead of tension softening after first cracking strain is exceeded. As a direct result of strain-hardening response, fracture energies and ultimate tensile strain values (i.e. ductility), which can be two orders of magnitude higher than in conventional cementitious materials, are obtainable with a fiber volume fraction of only 2%.^6–9 The superior ductility of ECCs is realized by the formation of many closely spaced multiple microcracks (with widths of less than 100 µm) and it is at least 100 times higher than conventional concrete used for common construction practices.⁷ Moreover, this level of ductility is rather easy to obtain without any sacrifice in compressive strength levels equivalent to normal concrete grades. As another special member of HPFRCCs, Ryerson mix concrete (RMC) was first introduced by the researchers of Ryerson University in Toronto, Canada.¹⁰ RMCs are characterized by their relatively high compressive strength and moderate tensile ductility. It should be mentioned that although RMCs exhibit tensile ductility lower than that of standard ECC mixtures due to tension softening behavior, their compressive strength is considerably higher and can reach levels of more than 150 MPa after 28 days of age. Although a few studies targeting at the evaluation of ECC and concrete mixtures with very high-strength levels and moderate ductility (similar to RMC herein),^11–14 no studies focusing on the load–deflection behavior, strength, initial stiffness, deflection ductility ratio, and energy absorption capacity of large-scale ECC and RMC beams were encountered. In this paper, this gap in literature was tried to be filled. Taking into account the individual characteristics of different concrete types, performance of ECCs and RMCs was evaluated in terms of shear behavior. The main motivation was to determine whether moderate ductility with ultra high strength or ultra high ductility with moderate strength is more effective on shear behavior of RC beam elements. Parameters considered during the experimental study were ratio of shear span (a) to effective depth (d), different concrete types used, and the tensile reinforcement ratios of reinforced beam elements. The influence of these parameters on shear capacity of beams, stiffness, ductility, energy absorption capacity, and failure modes was evaluated. It is believed that these results will play an important role in interpreting the shear behavior of large-scale RC beams produced with ECC and RMC.

Research significance

As is common with many emerging materials, research related to material and durability properties precedes research on structural behavior. Because they are relatively new to the research community, studies related to structural response of HPFRCCs such as ECC and RMCs are lacking, especially those investigating response under shearing effect. Within this context, this study aims to investigate the structural behavior of large-scale ECC- and RMC-reinforced beams under shear by giving priority to the individual effects of ultra high ductility with moderate strength and ultra high strength with moderate ductility.

Experimental program

Test specimens and material properties

Six reinforced ECC beams and another corresponding six RMC beams were produced and tested. To evaluate reinforced beam elements under shearing effect, ratios of shear span (a) to effective depth (d) were selected to be a/d = 1, 2, and 3. A low ratio of a/d between 1 and 3 was selected to monitor the behavior of test specimens under the influence of shear forces and to promote the possibility of shear failure occurrence. Changes in failure mode of reinforced concrete beams with variations in a/d ratio are presented in Figure 1.¹⁵ The figure shows that as the a/d ratio decreases, the possibility of shear failure increases, particularly when the a/d is between 1.0 and 3.0. For each a/d ratio, test specimens were reinforced with two different amounts of main longitudinal tensile reinforcements, whereas, for specimens with “A” in their notation, longitudinal tensile reinforcement of 2φ16 was used. For “B” specimens, longitudinal tensile reinforcement of 4φ16 were used. To prevent local failure at the support points of reinforced beam specimens, a small number of shear reinforcements were used in these areas. No additional shear reinforcements were added along the beam length. The geometrical dimensions and steel reinforcement details of test specimens are presented in Figure 2. For each ECC test specimen with certain reinforcement details and a/d ratio, there was an identical corresponding RMC specimen. Reinforcement used throughout the study was obtained from the same supplier and its properties were tested under uniaxial tension with five trials for each type. Ribbed steel bars with 16 mm and 10 mm diameters were used as longitudinal and shear reinforcements over end supports, respectively. The yield strength (f_sy), ultimate strength (f_su), and elastic modulus (E) of the 16 mm reinforcement were 520 MPa, 625 MPa, and 205 GPa, respectively. The corresponding values for 10 mm reinforcement were 428 MPa, 535 MPa, and 198 GPa, respectively. The values of the main parameters of test specimens are presented in Table 1.

Table 1.

Values of main parameters of test specimens.

Spec. no.	Beam name	Concrete type	Beam length (mm)	Longitudional tension reinforcement ratio	a/d Ratio
1	Beam 1A	ECC	1130	0.0129	1
2	Beam 1B		1080	0.0257
3	Beam 2A		1560	0.0129	2
4	Beam 2B		1460	0.0257
5	Beam 3A		1990	0.0129	3
6	Beam 3B		1840	0.0257
7	Beam 1A	RMC	1130	0.0129	1
8	Beam 1B		1080	0.0257
9	Beam 2A		1560	0.0129	2
10	Beam 2B		1460	0.0257
11	Beam 3A		1990	0.0129	3
12	Beam 3B		1840	0.0257

ECC: engineered cementitious composites; RMC: Ryerson mix concrete.

ECC mixtures used for all six test specimens were produced to have similar compressive strengths. The ECC mixture had a water-to-cementitious material ratio (W/CM) of 0.27 and a fly ash to Portland cement ratio (FA/PC) of 2.2, by mass. Components similar to those of typical fiber-reinforced concrete were used for ECC production, including CEM I 42.5 type cement (similar to ASTM Type I), sand, Class-F FA with lime content of 5.57%, water, fibers, and a high-range water-reducing admixture (HRWRA). Proportion details for the mixture of ECC are presented in Table 2. To minimize matrix fracture toughness, which is necessary to obtain multiple microcracking and consequently strain-hardening behavior, smaller size aggregates are preferred. Therefore, silica sand with a maximum aggregate size of 1 mm was incorporated into the ECC mixtures. Fibers used in ECC were polyvinyl alcohol (PVA) type with an average diameter of 39 µm, an average length of 8 mm, and a tensile strength of 1610 MPa. HRWRA was added to the mixture until the desired fresh ECC characteristics were visually observed described in Yang et al.¹⁶ The mixture proportion for RMC is also presented in Table 2; production incorporated CEM I 42.5 type cement, silica sand with maximum aggregate size of 0.4 mm, silica fume, straight steel fibers with the length of 13 mm and diameter of 0.2 mm, HRWRA, and water. During the production of RMC concrete, solid ingredients were initially mixed for a minimum of two minutes in a dry state until a homogenous distribution was achieved, and then water was slowly added with half of the superplasticizer and mixed for 4 min. After that, the second half of the superplasticizer was slowly added to the mixture. When the mixture became self-consolidating, usually 3 to 5 min after the addition of the second half of the superplasticizer, the fibers were dusted over the mixture to allow even dispersion and avoid clumping. Finally, the concrete was mixed for one additional minute after the inclusion of fibers.

Table 2.

Mixture proportions and basic mechanical properties.

Ingredients (kg/m³)	ECC	RMC
Portland cement	375	818
Fly ash	823	–
Slag	–	–
Water	318	221.8
Coarse aggregate	–	–
Fine aggregate	–	–
Silica sand	446	899
Silica fume	–	204
Steel fiber	–	164
PVA fiber	26.0	–
HRWRA	2.3	26
WRA	–	–
Mechanical properties (28-day)
Compressive strength (MPa)	46.1	150.5
Flexural strength (MPa)	7.41	11.6
Flexural deformation (mm)	4.52	1.53

ECC: engineered cementitious composites; RMC: Ryerson mix concrete; PVA: polyvinyl alcohol; HRWRA: high-range water-reducing admixture.

Six Ø100 × 200 mm cylinder specimens from each mixture were prepared for the determination of compressive strength. Flexural parameters (flexural strength and deflection) were evaluated with four-point bending tests, using six 400 × 100 × 75 mm³ beam specimens from each mixture. Four-point bending tests were performed at a loading rate of 0.005 mm/s using a universal testing machine. Typical bending test results are displayed in Figure 3 as flexural stress–deflection diagrams for ECC and RMC mixtures at the age of 28 days. As seen from Table 2, compressive strength test results at 28 days for ECC and RMC mixtures were 46.1 and 150.5 MPa, respectively. Even though a considerable difference was observed in the 28-day-old compressive strength results of the ECC and RMC mixtures due to different ingredients used for proportioning, flexural strength values after 28 days were closer to each other with averages of 7.41 and 11.6 MPa for ECC and RMC beams, respectively. Ultimate deflection capacities from the four-point bending test, which reflects the material ductility of ECC and RMC mixtures, were 4.52 mm and 1.53 mm after 28 days, respectively. As seen in Figure 3, both beams from different concrete types underwent plastic deformation, although this was much more significant for ECC.

Test setup and instrumentations

The four-point bending test setup was adopted to evaluate the behavior of reinforced beams under shear forces. Shear force was gradually applied to the specimens until failure, and measurements of mid-span beam deflections, widths of cracks in both shear spans and strains in tensile reinforcements were recorded. The selected test setup and locations of LVDTs are shown in Figure 4.

Load–deflection curves of the beam specimens were plotted by considering the vertical deflection values recorded at the mid-span of the maximum moment region and three points on both symmetrical sides of the beams. Strength, stiffness, ductility ratios, and energy absorption capacities of the specimens were obtained from the graphs as well explained later. Strain measurements were recorded at the maximum moment regions for tensile reinforcement of each beam specimen. Strain gages placed at the middle of each steel bar before casting are shown in Figure 5.

Shear deflections due to shear cracks were calculated using three LVDTs symmetrically placed on both left and right shear spans of the beams, to calculate the percentage of shear deflection in total mid-span deflection as shown in Figure 4. Since the distances of vertical shear deflections to mid-span were equal for the left and right spans, the results from both sides were averaged for each beam specimen. The approach used for this calculation is shown in Figure 6, along with the strain geometry. Equations (1) to (4) were used to calculate vertical shear deflection, which was the component of LVDT measurements recorded from left and right shear spans. Definitions of notations used in equations (1) to (4) are defined in Figure 6.

α = acr \tan (\frac{y a}{x a}) = acr \tan (\frac{(\frac{l 1' - l 2'}{2}) \sin (θ)}{(\frac{l 1' - l 2'}{2}) \cos (θ)}) = acrtan (\frac{l 1' - l 2'}{l 1' + l 2'} \tan (θ)) = acr \tan (\frac{l 1' - l 2'}{l 1' + l 2'} (h / h w)) = acr \tan (\frac{ɛ 1 - ɛ 2}{2 + ɛ 1 + ɛ 2} (h / w))

(1)

β = acr \tan (\frac{y c}{x c}) = acr \tan (\frac{(\frac{l_{1}^{'} - l 2'}{2}) \cos (θ)}{(\frac{l_{1}^{'} - l 2'}{2}) \sin (θ)}) = acr tan (\frac{l_{1}^{'} - l 2'}{l_{1}^{'} + l 2'} \cot (θ)) = acr \tan (\frac{l_{1}^{'} - l 2'}{l_{1}^{'} + l 2'} (w / w h h)) = acr \tan (\frac{ɛ 1 - ɛ 2}{2 + ɛ 1 + ɛ 2} (w / h))

(2)

γ xy = α + β

(3)

δ sh = γ xy \cdot h

(4)

Experimental results and discussion

General behavior and failure modes

Load–deflection curves obtained after the application of four-point bending loading are shown in Figure 7, grouped in accordance with concrete type: Figure 7(a) for ECC beams and Figure 7(b) for RMC specimens. From the load–deflection curves, general behavior, failure modes, strength, stiffness, ductility, and energy absorption capacity of test specimens were obtained and tabulated in Table 3. Crack distributions and failure modes of specimens after testing are also shown in Figure 8.

Table 3.

Overall results of the beam specimens after four-point bending tests.

Spec no.	Concrete type/ Beam name	Load capacity (kN)		Deflection (mm)		Ductility ratio	Yield stiffness (kN/mm)	Energy absorption capacity (kN-mm)	Maximum strain at tension reinf. (µɛ)	Failure mode
Spec no.	Concrete type/ Beam name	Yield	Ultimate	At yield	At failure	Ductility ratio	Yield stiffness (kN/mm)	Energy absorption capacity (kN-mm)	Maximum strain at tension reinf. (µɛ)	Failure mode
1	ECC/Beam 1A	403.60	501.47	3.71	16.01	4.32	108.79	6113	10,021	Flexure
2	ECC/Beam 1B	512.53	550.91	5.04	20.25	4.02	101.69	7758	4493	Comp. strut crushing
3	ECC/Beam 2A	211.45	220.66	12.58	27.36	2.17	16.81	4269	2536	Flexure shear
4	ECC/Beam 2B	290.83	290.83	10.36	13.18	1.27	28.07	5183	3460	Flexure
5	ECC/Beam 3A	120.96	143.27	11.79	32.75	2.78	10.26	3093	3983	Flexure
6	ECC/Beam 3B	180.70	190.59	13.54	20.49	1.51	13.35	4806	3154	Flexure
7	RMC/Beam 1A	437.82	548.19	3.38	11.08	3.28	129.53	3767	9472	Flexure
8	RMC/Beam 1B	752.23	875.59	4.14	13.26	3.20	181.70	7698	8993	Flexure
9	RMC/Beam 2A	220.97	242.96	6.54	13.10	2.00	33.79	1822	8484	Flexure
10	RMC/Beam 2B	402.61	443.25	6.04	12.95	2.14	66.66	3109	8358	Flexure
11	RMC/Beam 3A	171.00	189.43	12.50	21.60	1.73	13.68	1611	9416	Flexure
12	RMC/Beam 3B	259.02	282.10	14.51	22.17	1.53	17.85	2349	7377	Flexure

ECC: engineered cementitious composites; RMC: Ryerson mix concrete.

As the applied load increased, bending cracks occurred in ECC/Beam 1A and 1B specimens with a/d ratios of 1.0. Upon the increase in loading, the widths of bending cracks increased and tensile reinforcements reached their yielding points at 403.6 kN and 512.53 kN loading levels and corresponding 3.71 mm and 5.04 mm deflection levels for ECC/Beams 1A and 1B, respectively. For the same specimens, failures occurred after highly ductile behavior at maximum load-carrying capacities of 501.47 kN and 550.91 kN and mid-span deflections of 16.01 mm and 20.25 mm (Figure 8(a) for ECC/Beam 1A and Figure 8(b) for ECC/Beam 1B). The ECC/Beam 1A specimen failed in a typical flexural mode due to crushing at the top portion of the maximum moment region after yielding of the main longitudinal bottom steel reinforcement, whereas concrete crushing took place between the end support and point load after tensile reinforcement yielding in case of ECC/Beam 1B.

In ECC/Beam 2A and 2B specimens, load increase initially caused bending cracks at the maximum moment region, followed by shear cracks at the right shear span of the ECC/Beam 2A specimen. 2A and 2B ECC specimens showed yielding at 211.45 kN and 290.83 kN loading and corresponding 12.58 mm and 10.36 mm deflection levels, respectively. 2A and 2B ECC failed after reaching the maximum load-carrying capacities of 220.66 kN and 290.83 kN, respectively. After the tensile reinforcements of ECC/Beam 2A yielded, flexural-shear failure took place due to excessive widening of shear cracks in the right shear span, and the joining of these cracks between the right loading point and support. On the other hand, ECC/Beam 2B displayed flexural failure owing to crushing of concrete and yielding of steel reinforcement at the maximum moment region.

Both ECC/Beam 3A and 3B specimens with an a/d ratio of 3 showed typical flexural failure due to crushing of concrete at the maximum moment region after the tensile reinforcements yielded. Although none of the ECC beams included shear reinforcements, all tensile reinforcements in these specimens yielded and all except ECC/Beam 2A showed flexural failure. Attaining ductile flexural failure for RC beams with no shear reinforcements, especially in case of low a/d ratios with high potential of triggering shear failure, is quite an achievement and was mainly attributed to the superior tensile and flexural strength of ECCs.

With the effect of increasing shear forces, all RMC beams showed failure under bending after the occurrence and further development of flexural cracks in the maximum moment region. Changes in shear spans and longitudinal reinforcement ratios did not influence the failure mechanisms, and all RMC beams showed flexural failure after the yielding of tensile reinforcement along with the failure of fibers in this region. In beam specimens with an a/d ratio of 2, a limited number of small shear cracks occurred in the shear spans. Most were flexural cracks and occurred in the maximum moment region. Moreover, steel fibers used for RMC beams were monitored to restrict increments in number and widths of cracks visible in shear spans.

Tensile reinforcements of RMC/Beam 1A and RMC/Beam 1B specimens with shear span to effective depth ratios of 1 reached their yielding points after reaching load-carrying capacities of 437.82 kN and 752.23 kN at deflection values of 3.38 and 4.14 mm, respectively. With increased loading on RMC/Beam 1A and RMC/Beam 1B specimens, the number of flexural cracks occurring in the maximum moment region increased, and after reaching load-carrying capacities of 548.19 kN and 875.59 kN, beams failed at their compression zones with 11.08 mm and 13.26 mm final deflection capacities, respectively (Figure 8(g) and (h)). These findings show that increasing the tensile reinforcement ratio by two has a paramount influence on the escalation of load-carrying capacity.

Upon loading, RMC/Beam 2A and RMC/Beam 2B specimens with an a/d ratio of 2 exhibited some flexural cracks in the maximum moment region, and at load-carrying capacities of 220.97 kN and 402.61 kN, tensile reinforcements of the beams yielded after reaching deflection levels of 6.54 mm and 6.04 mm, respectively. When loading was continued, shear cracks started to be visible near the shear spans, although the number and widths of cracks were restricted. Before failure due to crushing of concrete in the compression region upon further continued flexural loading, RMC/Beam 2A and RMC/Beam 2B specimens exhibited deflection capacities of 13.10 mm and 12.95 mm after reaching maximum load-carrying capacities of 242.96 kN and 443.25 kN, respectively (Figure 8(i) and (j)).

RMC/Beam 3A and RMC/Beam 3B having shear span to effective depth ratios of 3 showed yielding of their tensile reinforcements after the occurrence of flexural cracks in the maximum moment regions. Steel reinforcement yielding took place after applied loads of 171.00 kN and 259.02 kN were reached by RMC/Beam 3A and RMC/Beam 3B at 12.50 mm and 14.51 mm deflection levels, respectively. A very limited number of cracks with small openings were observed in the shear spans and upon increased loading, and cracks in the maximum moment region increased in both number and width. At maximum load-carrying capacities of 189.43 kN and 282.10 kN, beam specimens showed failure with 21.60 mm and 22.17 mm deflection capacities after crushing of beams from the compression regions (Figure 8(k) and (l)).

The yield capacity of tensile reinforcements of RMC specimens was reached at higher levels compared to ECC specimens. In beams with a/d ratios of 1 and 2, yield capacities of RMC specimens were reached at lower deflection values than ECC specimens. RMC beams exhibited higher load-carrying capacities than ECC beams, while the opposite was true for deflection results at failure. In addition, increased tensile reinforcement ratios caused much larger enhancements in yield and maximum load-carrying capacities of RMC specimens in comparison to ECCs.

Although ECC and RMC beam specimens were designed to have shear span to effective depth ratios between 1 and 3, all of the specimens failed in flexure. The reason for both RMC and ECC beams to fail in flexure despite the high shear span to depth ratios is attributed to the increased ductility of specimens. Although ECC specimens are relatively more ductile than the RMC specimens, it should be kept in mind that the ductility of RMCs is still several hundred times that of conventional concrete material (Figure 3).

Load capacity and stiffness

The yield and ultimate load-carrying capacity values of the beam specimens are determined from the load deflection curves and presented in Table 3. Yield and ultimate strength values of beam elements were evaluated in terms of concrete type, shear span to effective depth ratio, and tensile reinforcement ratio. When the ratio of tensile reinforcements was increased in ECC beams, there was an average 38% increase in yield strength and 25% in ultimate strength. In the case of RMC beams, however, increments were more pronounced; the increase in tensile reinforcement ratio resulted in 68% and 64% increases in yield and ultimate strength results, respectively. As the shear span to effective depth ratios of ECC and RMC beams were increased, both yield and ultimate strength results of the specimens were reduced. Yield strength results dropped by 91% and 98% for ECC and RMC specimens when the a/d ratio was increased from 1 to 2. When the a/d ratio was increased from 2 to 3, however, reductions were less pronounced, with the same values decreasing by 75% and 29% for ECC and RMC beams, respectively. When tensile reinforcements were doubled, reductions in yield strength results were 76% and 61% for ECC beams and 87% and 55% for RMC beams. With the increased a/d ratios, the ultimate strength results of beam specimens were reduced. Reductions in the ultimate strength results were less pronounced compared to yield strength results, however. Reductions in yield and ultimate strength results were much more visible when a/d ratio was increased from 1 to 2.

The concrete type has also affected the strength results. In the A series specimens, average yield strength values of ECC beams were 18% less than in RMC beams. Average ultimate strength results of ECC beams were 17% less than RMC beams, and the difference between the two concrete types became more obvious when the amount of tensile reinforcement was doubled. Focusing on the results of B series specimens with higher reinforcement ratios, the average yield and ultimate strength results of ECC beams were 43% and 53% less than those of RMC beams, respectively.

The stiffnesses of the tested beams were calculated at the point where specimens reached their yielding points; the results are presented in Table 3. When the overall yield stiffnesses were taken into consideration, RMC beams results were an average 59% higher than in ECC beams. Although this finding was expected due to the significantly higher compressive strength of RMC, the amount was rather high. The fact that RMC beams exhibited a great deal of stiffness until the tensile reinforcements reached their yielding capacity is considered a success. Average stiffness values of RMC beam specimens with a/d ratios of 2 were 124% higher than those of ECC beams. It is notable that RMC beams showed relatively high yield stiffness values at this level of a/d ratio, where shear failure risk is higher. With the increase in tensile reinforcement ratio, yield stiffnesses increased as well. Yield stiffnesses of B series ECC beams were on average 30% higher than these of A series, the corresponding values were 56% for RMC beams. Increased reinforcement ratios were more effective on the yield stiffnesses of RMC beam specimens compared to ECCs.

Deflection ductility ratio

The calculated ductility ratios of the beam specimens are shown in Table 3. The ductility ratios were calculated by dividing the deflection at failure to the deflection at the yield point. The level where specimens dropped down to 85% of their ultimate load-carrying capacities was used as the failure point. Higher ductility ratios were acquired from ECC beams than from RMCs, excluding the ECC/Beam 2B specimen. RMC/Beam 3B and ECC/Beam 3B specimens had very similar ductility ratios. For both concrete types, increased a/d ratios caused deflection ductility ratios to show a decreasing trend, with the exception of the ECC/Beam 2B specimen. For an a/d ratio of 1, ECC beams had an average 29% higher ductility than RMC beams. When the a/d ratio was increased to 2, however, RMC beams had 17% higher average ductility values than ECC beams. For the a/d ratio of 3, the average deflection ductility ratio of ECC beams was 32% higher. With the exception of the RMC/Beam 2B specimen, deflection ductility ratios decreased as tensile reinforcement ratios were increased. Increments in tensile reinforcement ratios affected ECC beam elements more significantly so that the reduction in deflection ductility ratios was 54% on average, while the same value was only 3% for RMCs. ECC specimens in the A series displayed an average 34% higher ductility ratio compared to RMC specimens. With the exception of test specimens having an a/d ratio of 2, ECC beams in the B series showed an average ductility ratio 12% higher than RMC beams. For a shear span to effective depth ratio of 2, deflection ductility ratios of ECC and RMC beams were similar.

Energy absorption capacity

The energy absorption capacity of beam specimens was found by calculating the area under load–deflection curves. In the energy absorption calculations, only the area up to the failure point was considered. Failure points were kept similar to those used for ductility ratio calculations, as explained above. Energy absorption capacities for all beams tested are presented in Table 3. When the results were evaluated, it was noted that all ECC beams exhibited energy absorption capacities higher than those of RMCs. The average energy absorption capacity obtained from ECC beams was 71% higher than that obtained from RMC beams. For both concrete types, increasing the a/d ratio caused energy absorption capacity results to decrease dramatically. For a/d ratios of 1, 2 and 3, energy absorption capacities of ECC beams were 21%, 92%, and 98% higher than in RMC beams, respectively. Increased a/d ratios therefore also increased the differences in energy absorption capacities. When the amount of tensile reinforcement was increased, energy absorption capacities increased for both concrete types. ECC beams with higher amounts of tensile reinforcement showed 35% higher average energy absorption capacities, while this value went up to 74% in RMC beams. It, therefore, appears that the energy absorption capacities of RMC beams were more affected by the changes in tensile reinforcement ratio. Different concrete types were also influential on energy absorption capacities. A series ECC beams resulted in 96% higher energy absorption capacity than RMCs, whereas the same value was 57% for B series ECC and RMC beams.

Measured tensile reinforcement strains

Strain measurements were recorded at maximum moment regions for longitudinal tensile reinforcement of each beam specimen. Maximum strain recorded for each beam specimen is shown in Table 3, and selected examples of load–strain graphs are presented in Figure 9. Strain measurements were taken to check whether the tensile reinforcements reached their yielding capacities and to confirm the mode of failure of tested specimens. Results show that strain measurements complied with general load–deflection behaviors, ductility ratios, and failure modes of the specimens. The yield strain limit was determined to be 2549 mɛ for tensile reinforcements. All specimens showed strain values higher than the yield strain limit of tensile reinforcements, except in ECC/Beam 2A, which showed combined shear-bending failure. Moreover, recorded strain measurements complied with the failure mechanisms. Maximum average strain measurements were 4608 and 8683 mɛ for ECC and RMC beam elements, respectively, which indicates that average strain measurements for RMC beams were 88% higher than ECC beams. As expected, increased reinforcement ratios used in the beam specimens led to a decrease in strain values.

Shear deflection

Vertical shear deflections originating due to cracks occurred in shear spans, and their percentage to mid-span beam deflection results were calculated, providing important information about the shear behavior of beam specimens. Calculated shear deflections and their percentage of total mid-span beam deflection are shown in Table 4. The approach presented in Figure 6 was used to calculate the vertical shear deflections of the beams tested, and the values obtained from left and right shear spans were averaged to find the shear deflections. RMC beams showed lower shear deflection results than ECC beams. While average shear deflection percentage in RMC beams was 2.0%, this value was 9.5% for ECC beams. Increasing the tensile reinforcement ratio led the shear deflection percentages to increase as well; ECC and RMC specimens showed an average of 430% and 80% increase in shear deflection values. Under normal conditions, shear deflection percentages are expected to increase with a/d ratio. However, increased a/d ratios caused marked restrictions and reductions in the shear deflection results of ECC beams. When a/d ratios were increased from 1 to 2 to 3, shear deflection results decreased from 12% to 11% and 5%, respectively, which shows that ECC is effective in restricting shear cracks. In the RMC beams, increased a/d ratios did not make a noticeable difference in the shear deflection results, with values ranging between 1.7% and 2.7%, suggesting that RMC concrete successfully reduces shear deflection.

Table 4.

Shear deflections of specimens.

Spec. no.	Concrete type/ Beam name	Shear deflection (mm)	Mid-span deflection (mm)	Percentage^a
1	ECC/Beam 1A	0.44	16.01	2.75
2	ECC/Beam 1B	4.32	20.25	21.33
3	ECC/Beam 2A	1.25	27.36	4.57
4	ECC/Beam 2B	2.36	13.18	17.91
5	ECC/Beam 3A	0.66	32.75	2.02
6	ECC/Beam 3B	1.69	20.49	8.25
7	RMC/Beam 1A	0.19	11.08	1.71
8	RMC/Beam 1B	0.22	13.26	1.66
9	RMC/Beam 2A	0.11	13.10	0.84
10	RMC/Beam 2B	0.33	12.95	2.55
11	RMC/Beam 3A	0.51	21.60	2.36
12	RMC/Beam 3B	0.66	22.17	2.98

Ratio of shear deflection to total mid-point deflection in percentage.

ECC: engineered cementitious composites; RMC: Ryerson mix concrete.

Conclusions

In this study, the shear behavior of RC beams produced with ECC and RMC was investigated under four-point bending loading, with tensile reinforcement ratio and shear span to effective depth ratio as the main parameters. The relationships among the abovementioned parameters and load-carrying capacity, ductility ratio, energy absorption capacity, stiffness, failure modes, and general load–deflection behaviors of beam specimens were further examined, and the following conclusions were drawn from the results of physical testing:

Both ECC and RMC use in beam specimens successfully prevented the formation of shear failure and resulted in flexural failure. Tensile reinforcements of all reinforced beam specimens with no shear reinforcements, with a/d ratios ranging from 1 to 3 (risky in terms of shear failure), reached their yielding capacity. Beams failed under flexure after displaying different ductility and energy absorption capacity results.

RMC beams showed higher load-carrying capacity compared to ECC beams. Both yield and ultimate strength results of RMC beams were an average of 30% and 36% higher than those of ECC beams, respectively. The fact that shear effect beams produced with two different concrete types with no shear reinforcements reached their yielding capacity and exhibited flexural failure can be regarded as a significant achievement. Increasing the amount of shear span to effective depth ratio decreased yield and ultimate strength results of beams using both concrete types. Yield and ultimate strength results showed increasing behavior with the increase in tensile reinforcement ratios, although this effect was more evident in RMC beams.

Yield stiffnesses of RMC beams were substantially higher (average of 59%) than these of ECC beams. When a/d ratios were increased, significant reductions in yield stiffnesses were monitored in beams incorporating different concrete types. However, increased tensile reinforcement ratios resulted in escalated yield stiffness values for beams of both concrete types.

ECC beams exhibited 16% higher deflection ductility ratios than RMC beams, on average. The deflection ductility ratios of ECC beams having a/d ratios of 1 and 3 were higher than RMC beams by 29% and 32%, respectively. With the increments in the amount of tensile reinforcement, deflection ductility ratios of all beam specimens decreased.

ECC beams proved themselves rather ductile by exhibiting 53% higher energy absorption capacity than RMC beams. As the a/d ratio was reduced, energy absorption capacities for both concrete types decreased as well. Increased tensile reinforcement ratios led to an increase in energy absorption capacities.

Although no shear reinforcement was used in all beam specimens produced with different concrete types, shear deflection ratios measured for both concrete types were rather low. Average shear deflection ratios were calculated at 9.5% and 2.0% for ECC and RMC beams, respectively. These values were found to be very low for beams with no shear reinforcement at a/d ratios, which are at risk of shear failure. Overall, shear cracking was restricted and shear deflection ratios were significantly reduced for both concrete types.

RMC beams have higher load-carrying capacity and yield stiffness than ECC beams, while ECC beams have a higher deflection ductility ratio and energy absorption capacity than RMC beams.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors gratefully acknowledge the financial assistance of the Turkish Academy of Sciences, Young Scientist Award program (GEBIP) and Feyzi AKKAYA Scientific Activates Supporting Fund (FABED) Young Investigator Research Award.

Appendix

References

Bentz

Vecchio

Collins

. Simplified modified compression field theory for calculating shear strength of reinforced concrete elements. ACI Struct J 2006; 103: 614–624.

Collins

Bentz

Sherwood

. An adequate theory for the shear strength of reinforced concrete structures. Magaz Concr Res 2008; 60: 635–650.

Nilson

. Design implications of current research on high-strength concrete. ACI Special Publ 1985; 87: 85–109.

Swamy

. High-strength concrete-material properties and structural behavior. ACI Special Publ 1985; 87: 110–146.

Naaman AE and Reinhardt HW. Setting the stage: Toward performance-based classification of FRC composites. In: High performance fiber reinforced cement composites (HPFRCC-4), Published by RILEM S.A.R.L., 2003, pp. 1–4.

Li VC. Engineered cementitious composites tailored composites through micromechanical modeling. In: Fiber reinforced concrete: Present and the future. Westmount, QC: Canadian Society for Civil Engineering, 1998, pp. 64–97.

. On engineered cementitious composites (ECC): A review of the material and its applications. Adv Concr Technol 2003; 1: 215–230.

Lin

. Crack bridging in fiber reinforced cementitious composites with slip-hardening interfaces. Mech Phys Solids 1997; 45: 763–787.

Lin

Kanda

. On interface property characterization and performance of fiber reinforced cementitious composites. Concr Sci Eng 1999; 1: 173–184.

10.

Ametrano D. Bond characteristics of glass fibre reinforced polymer bars embedded in high performance and ultra-high performance concrete. MS Thesis, Ryerson University, Toronto, ON, Canada, 2011, pp. 1–132.

11.

Hou

L-J

Zhang

X-F

. Shear behavior of reinforced ultrahigh toughness cementitious composite beams without transverse reinforcement. J Mater Civil Eng 2012; 24: 1283–1294.

12.

. On the shear behavior of engineered cementitious composites. Adv Cem Mater 1994; 1: 142–149.

13.

Wang

. Flexural behaviors of glass fiber-reinforced polymer (GFRP) reinforced engineered cementitious composite beams. ACI Mater J 2002; 99: 11–21.

14.

Van Zijl

GPAG

. Improved mechanical performance: Shear behavior of strain-hardening cement-based composites (SHCC). Cem Concr Res 2007; 37: 1241–1247.

15.

MacGregor

Wight

Teng

. Reinforced concrete: Mechanics and design, Upper Saddle River, NJ: Prentice Hall, 1997.

16.

Yang

Sahmaran

Yang

. Rheological control in the production of engineered cementitious composites. ACI Mater J 2009; 106: 357–366.