In the present study, an experimental program will be prepared to provide a much needed understanding of the behavior of MRPC one-way slabs under line loading. The experimental program was conducted a series of reinforced one-way slabs of normal and modified reactive powder concrete which were casted to illustrate the effect of concrete type, steel fiber content, slab thickness and flexural reinforcement ratio. Initially, the mechanical description of constitute materials was given. As well as the mechanical properties of normal and modified reactive powder concrete were investigated by using control specimens were casted which were three cylinders for compressive strength, three cylinders for splitting tensile strength and three prisms for modulus of rupture.

Four variables were investigated in this study to show their effects on the flexural strength of the NC and MRPC slabs. These variables are:

1. Percentage of steel fibers volumetric ratio (0, 0.4, and 0.8) %.
2. Flexural steel reinforcement ratio (ρ₁=0.0033 and ρ₂=0.0066).
3. Thickness of slab (T=50 and T=70) mm.
4. Type of concrete [Normal Concrete (NC) and Modified Reactive Powder Concrete (MRPC)].

Table 1 illustrates the details of all the test slabs.

Slabs designations were as following:

NCS is Normal Concrete Slab,

MRPCS is Modified Reactive Powder Concrete Slab,

the symbol (1 and 2) from (ρ₁=0.0033 or ρ₂=0.0066),

the symbol (0, 0.4 and 0.8) from steel fibers content (V_f=0, 0.4 or 0.8%),

and the symbol (50 and 70) from slab thickness (T=50 or T=70) mm.

Ordinary Portland cement (Type Ι) produced in Iraq (TASLUJA-BAZIAN) was used in this study. The chemical and physical properties indicated that the adopted cement conforms to the specification.

Al-Ekhaider natural sand of 4.75mm maximum size was used as fine aggregate in normal concrete. However, for RPC, very fine sand with a maximum size of 600µm was used. Crushed gravel from the AL-Nibaey region was used for MRPC and reference concrete specimens with a maximum size of (8 and 10mm), respectively. Results indicate the grading of these materials (aggregates) within the requirements of the specification.

Fig. (1). A-Schematic sketch of typical experimental set-up. B-Specimen and loading instrumentation.

To improve the workability and strength of concrete, high range water reducing admixture used in this study is known commercially as Viscocrete-PC20. This type of superplasticizer is imported from Sika Company. In the present investigation, silica fume was used, which imported from Sika Company. Tap water was used in all mixes and the curing of the specimens.

Micro straight steel fibers with a diameter of 0.25 mm, length of 13.1mm, aspect ratio (L/d) of 52 and yield strength of 1130 MPa were used in MRPC mixes.

Deformed steel bar of nominal diameter (6) mm was used as slab reinforcement. Two reinforcement ratios (ρ) were used in each group of the tested slabs. The yield strength of the used bar was 435 MPa. The result of testing this bar meets the ASTM A615 requirements for Grade 60 steel.

Three types of modified reactive powder concrete MRPC mixes were used in the present research, as listed in Table 2. The variables adopted in these mixes were the volume ratio of steel fibers (three-volume ratios were considered 0, 0.4 and 0.8) %. The mixed type of normal concrete NC was adopted as a reference mix, (Table 2).

In the present work, mixing was performed by using a 0.19 m³ capacity horizontal rotary mixer. Firstly, the silica fume powder was mixed in a dry state with the required quantity of sand for 5 minutes. Then, cement and crushed gravel were loaded into the mixer and mixed for another 5 minutes. The required amount of tap water was mixed with superplasticizers and then added to the rotary mixer within 5 minutes. Finally, when steel fibers were used, they were introduced, and dispersed uniformly and mixed for an additional 2 minutes. This procedure is similar to the method used by Wille et al [2, 8] which was used successfully to produce RPC with compressive strength exceeding 150MPa without using heat curing; this is a major factor in the process of production.

Sixteen slabs of dimensions (1000×450×50 or 70) mm were cast and tested in flexure in this study. Four of these slabs were made with NC and twelve with MRPC. The slabs were simply supported at their ends with a clear span of (900) mm. The central deflection of the slabs had been measured by means of (0.01mm/div.) sensitivity dial gauge of (30)mm capacity placed at their center of the tension face. Two-line loads were applied at the third points of the slab by means of a hydraulic jack (Fig. 1). The test procedure included crack monitoring and central deflection measurements for load increments of 5 kN.

Table 3 shows the test results of mechanical properties obtained for the four mixes. These properties are concrete compressive strength (fʹ_c), splitting tensile strength (f_t) and modulus of rupture (f_r). Each value presented in this table represents the average value of the three specimens.

The effect of steel fibers on concrete compressive strength seems to be small for MRPC and about (20) % as compared with control specimens without steel fibers. However, the splitting tensile strength and modulus of rupture are more affected by using steel fibers and about (50) % compared with control specimens without steel fibers.

Table 4 summarizes the results of the first cracking load (P_cr), ultimate load (P_u) and mode of failure for all tested slabs together with their deflection.

The load- central deflection relationships for the tested slabs are shown in Figs. (2-4). From this relation, it can be observed that at initial loading, the relationship is approximately linear elastic characteristics with no crack occurrence. However, when the cracking load (Pcr) was exceeded and the first crack developed at the bottom of the slab within the middle third where maximum bending occurred, the gradient of the initial load-central deflection curve was reduced and continued to reduce gradually until the steel yielded. At the post-yield stage of the slab's reinforcement, where strain hardening occurred, such that a slight increase in the subjected load was lead a large increase in the central deflection until the slab failure was occurred.

Fig. (2). Load-deflection curve of slabs with variable steel fiber.

Fig. (3). Load-deflection curve of slabs with variable steel reinforcement.

Fig. (4). Load-deflection curve of slabs with variable thickness.

Generally, at a low level of load, concrete resists the stresses at the tension face of the slab and it's free from any cracks. When the load level was increased, the extreme fiber concrete stress reaches its limiting concrete tensile stress and hair-fine flexure cracks occurred. The load at this stage is called the first cracking load. As the load was increased above this stage of loading, the cracks seem to be widened and start to propagate, then, generally, extend (to be initiated) and deviate towards the slab free sides up to failure.

The first crack was observed visually and the recorded loads at this stage for all the tested slabs were to be at (23 to 31) % of the slab failure load. In general, the first cracking load for all the tested slabs appeared under the loading point. The cracking load and ultimate load for all tested slabs are listed in Table 4.

For a simply supported one-way slab subjected to equal transverse loads at the third point, the middle third of the span is subjected to pure bending (such that it is under zero shear and maximum bending moment); whilst the remaining sections experience maximum shear force and varying bending moment. The middle third experienced the largest strains and therefore, concrete beneath undergoes cracking first. However, many of the slabs were failed by flexural tension mode and the others by combined modes of flexural tension and flexural shear. One of these slabs was failed by shear mode. Each of the slabs developed at least one shear crack. The shear cracks developed after several flexural cracks had developed. Typical crack configurations have been illustrated in Fig. (5).

Effect of concrete type (NC and MRPC) and the compressive strength of their concrete (fʹc) on cracking and ultimate loads and the increasing ratio between them for all tested slabs are detailed in Table 5. The improvement in the first cracking and ultimate loads due to increasing the (fʹc) value, for modified reactive powder concrete, reached to (190 and 212) %, respectively. However, increasing (fʹ_c) value, approximately, reduced the deflection for all load stages. The increase in (fʹ_c) value results in higher modulus of elasticity, which ultimately results in higher flexural rigidity (EI); therefore, the deflection is smaller.

Fig. (5). Crack pattern of tested slabs.

Effect of volumetric steel fiber ratio (V_f) on cracking and ultimate loads and the ratio of them for all tested slabs are listed in the Tables 6, 7 and 8 for V_f= 0.4% and 0.8%, respectively. Comparison with MRPC slabs without steel fibers, the improvement in the ultimate load value of slabs inclusion (V_f) 0.4% and 0.8% was reached up to (47 and 66) %, respectively. This means that this percentage of improvement of the ultimate load was increased as increased the content of steel fibers in the MRPC slabs.

Therefore, it can be observed from these tables that the improvement in cracking a load of slabs due to inclusion (V_f) percentages of 0.4% and 0.8%, comparison with MRPC slabs without steel fibers, was reached up to (24 and 50) %, respectively. It is clear that the improvement became higher as the steel fiber content increased.

The presence of steel fibers results in a delay in crack initiation and propagation where they hold concrete particles and prevent them from initial separation. Therefore, the first crack in fibrous concrete slabs appears at a load level appreciably higher than the load, which causes crack initiation in non-fibrous concrete slabs. After cracking, the steel fibers prevent the crack widening and delay its growth by the absorption of a portion of tension stresses carried by concrete i.e., this action reduces the tension stresses applied to the concrete. Therefore, the failure takes place in fibrous concrete slabs at a load level higher than that of non-fibrous concrete slabs. Moreover, this is observed from that the effect of including the steel fibers on the enhancement of the mechanical properties (compressive strength (fʹ_c), splitting tensile strength (f_t) and modulus of rupture (f_r)) of MRPC as present in Table 2.

However, it can be seen from the Table 4 that when increasing the steel fiber content in the MRPC slab up to 0.8%, slab of MRPCS2-0.8-70, the mode of failure was changed from shear to tension. This may be due to the effects of steel fiber, which delay the propagation of a crack in the shear zone, thereby, transform the failure to flexural mode.

Table 8 shows the effect of flexural steel reinforcement ratio on the behavior of normal and modified reactive powder concrete slabs in terms of load characteristics. Generally, for both types of concrete, the ultimate flexural capacity was enhanced by increasing the ratio of steel reinforcement.

For normal concrete, increasing flexural steel reinforcement ratio from (0.0033 to 0.0066) was improved with the ultimate failure load up to (45 and 37) % for slabs with (50 and 70) mm thicknesses, respectively. However, for MRPC slabs with thickness of 50mm and with steel fiber contents of 0,0.4 and 0.8, this percentage of increase was reached to (33,7 and 3) %, respectively. The enhancement of ultimate load capacity for MRPC slabs of the thickness (70) with steel fibers of 0, 0.4 and 0.8 was reached to 14,-20 and 53, respectively. Except for the specimen of MRPCS2-0.8-70, it is clear that when increasing the ratio of steel reinforcement, the enhancement in the ultimate load capacity was decreased, this may be due to that the failure mode was converted from tension failure to shear. The maximum increasing percentage of ultimate load was reached for the slab of thickness 70mm and had a higher content of steel fiber, MRPCS2-0.8-70, which delay propagation the cracks in shear zone, thereby, that permit to yield the flexural reinforcement and failed by tension mode, (Table 4).

From the results shown in Table 9, it can be seen that the ultimate flexural capacity was increased with increasing the slab thickness.

For normal concrete specimens, increasing slab thickness from (50) to (70) mm has increased the ultimate failure load by (49 and 33%) for slabs with flexural steel reinforcement ratio of (0.0033 and 0.0066), respectively. This is the percentage for slabs with 0.0033 flexural steel reinforcement (75, 78 and 33%) for MRPC slabs with steel fibers content (0, 0.4 and 0.8%), respectively. The improving percentage of the ultimate failure load for slabs with (0.0066) flexural steel reinforcement was (50, 38 and 98%) for MRPC slabs with steel fibers content (0, 0.4 and 0.8%), respectively. From Table 5, it can be noted that the thickness of slabs had no effect on the mode of failure, excepted slab of MRPCS2-0.8-70, where the failure mode was transformed from tension-shear to pure tension failure due to high content of steel fiber in this slab.

From the results in Table 9, for both normal and MRPC slabs, it is clear that the percentages of increasing the ultimate failure load for slabs with (0.0033) flexural steel reinforcement ratio were higher than of slabs with flexural steel reinforcement ratio (0.0066). This may be due to the slabs with (0.0033) flexural steel reinforcement ratio failed in tension mode, while the mode of failure in slabs with (0.0066) flexural steel reinforcement ratio was already shear mode. An excepted slab of MRPCS2-0.8-70 was failed by tension mode due to the high content of steel fiber in this slab.