EFFECT OF NEAR-SURFACE MOUNTED BARS ON THE STRUCTURAL BEHAVIOR OF ONE-WAY RC SLAB

The present research paper deals with the finite element modelling of continuous one-way reinforced concrete (RC) slabs strengthened by near surface mounted (NSM) bars under static loads. Seven specimens that consisted of one reference slab without strengthening and six strengthened slabs were used, with 2400 mm total length, 500 mm width and 100 mm thickness. The average compressive concrete strength at 28 days’ age was about 31MPa was used. Slabs were simulated with finite element computer program. The results of the control and strengthened slabs were compared with the experimental results. It is observed that the results of the strengthened slabs are in close agreement with the experimental results with a maximum difference in the ultimate load of about 6.64%.


INTRODUCTION
In Europe, the use of near surface mounted (NSM) bar technique was advanced for strengthening of reinforced concrete (RC) structures in the early 1950s. In 1948, a RC bridge deck in Sweden wanted to be upgraded in its hogging zone due to an excessive settlement of the steel cage through building. This was done by adding steel reinforcement bars into concrete grooves and by filling them with cement mortar (Parretti and Nanni, 2004). Steel corrosion in RC structures and elements is a main problem in the United States and around the world, costing billions of dollars in repairs and damages that are needed. It is also very useful to use externally bonded reinforcement for FRP and NSM in order to stabilize and repair but also to avoid further corrosion in concrete structures (Augustine 2013). Using NSM reinforcement can raise the flexural capacity, stiffness and reduces the crack width more than using externally bonded (EB) reinforcement. Experimentally and numerically study on repairing and strengthening of selfcompacted concrete slabs was presented by Ashteyat et al, 2019. The effectiveness of straight and inclined NSM carbon-fiber-reinforced polymers CFRP on the strengthening and repairing of self-compacted concrete slabs. It was found that the ultimate strength of one-way slabs increased by 45%-163% when the strengthened by either inclined and straight NSM-CFRP strips. In addition, repairing of one-way slabs resulted in a recovered the ultimate strength to about 223% of the control slabs ultimate strength, with a considerable enhancement in stiffness that reached to 296%. Abdallah et al. (2020) used NSM carbon and fiber-reinforced polymer (FRP) rods to strengthen the continuous reinforced RC beams. Length, type and ratio of FRP rods in addition to the filling material were the main tested parameters. Test results showed that all the capacity of al strengthened specimens with NSM-FRP increased compared to the control beam. In addition, the failure mode was changed from pull out of the FRP-rods to premature peeling of concrete cover when the rods are terminated before the zero moment point. Most researchers have adopted carbon fiber reinforced polymer (CFRP) as strengthening materials due to its high strength to weight ratio. But there are very limited researches dealt with using steel bars as strengthening materials such as strengthening of corbels (Shakir and Kamonna 2018) shear strengthening of beams (Thamrin et al. 2019), torsional strengthening of beams (Askandar and Mahmood 2020) flexural strengthen of continuous beam (Abdzaid and Kamonna 2019) and deep beams (Kamonna and Al-Issawi 2018, Al-Issawi and Kamonna 2020). There is very limited knowledge of adopting the NSM method by steel bars for strengthening of continuous one-way RC slabs, although it is availability in addition to it is ability of accounting of redistribution load beyond formation of plastic hinge which may occur 14 Kamonna and Abd Al-Sada in continuous members. This phenomenon could not be account to in case of using CFRP martial due it is brittle nature.

EXPERMENTAL TEST
Six RC one-way continuous slabs that experimentally were numerically analyzed. These slabs consist of one unstrengthened slab (CS) as a control specimen and five strengthened slabs. Four of the strengthened slabs were strengthened by #10 steel bars that have different lengths and quantities as shown in Table 1. Carbon fiber rods of 10 mm in diameter that having tensile strength and modulus of elasticity of 2172 MPa and 124 GPa was adopted to strengthen one of the slab specimens. The last specimen was strengthened by 6 #10 steel bars. Epoxy-paste known as Sikadur ® 330 and produced by Sika was used to bond the strengthening bars. All slabs have 100 mm thick, 500 mm width, and 2400 mm total length. Tests were done under two-point loads centered at two spans that having clear lengths of 1100 mm. The compressive strength, splitting tensile strength and modulus of rupture of concrete were equal to 31 MPa, 2 MPa and 4.57 MPa respectively. Two meshes of steel reinforcement were used to reinforce slab specimens located at their top and bottom faces as shown in Fig 1. Each mesh consisted of #8 at 200 mm c,c as main reinforcement, and #8 at 380 mm c,c as secondary reinforcement. Table   2 lists properties of reinforcing and strengthening steel bars.

FINITE ELEMENT MODELING
In the FEM, the structure was divided into finite elements connected together by their nodes.
ANSYS software is the comment tool that used for this purpose. To achieve analysis process by ANSYS, it is required to define (element type, section and real constant for each element in addition to the applying loads, and boundary conditions).

Modeling.
Modeling of RC slab is done using: SOLID65 element is used for representing the concrete because of its ability in taking account of crushing and cracking of concrete that occurred when compression and tension stresses exceeded concrete compression and tension strengths. Besides to that, it has the feature of adding reinforcement material to the concrete element as shown in Fig 2. LINK180 element which is a uniaxial 3D bar-element is used for simulating steel reinforcement and NSM strengthening bars as shown in   Element attributes which adopted in the present works are shown in Table 3. A quarter of the entire model is shown in Fig. 5. The main and transverse steel reinforcement modeling are shown in Fig. 6. Modeling of NSM steel reinforcement is shown in Fig. 7.

Material Properties
In this analysis, three types of material properties were used. Solid 65 element was the first material (material number 1), this element requires the definition of linear isotropic, concrete parameters and multilinear properties show Table 4. The material number 2 was link 180. The element requires defining linear isotropic and bilinear properties show Table 5. The material number 3 for solid 185 was the last material and requires only linear isotropic, show Table 6.
The stress-strain constitutive relationships shown in Fig. 8 were used to simulate concrete behavior under compression and tension. A bilinear and linear stress-strain relationships are used to represent steel and carbon bars, as shown in Fig. 9. For simplicity, epoxy was modeled as concrete with full bond assumption between concrete and epoxy in one side and epoxy and steel on another side. So, Solid65 element is used to represent the epoxy.

Loads and Boundary Conditions
The simulation of a one-way RC slab was achieved for just a quarter of a continuous concrete slab using two planes of symmetry. Therefore, displacements perpendicular to planes of symmetry in the direction were set as zero (Uz= Ux=0). In addition, the displacements in all supports were modeled in such way they will be work as a roller (Uy=0)    concrete and between strengthening bars and epoxy. Beside to those, failure criteria of the concrete depended on uniaxial test whereas concrete elements in the slab body were exposed to either biaxial or triaxil stresses. In the other side, the solution improves more in the cases contained higher strengthening ratio due to more dependence on steel which considered as a more homogeneous material. Modeling of such materials will have more reasonable assumptions.

Ultimate load
It is apparent that the numerical ultimate loads from the FE analysis accord well with the subtending values of the ultimate experimental tests as appeared in Table 7. The maximum variance in the ultimate load was 6.64% for the tested and analyzed slabs that strengthened by steel bars.
Summarizing of the difference between the ultimate loads and maximum deflection obtained experimentally and numerically depends on the embedded length of NSM bars that put in the sagging and hogging zones (0.9 or 0.6) L, spacing of NSM bars (100mm or 70mm) as well as the material type of NSM bars (CFRP or steel). Results revealed that the ultimate load of one way slab increased by (131-214) % when the strengthened by NSM-technique. Increasing strengthening quantity by 50% resulted in an increasing ultimate load by about 12%. Also, using CFRP bars instead of steel bars as strengthening material resulting in a decrease in the ultimate load by about 8%. This may be due to that the CFRP bar is a brittle material that could not help with the formation of plastic hinges. Therefore, in statically indeterminate structure adopting steel bars as strengthening material is a better choice than CFRP bars due to their high ductility. It can be also seen that increasing the embedded bars in the positive moment region gave greater ultimate load than the case in which higher embedded length is used in the negative moment region.

PARAMETRIC STUDY
To extend of the knowledge on the behavior of one-way RC continues slabs strengthened by NSM bars, some parameters that may effect on the behavior of such member are presented in this section.

Concrete strength.
In the experimental test, concrete with 31 MPa compressive strength was used. The parametric study included four compressive strengths, one less than 31 MPa which is 20 MPa and the other more than 31 MPa that assumed as 40, 50 and 60 MPa. This new material quality will be implemented in SS94.9 specimen slab.
It can be seen from Fig. 24  where f`c is concrete compressive strength, 1 is a factor depends on concrete compressive strength and c is the depth of neutral axis measured from the extreme of compression face (Hassoun and Al-Manaseer, 2015).

Location of NSM bars.
In this section, effect of changing the location NSM bars i.e. (but them in the negative moment region only, positive moment region only or both regions) is investigated. Two values of NSM bars spacing are adopted which are 70 mm and 100mm.

Effect Location of NSM when adopting 70 mm NSM bars spacing
In the first specimen only the negative moment zone is strengthened. The results showed increasing in the ultimate load of about 135.5% compared with the CS, whereas only the positive moment zone is strengthened in the second specimen. The ultimate load was increased by 108% relative to CS as shown in Fig. 26. When using the strengthening bars just in the negative moment zone led to increase in the ultimate load about 13.2 % relative to the strengthening case when using strengthening bars in only the positive moment zone. "Deflection mm" f'c20 f'c31 f'c40 f'c50 f'c60

Effect of NSM Location when adopting 100 mm NSM bars spacing
The first specimen is strengthened only in the negative moment zone. The results showed increasing of about 108.47% in ultimate load compared with the CS. The second sample is only strengthened at the positive moment zone. The ultimate load raised by 143.23% in comparison to the as seen in Fig. 28. When using the strengthening bar in only the positive moment zone led to increase on ultimate load about 16.67 % relative to the case when the strengthening bars is put in only the negative moment region. Therefore, it can be concluded that when adopting a high strengthening ratio, strengthening only the negative moment region is better than strengthening only the positive moment region. A reverse action was obtained in case of adopting a small strengthening ratio. This is maybe attributed to that when adopting a high quantity of strengthening the probability of the formation of plastic hinges in the negative moment region may increase. So, the redistribution of loading will occur. This is due to higher capacity which increases the corresponding failure load. Therefore, strengthening the negative moment region gave better performance. On another side, when using small strengthen quantity causes a reduction in the slab capacity. This resulted in reducing the probability of the formation of the plastic hinge. So, the importance of strengthening the negative moment region may decrease, and strengthening the positive moment region will be better.
Effect of changing the strengthening location when adopting 100 mm NSM reinforcement spacing on the load-deflection curves is shown in Fig. 29, linear and nonlinear behavior is also recorded in these specimens. After that, the plastic hinges occurred approximately at the load 175 KN when using the strengthening bars just in the negative moment zone. Then, the slab failed gradually and showed higher ductility before failure.

Effect of bar diameter when using 70mm NSM bars spacing
The effect of NSM diameter bar is presented in Fig. 30. For slab SS96.9 steel bars having diameter of 10 mm have been used as NSM reinforcement. This Specimen illustrated an increase in the load capacity of about 9.92% compared to those strengthened by 8 mm bar diameter. While using of 12mm bar diameter gave an enhancement of about 3.74 % more than specimen SS96.9 which strengthened by 10 mm in diameter. Fig. 31 illustrates effect of changing the bar diameter of NSM strengthening bars on the loaddeflection curves. It can be seen that the stiffness considerably effected by changing bars diameter of strengthening bars beyond cracking stage. It can also be observed that when using 12mm bars diameter as NSM bars gave an enhancement of deflection about 6.73 % higher than those contain 10mm bars diameter and about 6.45% when using 8mm in diameter. Kamonna and Abd Al-Sada

Effect of bar diameter when using 100mm NSM bars spacing
Specimen SS94.9 showed an increase in load capacity of about 20.23% when steel bars diameter is changed from 8mm to 10mm. While increasing bars diameter from 10 mm to 12mm gave a decrease of about 8.15%. This may because that when using a greater diameter, steel bars will apply higher stresses on the concrete cover then those of a smaller diameter, thus causing a decrease in the slab capacity. The effect of NSM diameter bar is presented in Fig. 32.
Effect of changing strengthening bars diameter on the load-deflection curves is illustrated in Fig. 33. It is appearing that the deflection has the same path with load at the linear stage. Plastic hinge formed at the load (190, 230 and 220) KN when using the (8, 10 and 12) NSM bars diameter and after that, the slab failed gradually and showed more ductility before failure. Also it is shown that higher stiffness when using 12 mm bar diameter NSM at positive and negative moment zone, while smaller stiffness is obtained when the bar diameter became10mm or 8mm.