Experimental Study of Self-Compacting Reinforced Concrete Hollow Beams Using Recycled Aggregate under Torsion

The characteristics of the materials utilised in the fabrication of the tested reinforced SCC beams with RCA are outlined as follows. The tests were conducted at the laboratories of the department is civil engineering at the University of Basrah.

2.1 Concrete

Four concrete mixes were utilised, varying based on the proportion of recycled concrete aggregate replacing natural material. Typically, chemical admixtures were used to achieve workable concrete with a low w/c ratio. Below is a concise overview of the components utilised in the production of concrete.

2.1.1 Cement

Type I Ordinary Portland Cement (Mabroka) is a widely utilised cement type that is suited for all general concrete construction purposes. Hence, it was used in this work. This type adheres to the specified condition outlined in IQS 5-1984 [8]. Tables 1 and 2 provide properties of the chemical and physical, respectively.

Table 1. Cement physical characteristics

Table 2. Cement chemical profile

Table 3. Characterization of fine aggregate

2.1.2 Water

Unadulterated tap water was utilized in the mixing procedure for all concrete compositions, according to the prescribed proportions specified in the design. In addition, the concrete curing procedure involved using pure water, which was replaced every two days for a period of 28 days. The curing process involved immersing the control specimens, including cubes, cylinders, and prisms, as well as the beam specimens, in a water tank filled with fresh water. This was done to evaluate the mechanical properties of the concrete. The specimens were continuously submerged in water for the entire duration of the test.

2.1.3 Fine aggregate (sand)

The manufacturing of SCC mixes involved the utilisation of natural sand sourced from Zubair-Basrah, Iraq. The fine aggregate had a maximum size of 4.75 mm and had a smooth texture with a spherical particle form. The sand was rinsed twice and subsequently air-dried, for the purpose of removing unwanted materials. Table 3 displays the classification of sand. The results indicate that the sulphate content of the sand was within the limitations specified in Iraqi Specifications No. 45/1984 [8], where the value was equal to 0.33.

2.1.4 Coarse aggregate

Two types of coarse aggregate were used:

(1) Gravel

Sanam Mountain location in the city of Basra offers natural crushed gravel (NCA) with a maximum size of 19 mm, the crushed coarse aggregate was washed, and then kept in an internal saturated and dry surface condition before being utilized. Where the results indicate that the coarse aggregate grading is within the requirements of Iraqi Specification No. 45/1984 [9], Table 4 shows the properties of the coarse aggregate used.

(2) Recycled concrete aggregate

RCA is prepared by crushing old concrete cubes under a compressive force of 25-32 MPa manually. The crushed concrete is divided into three parts with sizes 14 to 20, 10 to 14, and 5 to 10 mm using sieves as depicted in Figure 1. The grades and physical characteristics of RCA are shown in Tables 4 and 5, respectively.

1.png

Figure 1. RCA preparation

Table 4. Grading of coarse aggregate

Table 5. Physical properties and chemical composition of coarse aggregate

2.1.5 Superplasticizer (HRWR)

To produce SSC with and without recycled concrete aggregate a high-range water reducer must be used to satisfy the SCC requirement even with a low (w/c) ratio. The superplasticizer used in this work is a new generation modified polycarboxylate, which can be commercially named. GLENIUM51. This material is compatible with all types of Portland cement, free of chlorides, and compatible with ASTM C494 [10]. Ultra-ductile concrete shows better workability without separation. It also provides sufficient time for mixing, pouring, and finishing the final concrete surface.

2.1.6 Limestone powder (LSP)

The self-compacted concrete was achieved by using a higher proportion of LSP in the mixture, which includes both cement and filler materials. This material is commonly referred to as AL-Gubra and is obtained from the local market. It is a white grinding material derived from limestone quarried from several places in Iraq, commonly employed in construction procedures. Tables 6 and 7 display the physical attributes and chemical makeup of LSP.

Table 6. Physical properties of limestone (LS)

Table 7. Chemical composition of LSP

2.2 Steel reinforcement

The current investigation utilised steel reinforcing bars with sizes of Ø8, Ø10, and Ø12 mm for both the longitudinal reinforcement and stirrups. Three samples of each bar diameter were subjected to testing utilising the available testing machine at the laboratory of the Civil Engineering Department at Basra University. where the properties of the reinforcing bars are presented in Table 8.

Table 8. Test results of steel reinforcement

2.3 Mixing procedure

Two types of self-compacting concrete were produced: one using natural coarse aggregates and the other using recycled concrete aggregates to partially replace the natural coarse aggregate.

Various replacement ratios (0%, 25%, 50%, and 75%) were employed to substitute the natural coarse aggregate with recycled aggregate. Table 9 shows the quantities of materials included in each type of mixture, and the highlighted mix (number two) was used in the present experimental program.

Where the concrete mixture was produced using a Rotary drum mixer with a capacity of 0.1 m³.

The process can be succinctly described as follows: First, a combination of cement, limestone powder, and aggregate was mixed together for a duration of 1 minute. Second, 80% of the mixing water was gradually contained in the mixture, and the mixing process was continued for an additional minute. Third, the remaining 20% of water mixed with the dissolved superplasticizer was slowly added to the mixture while continuing to mix for an additional minute. To avoid the conglomerate in the mix and offer higher workability and a good consistency for the concrete mix, Fourth, the mixing process was extended for an additional duration of three minutes. Fifth, the mixture was allowed to sit undisturbed for a duration of three and a half minutes.

Finally, the mix was remixed for half a minute before being discharged to conduct fresh concrete testing and cast the specimens. For concrete mixes with recycled aggregate replacement, mixing took longer than expected. A quick mixing process must be considered to ensure a uniform distribution of the concrete components, especially the recycled aggregates.

Table 9. The trail mixes investigated

The loading frame consists of an I-section girder that transfers the load from the testing machine, developed to fit the designed models, to two I-section wing arms. These arms are bolted to the ends of the concrete beams as shown in Figures 3 and 4. The testing machine applies the load centrally to the middle of the main I-section steel arm, which then directly transfers the load to the wing arms, generating torque on the tested beam. The main steel arm is connected to the wing arms by hinges to allow free rotation. The load is precisely applied at the center of the main arm.

3.png

Figure 3. Test specimen

4.png

Figure 4. Steel frame and beam setup

6.1 General behaviour of beams under loading and cracks patterns

Initially, cracks manifest on the vertical surfaces of the beam, and subsequently spread to all sides. Geometrically, the phenomenon can be referred to as torsion, while visible cracks manifest as crack torsion. Given that all the beams tested have rectangular sections, it follows that the initial formation of cracks can occur uniformly from one side.

With an increase in the load, further torsional cracks formed in the pure torsion span. When the maximum torque is applied, the number of diagonal cracks on each side grows and eventually forms a spiral shape until the specimen fails due to complete breaking caused by torsion.

Beams lacking longitudinal bars and stirrups exhibited cracking torque that was equivalent to the ultimate torque. These beams had no torsional ductility due to the absence of steel reinforcement in both the longitudinal and transverse directions, which would have provided resistance against the applied torque beyond the cracking stage. The beam experienced an abrupt failure, resulting in its separation into two distinct parts. This occurred when the initial cracks appeared, as depicted in Figure 5 (referred to as failure mode 'A').

5.png

Figure 5. Failure mode of beams (mode “A”)

6.png

Figure 6. Failure mode of beams (mode “B”)

7.png

Figure 7. Failure mode of beams (mode “C”)

Figure 6 (failure mode "B") depicts beams that solely contain longitudinal bars. The beams exhibited brittle torsional failure, characterised by the development of a single intense helical crack. However, the presence of bars prevented the specimens from breaking into two separate pieces. Following the initial stage of cracking, one of the inclined individual cracks propagates and forms a spiral pattern around the surfaces of the member, gradually increasing in breadth until the point of collapse. By steadily increasing the load applied after the first cracking load, one of the inclined separate cracks expanded further along the beam until it reached the failure point known as the ultimate torque stage.

The beam's resistance increases after the initial breaking due to the presence of longitudinal reinforcing steel bars. This leads to an enhancement in the bonding of the concrete under loading, preventing a catastrophic collapse when the concrete reaches its ultimate tensile strength.

Structural elements reinforced with both longitudinal and transverse reinforcement. These exhibited a rise. The post-cracking behaviour leads to the creation of several helical and radial cracks.

In these instances, when the torque increased, the fissures expanded, grew in size, and became more noticeable. Ultimately, the beams broke, as depicted in Figure 7 (failure mode "C").

6.2 Cracking and ultimate loads

In this study, the prediction of pure torsion loading on beam specimens reveals two stages of behaviour, adopting a combination of two different theories. Initially, where no cracks form and the beams remain in the elastic state, the loading is described by a smeared crack analysis for plain concrete in torsion [13]. At the final stage, when loading reaches the maximum principal tensile stress of the concrete, the post-cracking behaviour is modeled by the well-known softened truss model [14-16], causing cracks to form and leading to the failure of an unreinforced member.

In the first stage, the load is resisted solely by the concrete. The addition of longitudinal steel without stirrups has little effect on the beam's strength under pure torsion, as it only helps resist the longitudinal component of diagonal tension forces. In the second stage, the load is resisted by a combination of both longitudinal and stirrup reinforcements. The cracking and ultimate load are detailed in Table 20.

Table 20. The results of the load of the tested beams

Figures 8 and 9 show beam is tested by subjection to torsional moments. The loading of the specimens causes certain changes in the behaviour as it transitions from the initial stage to the final stage. The cracking load is recorded as the load in which the cracks become visible to the recording camera by concrete crack width gauge.

8.png

Figure 8. Beam tested by subjection to torsional moments

9.png

Figure 9. The instruments used in the tests

6.3 Cracking and ultimate loads

Effect of RCA replacement ratio, by observing the results, no significant difference in behaviour was identified in the RCA-containing beams at the cracking stage, where all the beams exhibited similar linear torque-rotation behaviour. At the same time, RCA-containing beams show a decrease in cracking torque depending on the substitution ratio of RCA. The reduction in splitting tensile strength of RAC depends on what are the weak points between the binder and the old mortar in the RCA.

The torque is the moment that beams fail and cannot resist any increase in applied load. The ultimate capacity of the tested beam is determined by the highest torsional moment it can withstand. If the machine reading drops and the beam undergoes rapid deformation, it is considered a failure. Therefore, the cracking and ultimate torsional resistance of reinforced concrete members had been improved by adding steel although transverse reinforcement., such that any decrease in the machine reading accompanied by rapid deformation of the beam was described as failure.

Effect of longitudinal and transverse reinforcement, the load was applied on tested beam specimens up to failure. cracks manifested on the vertical surfaces and thereafter spread to all sides of the beam. Cracking torsion refers to the torsion level at which visible cracks become apparent. The augmentation in the initial fracture load for beams solely reinforced in the longitudinal direction, where the addition of longitudinal steel bars without stirrups has been effective in resisting the longitudinal component of the diagonal tension forces [17].

Where the results of the experimental test for beams showed that the effect of using transverse steel bars is clear with respect to the first cracking torsion resistance. the beams reinforced with transverse reinforcement demonstrated increment in both first cracking torque and ultimate torsional strength. This was due to the reinforcing steel being placed in an orderly manner in places exposed to tensile forces, which leads to an increase in the resistance of the material to bearing loads after it reaches the greatest tensile strength. cracking load but has a great effect on the ultimate load [16].

6.4 Torque - twist relationships

The angle of twist refers to the rotational displacement of the free end of a beam relative to its fixed end. Torsion occurs when a force is applied to a body, causing one end or part to rotate around a longitudinal axis while the other end or part rotates in the opposite direction. Table 21 illustrates the correlation between the torsion angle and the thresholds that were evaluated. Where the method used to calculate the angle of twist is performed by using a dial gauge attached to the bottom fiber of each end of the tested beam at a point (50 mm) from the centre of the longitudinal axis of the beam, as shown in Figure 9. The dial gauges recorded the downward displacement values to find the twist angle in radians, as follows:

The angle of twist (θ) = tan -1 Δ/L                 (1)

where,

Δ= dial gauge reading;

L= distance from the centre of the beam to the position of the dial gauge = 50.

The test findings indicate that the relationship between torque and twist angle is linear until the occurrence of the first cracks. After that, the behaviour becomes nonlinear until the failure stage, which is referred to as TWIST. The twist angle was measured during the application of loads until the failure stage.

Table 21. Values of torque and corresponding angle of twist of beams

The results of the angle of twist analysis indicate that the extent of concrete damage is directly influenced by the amount of replacement, with a corresponding increase observed as the replacement percentage rises. The beam experienced an increased impact due to an elevated twist immediately following cracking since the released energy resulted in a more brittle crushing, as observed in Figures 10-12, whilst at comparison among beams that also have the same characteristics but with different reinforcement content, The result revealed that increasing the reinforcement content in the section will increase the value of the angle of twist of the beam clearly and result in an increase in stiffness, due to increasing the area under the curve (stiffness) and the energy absorption capacity (ductility).is elucidated in Figures 13 and 14.

10.png

Figure 10. Torque-twist angle (RCA 25%)

11.png

Figure 11. Torque-twist angle (RCA 50%)

12.png

Figure 12. Torque-twist angle (RCA 75%)

13.png

Figure 13. Torque-twist angle (beams with longitudinal reinforcement)

14.png

Figure 14. Torque-twist angle (beams with longitudinal and stirrups reinforcement)

The Figure 15 shows the results of cracking with the ultimate of the tested beams. At ultimate torque, the angle of twist increases due to the addition of reinforcement content, but still, the solid section has a smaller angle of about 13.89% when compared with hollow sections.

15.png

Figure 15. Loading (cracking, ultimate) for specimens