- Email: [email protected]
Effect of ferrocement jacketing on the flexural behaviour of beams with corroded reinforcements
Construction and Building Materials 121 (2016) 92–99
Contents lists available at ScienceDirect
Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat
Effect of ferrocement jacketing on the flexural behaviour of beams with corroded reinforcements S. Jayasree a,⇑, N. Ganesan b, Ruby Abraham c a
Department of Civil Engineering, MBCET, Trivandrum, Kerala, India Department of Civil Engineering, NIT Calicut, Kerala, India c RIT Kottayam, Kerala, India b
h i g h l i g h t s Influence of cover thickness and strength of concrete on accelerated corrosion was investigated. Effect of corrosion on the flexural behaviour of RC beams was studied. Flexural behaviour of corroded RC beams after retrofitting with ferrocement jacketing was evaluated.
a r t i c l e
i n f o
Article history: Received 30 December 2015 Received in revised form 12 May 2016 Accepted 22 May 2016
Keywords: Corrosion Accelerated corrosion Flexural behaviour Retrofitting Ferrocement jacketing
a b s t r a c t Corrosion of reinforcement is one of the main causes of deterioration of Reinforced Cement Concrete (RCC) structures which affects the load carrying capacity and its durability. Though it is very difficult to completely eliminate the chances of corrosion, suitable retrofitting strategies can be introduced as a measure for the retrofitting of corrosion damaged structures to gain its original strength. The present work deals with the study of degradation of the ultimate load carrying capacity of the flexural members due to corrosion. Twenty-one RCC beams were cast, out of which, three beams were kept as control specimens, and the remaining were subjected to varying levels of corrosion (5, 10 and 15%) so that six specimens are obtained for each level of corrosion. Accelerated corrosion was induced by means of impressed current method. From each level of corrosion, three beams were subjected to loading corresponding to 70% of the ultimate load given by the remaining three specimens corresponding to that level of corrosion. Subsequently these beams were retrofitted by means of a U-wrap ferrocement jacketing. All the beams were tested under two point loading and the strength and behaviour of retrofitted beams were compared with control specimens. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction Corrosion of steel reinforcements is considered to be the reason behind the reduced service-life and failure of RCC structures. Use of pre-rusted steel, exposure to corrosive environment, loading conditions etc are some of the factors which impart corrosion in RCC structures. The effect of corrosion on RCC structures include cracking of the concrete cover, reduction and eventually loss of bond between concrete and corroded reinforcement and reduction of cross-sectional area of reinforcing steel [1]. Basically, the corrosion process has the ability to start in any environmental conditions and the rate of corrosion is determined by the harshness of the environment. Reinforcing steel is normally passive in concrete due to ⇑ Corresponding author. E-mail address: [email protected] (S. Jayasree). http://dx.doi.org/10.1016/j.conbuildmat.2016.05.131 0950-0618/Ó 2016 Elsevier Ltd. All rights reserved.
high alkalinity of the concrete pore solution. However, the penetration of harmful gases like CO into the concrete destroys this inhibitive property of the concrete and leads to corrosion [2]. The effect of corrosion on the load-bearing capacity and mechanical performance of corroded beams, exposed to wetting and drying cycles in a chloride environment under sustained loading without impressed current was also investigated by Yu et al. [3]. The seismic performance of reinforced concrete beams with corrosion induced in the transverse steel reinforcement by means of cyclic loading was studied by Ou and Chen and the results indicated that pitting corrosion increased with an increase in the corrosion level and the beams could sustain a corrosion weight loss of 6% in the hoops and maintain ductile flexural behaviour [4]. Experimental investigation of the behaviour of highly corroded reinforced concrete beams and also a study on the performance of a 27 year old corroded beam, revealed, brittle failure mode for the corroded RC
93
S. Jayasree et al. / Construction and Building Materials 121 (2016) 92–99
beams [5,6]. The influence of simultaneous loading and corrosion of reinforcement on the structural performance of concrete beams was investigated and the results indicated that corrosion or an occasional over-loading or both are likely to cause concrete structures under service loads to collapse suddenly without significant deflection [7]. An experimental investigation on the flexural failure behaviour of RC beams having different weight loss of 0, 5, 10 and 30% due to corrosion was carried out by Okude et.al and it was reported that the rebars were broken at the final stage of loading [8]. With increasing duration of exposure to a corrosive environment, the steel mass loss increases appreciably, which lead to a significant reduction of the tensile ductility of the material [9]. Studies in the past indicated that the bond strength slightly improves at lower levels of corrosion. This is mainly due to the development of Ferrous oxide surrounding the reinforcements which offer a higher frictional force. However as the degree of corrosion increases the bond strength decreases rapidly due to the drastic reduction of diameter of bars and the ineffective portion of corrosion products on the reinforcements. Suitable retrofitting strategies like jacketing using RCC, steel, Fibre-reinforced polymer and ferrocement can be introduced to reinstatement of corrosion damaged structures. The ultimate compressive strength increases with the change in orientation of square mesh from 90° to 45° if ferrocement is used as an external confinement to concrete specimen [10]. Also it was reported that strengthening through cast in situ ferro-mesh layer is the most efficient method [11]. CFRP sheets increased the fatigue capacity of the beams with corroded steel reinforcement beyond that of the control beams with un-corroded steel reinforcement [12,13]. The use of ferrocement as an external confinement to concrete specimens enhances the ultimate concrete compressive strength and failure strains [14]. The impressed current technique has been frequently used to study the effect of reinforcement corrosion on the cracking of concrete cover, bond behaviour, and load-bearing capacity of reinforced concrete structural members [15]. The advantage of this method over other accelerated techniques is the ability to control the rate of corrosion, which usually varies due to changes in the resistivity, oxygen concentration, and temperature. The predicted and actual mass loss values using impressed current method are very close and hence this technique can be used for studies involving accelerated corrosion [16,17]. The present paper deals with 1. The flexural behaviour of RC beams containing varying degrees of corroded reinforcement (5%, 10%, and 15% mass loss of reinforcement), and to compare the results with the control beam (beam with non-corroded reinforcement). 2. The effect of ferrocement jacketing in beams with different levels of corroded reinforcement and to compare the results with the non-retrofitted corroded beam, as well as with the control beam.
Fig. 1. Accelerated corrosion setup (Schematic diagram).
where, m is the mass of steel consumed (g), M is the atomic weight of metal (55.8 g for Fe), I is the current (amperes), t is the time (seconds), z is the ionic charge (2), and F is the Faraday’s constant (96,485 A/sec). 2. Material properties and mix proportion Materials and mix obtained in this investigation are suitable for M20 grade concrete which is generally used in all the RCC structures in the past and these structures are subjected to different degrees of corrosion depending on the environmental condition. As this investigation is deals with the effect of rehabilitation by ferrocement, the flexural members considered in this study were made of M 20 grade concrete. Portland Pozzolana Cement, crushed stones of 20 mm coarse aggregate, manufactured sand passing through sieve of size 4.75 mm and confirming to zone II of IS 383-1970 (reaffirmed 2002) as fine aggregates were used [18,19]. The mix design was done as per IS 10262-2009, to obtain a M20 grade concrete [20]. The mix proportion thus obtained was 1:1.8:2.9. 3. Experimental programme 3.1. Preliminary studies Preliminary studies were conducted in order to determine whether the desired percentage of mass loss is obtained during the time calculated using Faradays’ equation for the corrosion of steel reinforcements. 3.1.1. Determination of actual corrosion and corresponding strength of single bar Bars of length 800 mm and varying diameters 6 mm, 8 mm, and 10 mm (3 samples each), were subjected to different levels of corrosion (5, 10 and 15% of mass loss), using accelerated corrosion technique. For this study the current was fixed as, 8A. The time
1.1. Accelerated corrosion setup Accelerated corrosion was induced by means of impressed current technique. Fig. 1 shows the schematic diagram of accelerated corrosion setup. In order to induce corrosion in the steel rebars, an electrolytic cell with the rebar acting as anode and stainless steel rod as cathode, dipped in 4% NaCl solution acting as electrolyte was used. A constant DC voltage was applied to the electrolytic cell. The time required for achieving different levels of corrosion by mass loss was calculated using Faraday’s Law [15] as per the following Eq. (1).
m¼
M It zF
ð1Þ
Table 1 Time and mass loss for single bare bars. Diameter (mm)
Theoretical mass loss (%)
Time (seconds)
Actual mass loss (%)
10
5 10 15
9585.00 19170.00 28755.00
6.61 9.75 15.03
8
5 10 15
6590.54 13181.31 19771.64
6.02 9.83 14.98
6
5 10 15
3801.90 7603.81 11405.72
5.13 10.03 15.10
94
S. Jayasree et al. / Construction and Building Materials 121 (2016) 92–99
required for each degrees of corrosion level was calculated for different bar sizes and the details are shown in Table 1. In order to compare the theoretical and actual mass loss, the weights of individual bars were measured before the corrosion process. For the study, stainless steel rod of 6 mm diameter was used as cathode, and the reinforcement to be corroded was used as the anode. The electrodes were immersed in a 4% NaCl solution and electric potential was applied between cathode and anode using a DC power source of 10A capacity. The bars were protected with a heat shrink tubing at the ends, for a distance of 200 mm, in order to avoid corrosion at the ends, thereby, providing sufficient grip for the tension test in the UTM. After the corrosion process, the bars were cleaned to free them from rust, as per ASTM G-1 [21] and the specimens were weighed, and the actual mass loss was found out. Fig. 2 shows the experimental setup. The experimental values of actual mass loss and theoretical mass loss were given in Table 1. From Table 1, it may be noted that the values of actual mass loss and theoretical mass loss were almost same for bare bar specimens. Physical appearance of the corroded bars shows a more or less uniform loss in the diameter and general corrosion predominates than pitting corrosion. Tension test was conducted on all the bars and the yield stress, ultimate stress and percentage elongation were calculated and are given in Table 2. It can be observed that a considerable reduction in yield and ultimate stress were imparted due to corrosion. 3.1.2. Determination of actual corrosion in bundle of bars Bars in groups were tested in order to check the suitability of the impressed current method when electric potential is applied to one of the reinforcements, in a bundle of bars which were connected together by stirrups at the two ends. The main reinforcements comprises of two numbers of 10 mm and two numbers of 8 mm diameter bars and were connected by stirrups of diameter 6 mm. The bars were then subjected to accelerated corrosion, in varying levels (5, 10, and 15%). It was noted that each of the bars, including the stirrups suffered almost equal and uniform metal loss throughout the surface. Also the bars were individually weighed before and after the corrosion process. Fig. 3 shows the specimens subjected to corrosion and Table 3 gives actual percentage of mass loss. 3.1.3. Effect of concrete cover on the corrosion of reinforcements The fore mentioned preliminary study was also extended to RCC specimens to understand the effect of concrete resistance to the time calculated using Eq. (1) for corrosion. Specimens of 100 100 500 mm were prepared, as shown in Fig. 4, having two numbers of 8 mm diameter bars at the top and two numbers of 10 mm diameter bars at the bottom with 6 mm diameter
Table 2 Results of the tension test. Diameter of bars (mm)
Corrosion (%)
Yield strength (N/mm2)
Decrease in yield strength (%)
Ultimate strength (N/mm2)
Elongation (%)
10
0 5 10 15
308.13 285.31 256.36 228.89
0.00 7.50 16.90 25.70
451.93 413.12 387.97 315.94
18.20 18.00 11.50 08.30
8
0 5 10 15
410.31 359.02 348.76 307.73
0.00 12.40 14.90 24.90
461.59 410.31 379.53 359.02
21.10 20.60 13.92 09.07
6
0 5 10 15
499.26 445.77 427.94 410.11
0.00 10.80 14.20 17.80
552.75 517.09 481.43 463.59
17.30 16.83 11.40 08.65
Fig. 3. Bare bars in group.
Table 3 Time and mass loss for bare bars in group. Theoretical mass loss (%)
Current (A)
Time (s)
Actual mass loss (%)
5 10 15
8 8 8
15097.10 30194.19 45291.29
4.94 10.22 15.12
Fig. 4. Accelerated corrosion in rebars embedded in concrete specimens.
Fig. 2. Accelerated corrosion on bare bars (experimental setup).
stirrups with a concrete mix of M 20 grade. Before casting initial weight of the individual bars were measured. The specimens were then subjected to accelerated corrosion in order to achieve varying
S. Jayasree et al. / Construction and Building Materials 121 (2016) 92–99
levels of corrosion (5, 10, and 15%) in reinforcements. After the corrosion process, the specimens were demolished and weight of the cleaned corroded reinforcement was taken to find out the actual mass loss and which are shown in Table 4. It may be seen from the Table 4 that the actual mass loss was less when compared to the theoretical mass loss. In order to compensate this, the calculated time was suitably modified by a factor k as given below.
Theoretical mass loss k¼ Actual mass loss
ð2Þ
By modifying Eq. (1), the time required to provide a certain level of mass loss for reinforcements embedded in concrete is given by:
t ðsecÞ ¼
kmzF IM
ð3Þ
The average value of k obtained for specimens subjected to various corrosion levels is also given in Table 4. Since the values are approximately same, an average value of 1.65 was considered for future study. 3.2. Flexural behaviour of RCC beams Flexural behaviour of RCC beams with corroded reinforcements (corroded beams) were compared with that of beams with noncorroded reinforcements (control beams). Also the study was extended to ferrocement retrofitted RCC beams with corroded reinforcements (5%, 10% and 15%) and the behaviour was compared with that of control beam and corresponding non retrofitted corroded beams. 3.2.1. Specimen details Fig. 5 shows the geometry and reinforcement details of the specimens. The overall dimension of the specimens is 100 150 1000 mm. All the beams were designed as under reinforced sections [22]. The reinforcement consists of two numbers of 10 mm diameter bars at the tension face and two numbers 8 mm diameter bars at the compression face. Stirrups consists of 6 mm diameter bars at a spacing of 90 mm. Electrical strain gauges were attached to the reinforcement. A total number of 21 specimens were prepared in steel moulds, using M20 grade concrete. After 24 h of casting, the specimens were demoulded and kept immersed in water for 28 days. Three beams were designated as control specimens (B0). The remaining 18 beams were subjected to varying levels of 5, 10
and 15% of accelerated corrosion. Three beams in each group corresponding to 5, 10 and 15% of corrosion are designated as B5, B10 and B15. They were kept for studying the properties of corroded beams and the remaining specimens were kept for retrofitting and designated as RB5, RB10 and RB15(three in each group). 3.2.2. Accelerated corrosion Specimens were kept in a stainless steel chamber which serves as cathode and one of the 10 mm bar to which a terminal already connected before casting was serving as anode. The electrolytic medium in the corrosion cell was 4% NaCl solution. When the reinforcing steel was used as anode, there were chances of loss of connection between the electrical supply and the rebar due to loss of metal at the terminal. In order to avoid this, internal threading was done on the rebar and the terminal was rigidly connected to the rebar by means of screws as shown in Fig. 6. The end of the rebar, which houses the terminal, was protected by heat shrink tubing as shown in Fig. 7. Fig. 8 shows the electrical connection to the reinforcement. The time required for different corrosion levels were computed using Eq. (2). A constant current at 2.24 A was applied between the terminals. In order to keep the specimens, during corrosion process, masonry tanks of size 200 250 1200 mm were used as shown in Fig. 9. The stainless steel cathode was placed inside the tanks. Small wooden blocks were placed under the beams in order to separate the specimen from cathode to facilitate free flow of electrons through the electrolyte. Constant current was applied to three specimens simultaneously, by means of three different power sources. The setup was left undisturbed, until the time required for a particular level of corrosion was attained. Then the beams were taken out and kept ready for testing in a dry environment. 3.2.3. Testing All the beams were tested under 2 point loading in a Universal Testing Machine of 1000 kN capacity and the experimental setup is
Table 4 Time and mass loss in steel embedded concrete specimen. Theoretical mass loss (%)
Time (s)
Initial weight (Wi)
Final weight (Wf)
Actual corrosion (%)
Value of k
5 10 15
64395.00 124388.80 185582.40
988.03 988.06 981.04
957.56 928.19 892.68
3.08 6.06 9.01
1.62 1.65 1.665
95
Fig. 5. Specimen details.
Fig. 6. Terminals connected to the rebars by screws.
96
S. Jayasree et al. / Construction and Building Materials 121 (2016) 92–99
Fig. 10. Testing of beam.
Fig. 7. Terminal housed in heat shrink tubing.
load increments of 2.5 kN. In addition to this width of cracks was measured using a crack detection microscope of 50 magnification. Three beams from each group were tested to investigate the flexural properties for different degrees of corrosion levels. 3.3. Rehabilitation of specimens using ferrocement jacketing After accelerated corrosion, beams designated as RB5, RB10 and RB15 were preloaded to 70% of their ultimate load capacities, prior to the retrofitting process. Fig. 11 shows the preloaded specimens exhibiting flexural cracks ready for retrofitting. A suitable epoxy was used for developing bond between concrete surface and wire mesh. For ferrocement retrofitting, two layers of wire mesh 6 22 SWG (volume fraction 1.2%) of tensile strength 862.3 N/mm2 were wrapped around the beams, as shown in Fig. 12. The matrix consists of cement sand mortar of 1:2 ratio by weight and the compressive strength of mortar is 36 N/mm2. After retrofitting, the beams were cured under water for 14 days. All the beams were then tested under two point loading in the same way as that of corroded beams.
Fig. 8. Electrical terminal connected reinforcement.
4. Results and discussion 4.1. Behaviour of specimens under load
Fig. 9. Induction of accelerated corrosion to the beams.
shown in Fig. 10. The span of the beam was kept as 900 mm. Two LVDT’s (Linear Variable Differential Transformer) were employed in the compression and tensile regions to measure the deformations. Dial gauges were placed at midpoint of the span and under the loading points in the soffit of the beam. Since electrical strain gauges were damaged during accelerated corrosion the strains could not be monitored. For control specimens the electrical strain gauge readings were taken. All the measurements were made at
During testing, a large number of multiple fine cracks were observed in the control specimens. As the level of corrosion increased, wider cracks were noticed. Cracks observed parallel to main reinforcements in corroded beams were absent in the case of retrofitted beams. Also spalling of cover concrete were observed for the corroded specimens. Crack pattern of retrofitted beams, were almost similar to that of control beams. Figs. 13–16 show the crack patterns of corroded and retrofitted beams. Propagation of cracks and corresponding crack width of specimens for each load interval is shown in Fig. 17. The cracks developed in retrofitted beams, are finer than those in corroded beams. Also large number of closely spaced finer cracks were observed in the retrofitted specimens. For 15% corroded specimens, the maximum deflection was observed at the load point and not in the middle, may be due to the formation of localised corrosion. Moreover, spalling of cover was noticed in corroded beams during testing. 4.2. First crack load and ultimate load The first crack load was obtained from load deflection graph and it is the load point at which the curve deviates from linearity. The ultimate load was obtained from the experimental tests.
S. Jayasree et al. / Construction and Building Materials 121 (2016) 92–99
97
Fig. 11. Pre-loaded specimens exhibiting flexural cracks.
Fig. 12. Wrapping of wire mesh (U-Wrap).
Fig. 14. RB5 compared with B5 and B0.
Fig. 13. Corroded beams with varying levels of corrosion.
Fig. 15. RB10 compared with B10 and B0.
Both values are shown in Table 5. There was a decrease in first crack load as well as ultimate load, as the level of corrosion increased. It can be observed that due to ferrocement jacketing with two layers of wire mesh, the ultimate load carrying capacities were completely restored except for RB15, which had a higher degree of corrosion (15%). This indicates that the number of layers of wire mesh provided may not be adequate and further tests are required with more number of layers to have effective retrofitting.
4.3. Load deflection behaviour Fig. 18 shows the load – deflection behaviour of corroded and retrofitted beams. The curve is linear up to the formation of first crack load and then it became nonlinear with multiple cracks. Due to experimental limitations, the load deflection could be traced only up to 80% of the post peak loading in the descending
98
S. Jayasree et al. / Construction and Building Materials 121 (2016) 92–99
portion of the curve. It may be noted that the corroded specimens show high values of deflection at lower loads, compared to that given by control specimens. In the case of retrofitted beams, it may be noted that the deflections were less, compared to the control specimens. The area under the load deflection curve indicates the energy absorption capacity (EAC) and the values were calculated for all the specimens and are shown in Table 5. It can be seen that the energy absorption capacity of corroded beams were less compared to that of control beams. This property may be due to the loss of ductility of reinforcement due to corrosion. For RB15 specimens, the calculated value of EAC was less when compared with B0. This may be due to the fact that the two layer ferrocement retrofitting could not restore the ultimate load to the level of control beams. The stiffness was obtained from the slope of the tangent drawn to the load deflection curve and are given in Table 5. There was a decrease in stiffness, with an increase in level of corrosion and it may be noted from the table that there is a significant reduction in stiffness as the level of corrosion increases. Also it may be noted from Table 5 that the stiffness of retrofitted specimens were higher than those of control specimens.
Fig. 16. RB15 compared with B15 and B0.
B0
70
RB5
RB10
RB15
4.4. Moment curvature relationship
LOAD (KN)
60
Fig. 19 shows the comparison of moment curvature relationship of corroded and retrofitted beams. The curve is linear up to first crack moment and beyond which, the curve shifts from linearity. When it reaches yield moment, the curves becomes more or less flat. When steel yields, a large increase in curvature occurs with a small change in moment.
50 40 30 20 10 0
0
0.2
0.4
0.6
0.8
1
CRACK WIDTH (MM)
4.5. Ductility index Displacement ductility was calculated as the ratio of the displacement at ultimate load to the displacement at yield load and the values are given in Table 6. The large increase in curvature, before collapse of the beam is an indication of ductile failure of
Fig. 17. Crack width propagation in corroded beams.
Property
B0
B5
B10
B15
RB5
RB10
RB15
First crack load Ultimate load Yield load EAC (kN mm) % variation Stiffness (kN/mm) Stiffness variation Toughness % variation
11.5 57.0 45.0 107.0 1.0 30.0 1.0 38.9 1.0
11.0 52.3 37.0 79.0 26.0 27.2 9.3 28.7 26.2
10.0 44.8 36.5 63.0 41.0 24.3 19.0 26.1 32.9
37.0 31.0 22.6 46.0 59.0 20.6 31.3 24.4 37.4
13.0 66.3 37.0 118.0 10.3 50.6 68.7 70.1 80.0
12.0 62.3 31.0 113.0 5.6 45.4 51.3 69.7 78.9
9.5 52.0 22.6 80.0 25.3 36.0 20.0 66.0 69.5
12
B0
10
B5
8
B10
6
B15
4
RB5
2
RB10
Moment (kNm)
Table 5 Comparison of engineering properties of corroded and retrofitted beams.
0
RB15
0
0.01
0.02
0.03
0.04
0.05
Curvature (M-1) 80
LOAD (kN)
B5
60 50
B10
40
B15
30
RB5
20
RB10
10 0
Fig. 19. Comparison of Moment-Curvature plots.
B0
70
RB15
0
2
4 6 8 DEFLECTION (mm)
10
Fig. 18. Comparison of Load-Deflection plots.
12
Table 6 Comparison of ductility index. Specimen
B0 B5 B10 B15 RB5 RB10 RB15
Displacement ductility
Curvature ductility
Absolute
Relative
Absolute
Relative
2.41 4.32 5.11 7.57 2.10 2.855 3.71
1.00 1.79 2.12 3.10 0.87 1.18 1.54
2.98 5.44 5.90 9.07 3.41 6.13 7.40
1.00 1.83 1.98 3.04 1.15 2.05 2.48
S. Jayasree et al. / Construction and Building Materials 121 (2016) 92–99
beam. Curvature ductility was also calculated as the ratio of curvature at ultimate load to that of curvature at yield load. As the percentage of corrosion is increased, the value of calculated ductility factor increased, and this may be attributed to the large deflections recorded in the corroded beams, due to loss of bond between steel and concrete, and spalling of cover concrete due to corrosion. Table 6 shows the displacement ductility and curvature ductility of both corroded and retrofitted beams. 5. Conclusions Most of the countries in the world have long coastal lines. Structures located along the coastal line are subjected to different degrees or levels of corrosion. Structures located along the coastal line are subjected to different degrees or levels of corrosion. These structures have to be audited to assess the amount of corrosion that have already occurred and suitable rehabilitative measures have to be arrived at. Considering these aspects the present investigation is carried out and the main aim of this investigation is to study the type of rehabilitative procedure to be adopted in the case of structures subjected to partial corrosion. The following conclusions were arrived based on the present experimental investigation. &
&
&
&
&
&
The impressed current technique can be used for accelerated corrosion in concrete specimens by modifying Faradays’ law. As the corrosion rate increases, the ultimate stress and yield stress of rebars were found to decrease considerably. At 15% level of corrosion, width of cracks due to corrosion surpassed the allowable crack width as per IS 456:2000. Mechanical properties such as ultimate load carrying capacity and stiffness were found to decrease as percentage loss of mass due to corrosion in reinforcement increases. As the percentage of corrosion increases, the deflections observed in the corroded specimens were greater than that of control specimens due to loss of bond between steel and concrete. The deflection of the corroded beams were found to reduce considerably due to ferrocement retrofitting. Ferrocement retrofitting with 1.2% volume fraction of mesh reinforcement improved the ultimate load carrying capacity of corroded beams (B5 & B10) significantly higher than that of control specimens. On the other hand, RB15 specimens with same volume fraction of mesh reinforcement improved only 42.5% of ultimate load carrying capacity of B15. This indicates that ferrocement retrofitting with 1.2% volume fraction of mesh reinforcement is inadequate for higher corrosion levels and hence additional layer of mesh reinforcement may be required for restoring the ultimate load carrying capacity.
Acknowledgements The authors would like to thank the Kerala State Council for Science Technology and Environment (KSCSTE) for the financial
99
assistance for the purchase of equipments required for this work and the Mar Baselios College of Engineering and Technology, Trivandrum and National Institute of Technology Calicut for supporting the work. References [1] D.V. Val, L. Chernin, M.G. Stewart, Experimental and numerical investigation of corrosion-induced cover cracking in reinforced concrete structures, J. Struct. Eng. 135 (2009) 376–385. [2] D.V. Val, Deterioration of strength of RC beams due to corrosion and its influence on beam reliability, J. Struct. Eng. (2007) 1297–1306. [3] L. Yu, R. François, V.H. Dang, V. L’Hostis, R. Gagné, Structural performance of RC beams damaged by natural corrosion under sustained loading in a chloride environment, Eng. Struct. 96 (2015) 30–40, http://dx.doi.org/10.1016/j. engstruct.2015.04.001. [4] Y. Ou, H. Chen, Cyclic behavior of reinforced concrete beams with corroded transverse steel reinforcement, J. Struct. Eng. (2014) 1–10, http://dx.doi.org/ 10.1061/(ASCE)ST.1943-541X.0000932. [5] W. Zhu, R. François, D. Coronelli, D. Cleland, Effect of corrosion of reinforcement on the mechanical behaviour of highly corroded RC beams, Eng. Struct. 56 (2013) 544–554, http://dx.doi.org/10.1016/j. engstruct.2013.04.017. [6] V.H. Dang, R. François, Influence of long-term corrosion in chloride environment on mechanical behaviour of RC beam, Eng. Struct. 48 (2013) 558–568, http://dx.doi.org/10.1016/j.engstruct.2012.09.021. [7] Y. Du, M. Cullen, C. Li, Structural performance of RC beams under simultaneous loading and reinforcement corrosion, Constr. Build. Mater. 38 (2013) 472–481, http://dx.doi.org/10.1016/j.conbuildmat.2012.08.010. [8] N. Okude, M. Kunieda, T. Shiotani, Flexural failure behavior of RC beams with rebar corrosion and damage evaluation by acoustic emission, J. Acoustic Emission 27 (2009) 263–271. [9] C.A. Apostolopoulos, V.G. Papadakis, Consequences of steel corrosion on the ductility properties of reinforcement bar, Constr. Build. Mater. 22 (2008) 2316–2324, http://dx.doi.org/10.1016/j.conbuildmat.2007.10.006. [10] V.M. Shinde, J.P. Bhusari, Response of ferrocement confinement on behavior of concrete short column, IOSR J. Mech. Civil Eng. (1989) 24–27. [11] S.U. Khan, S.F.A. Rafeeqi, T. Ayub, Strengthening of RC beams in flexure using ferrocement, IJST Trans. Civil Eng. 37 (2013) 353–365. [12] R. Al-hammoud, K. Soudki, T.H. Topper, Fatigue flexural behavior of corroded reinforced concrete beams repaired with CFRP sheets, J. Compos. Constr. 15 (1) (2011) 42–51, http://dx.doi.org/10.1061/(ASCE)CC.1943-5614.0000144. [13] M.J. Shannag, Bending behavior of ferrocement plates in sodium and magnesium sulfates solutions, Cem. Concr. Compos. 30 (2008) 597–602, http://dx.doi.org/10.1016/j.cemconcomp.2008.03.003. [14] B. Kondraivendhan, B. Pradhan, Effect of ferrocement confinement on behavior of concrete, Constr. Build. Mater. 23 (2009) 1218–1222, http://dx.doi.org/ 10.1016/j.conbuildmat.2008.08.004. [15] S. Ahmad, Techniques for inducing accelerated corrosion of steel in concrete, Arab. J. Sci. Eng. 34 (5) (2009) 95–104. [16] T.A. El Maaddawy, K.A. Soudki, Effectiveness of impressed current technique to simulate corrosion of steel reinforcement, Concrete (2003) 41–47. [17] A. Meda, S. Mostosi, Z. Rinaldi, P. Riva, Experimental evaluation of the corrosion influence on the cyclic behaviour of RC columns, Eng. Struct. 76 (2014) 112–123, http://dx.doi.org/10.1016/j.engstruct.2014.06.043. [18] IS 1489 (Part 1) – 1991 (Reaffirmed 2005), Specification for Portland Pozzolana Cement, Bureau of Indian Standards, New Delhi, India, 1991. [19] IS 383 – 1970 (reaffirmed 2002), Specifications for Coarse and Fine Aggregate From Natural Sources for Concrete, Bureau of Indian Standards, New Delhi, India, 2002. [20] IS 10262-2009, Concrete Mix Proportioning – Guidelines, Bureau of Indian Standards, New Delhi, India, 2009. [21] ASTM G1-03, Standard Practice for Preparing, Cleaning, and Evaluation Corrosion Test Specimens, American Society for Testing and Materials Standard Practice, Philadelphia, Pennsylvania, 2004. [22] IS 456 – 2000 (Reaffirmed 2005), Plain and Reinforced Concrete – Code of Practice, Bureau of Indian Standards, New Delhi, India, 2000.
Contents lists available at ScienceDirect
Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat
Effect of ferrocement jacketing on the flexural behaviour of beams with corroded reinforcements S. Jayasree a,⇑, N. Ganesan b, Ruby Abraham c a
Department of Civil Engineering, MBCET, Trivandrum, Kerala, India Department of Civil Engineering, NIT Calicut, Kerala, India c RIT Kottayam, Kerala, India b
h i g h l i g h t s Influence of cover thickness and strength of concrete on accelerated corrosion was investigated. Effect of corrosion on the flexural behaviour of RC beams was studied. Flexural behaviour of corroded RC beams after retrofitting with ferrocement jacketing was evaluated.
a r t i c l e
i n f o
Article history: Received 30 December 2015 Received in revised form 12 May 2016 Accepted 22 May 2016
Keywords: Corrosion Accelerated corrosion Flexural behaviour Retrofitting Ferrocement jacketing
a b s t r a c t Corrosion of reinforcement is one of the main causes of deterioration of Reinforced Cement Concrete (RCC) structures which affects the load carrying capacity and its durability. Though it is very difficult to completely eliminate the chances of corrosion, suitable retrofitting strategies can be introduced as a measure for the retrofitting of corrosion damaged structures to gain its original strength. The present work deals with the study of degradation of the ultimate load carrying capacity of the flexural members due to corrosion. Twenty-one RCC beams were cast, out of which, three beams were kept as control specimens, and the remaining were subjected to varying levels of corrosion (5, 10 and 15%) so that six specimens are obtained for each level of corrosion. Accelerated corrosion was induced by means of impressed current method. From each level of corrosion, three beams were subjected to loading corresponding to 70% of the ultimate load given by the remaining three specimens corresponding to that level of corrosion. Subsequently these beams were retrofitted by means of a U-wrap ferrocement jacketing. All the beams were tested under two point loading and the strength and behaviour of retrofitted beams were compared with control specimens. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction Corrosion of steel reinforcements is considered to be the reason behind the reduced service-life and failure of RCC structures. Use of pre-rusted steel, exposure to corrosive environment, loading conditions etc are some of the factors which impart corrosion in RCC structures. The effect of corrosion on RCC structures include cracking of the concrete cover, reduction and eventually loss of bond between concrete and corroded reinforcement and reduction of cross-sectional area of reinforcing steel [1]. Basically, the corrosion process has the ability to start in any environmental conditions and the rate of corrosion is determined by the harshness of the environment. Reinforcing steel is normally passive in concrete due to ⇑ Corresponding author. E-mail address: [email protected] (S. Jayasree). http://dx.doi.org/10.1016/j.conbuildmat.2016.05.131 0950-0618/Ó 2016 Elsevier Ltd. All rights reserved.
high alkalinity of the concrete pore solution. However, the penetration of harmful gases like CO into the concrete destroys this inhibitive property of the concrete and leads to corrosion [2]. The effect of corrosion on the load-bearing capacity and mechanical performance of corroded beams, exposed to wetting and drying cycles in a chloride environment under sustained loading without impressed current was also investigated by Yu et al. [3]. The seismic performance of reinforced concrete beams with corrosion induced in the transverse steel reinforcement by means of cyclic loading was studied by Ou and Chen and the results indicated that pitting corrosion increased with an increase in the corrosion level and the beams could sustain a corrosion weight loss of 6% in the hoops and maintain ductile flexural behaviour [4]. Experimental investigation of the behaviour of highly corroded reinforced concrete beams and also a study on the performance of a 27 year old corroded beam, revealed, brittle failure mode for the corroded RC
93
S. Jayasree et al. / Construction and Building Materials 121 (2016) 92–99
beams [5,6]. The influence of simultaneous loading and corrosion of reinforcement on the structural performance of concrete beams was investigated and the results indicated that corrosion or an occasional over-loading or both are likely to cause concrete structures under service loads to collapse suddenly without significant deflection [7]. An experimental investigation on the flexural failure behaviour of RC beams having different weight loss of 0, 5, 10 and 30% due to corrosion was carried out by Okude et.al and it was reported that the rebars were broken at the final stage of loading [8]. With increasing duration of exposure to a corrosive environment, the steel mass loss increases appreciably, which lead to a significant reduction of the tensile ductility of the material [9]. Studies in the past indicated that the bond strength slightly improves at lower levels of corrosion. This is mainly due to the development of Ferrous oxide surrounding the reinforcements which offer a higher frictional force. However as the degree of corrosion increases the bond strength decreases rapidly due to the drastic reduction of diameter of bars and the ineffective portion of corrosion products on the reinforcements. Suitable retrofitting strategies like jacketing using RCC, steel, Fibre-reinforced polymer and ferrocement can be introduced to reinstatement of corrosion damaged structures. The ultimate compressive strength increases with the change in orientation of square mesh from 90° to 45° if ferrocement is used as an external confinement to concrete specimen [10]. Also it was reported that strengthening through cast in situ ferro-mesh layer is the most efficient method [11]. CFRP sheets increased the fatigue capacity of the beams with corroded steel reinforcement beyond that of the control beams with un-corroded steel reinforcement [12,13]. The use of ferrocement as an external confinement to concrete specimens enhances the ultimate concrete compressive strength and failure strains [14]. The impressed current technique has been frequently used to study the effect of reinforcement corrosion on the cracking of concrete cover, bond behaviour, and load-bearing capacity of reinforced concrete structural members [15]. The advantage of this method over other accelerated techniques is the ability to control the rate of corrosion, which usually varies due to changes in the resistivity, oxygen concentration, and temperature. The predicted and actual mass loss values using impressed current method are very close and hence this technique can be used for studies involving accelerated corrosion [16,17]. The present paper deals with 1. The flexural behaviour of RC beams containing varying degrees of corroded reinforcement (5%, 10%, and 15% mass loss of reinforcement), and to compare the results with the control beam (beam with non-corroded reinforcement). 2. The effect of ferrocement jacketing in beams with different levels of corroded reinforcement and to compare the results with the non-retrofitted corroded beam, as well as with the control beam.
Fig. 1. Accelerated corrosion setup (Schematic diagram).
where, m is the mass of steel consumed (g), M is the atomic weight of metal (55.8 g for Fe), I is the current (amperes), t is the time (seconds), z is the ionic charge (2), and F is the Faraday’s constant (96,485 A/sec). 2. Material properties and mix proportion Materials and mix obtained in this investigation are suitable for M20 grade concrete which is generally used in all the RCC structures in the past and these structures are subjected to different degrees of corrosion depending on the environmental condition. As this investigation is deals with the effect of rehabilitation by ferrocement, the flexural members considered in this study were made of M 20 grade concrete. Portland Pozzolana Cement, crushed stones of 20 mm coarse aggregate, manufactured sand passing through sieve of size 4.75 mm and confirming to zone II of IS 383-1970 (reaffirmed 2002) as fine aggregates were used [18,19]. The mix design was done as per IS 10262-2009, to obtain a M20 grade concrete [20]. The mix proportion thus obtained was 1:1.8:2.9. 3. Experimental programme 3.1. Preliminary studies Preliminary studies were conducted in order to determine whether the desired percentage of mass loss is obtained during the time calculated using Faradays’ equation for the corrosion of steel reinforcements. 3.1.1. Determination of actual corrosion and corresponding strength of single bar Bars of length 800 mm and varying diameters 6 mm, 8 mm, and 10 mm (3 samples each), were subjected to different levels of corrosion (5, 10 and 15% of mass loss), using accelerated corrosion technique. For this study the current was fixed as, 8A. The time
1.1. Accelerated corrosion setup Accelerated corrosion was induced by means of impressed current technique. Fig. 1 shows the schematic diagram of accelerated corrosion setup. In order to induce corrosion in the steel rebars, an electrolytic cell with the rebar acting as anode and stainless steel rod as cathode, dipped in 4% NaCl solution acting as electrolyte was used. A constant DC voltage was applied to the electrolytic cell. The time required for achieving different levels of corrosion by mass loss was calculated using Faraday’s Law [15] as per the following Eq. (1).
m¼
M It zF
ð1Þ
Table 1 Time and mass loss for single bare bars. Diameter (mm)
Theoretical mass loss (%)
Time (seconds)
Actual mass loss (%)
10
5 10 15
9585.00 19170.00 28755.00
6.61 9.75 15.03
8
5 10 15
6590.54 13181.31 19771.64
6.02 9.83 14.98
6
5 10 15
3801.90 7603.81 11405.72
5.13 10.03 15.10
94
S. Jayasree et al. / Construction and Building Materials 121 (2016) 92–99
required for each degrees of corrosion level was calculated for different bar sizes and the details are shown in Table 1. In order to compare the theoretical and actual mass loss, the weights of individual bars were measured before the corrosion process. For the study, stainless steel rod of 6 mm diameter was used as cathode, and the reinforcement to be corroded was used as the anode. The electrodes were immersed in a 4% NaCl solution and electric potential was applied between cathode and anode using a DC power source of 10A capacity. The bars were protected with a heat shrink tubing at the ends, for a distance of 200 mm, in order to avoid corrosion at the ends, thereby, providing sufficient grip for the tension test in the UTM. After the corrosion process, the bars were cleaned to free them from rust, as per ASTM G-1 [21] and the specimens were weighed, and the actual mass loss was found out. Fig. 2 shows the experimental setup. The experimental values of actual mass loss and theoretical mass loss were given in Table 1. From Table 1, it may be noted that the values of actual mass loss and theoretical mass loss were almost same for bare bar specimens. Physical appearance of the corroded bars shows a more or less uniform loss in the diameter and general corrosion predominates than pitting corrosion. Tension test was conducted on all the bars and the yield stress, ultimate stress and percentage elongation were calculated and are given in Table 2. It can be observed that a considerable reduction in yield and ultimate stress were imparted due to corrosion. 3.1.2. Determination of actual corrosion in bundle of bars Bars in groups were tested in order to check the suitability of the impressed current method when electric potential is applied to one of the reinforcements, in a bundle of bars which were connected together by stirrups at the two ends. The main reinforcements comprises of two numbers of 10 mm and two numbers of 8 mm diameter bars and were connected by stirrups of diameter 6 mm. The bars were then subjected to accelerated corrosion, in varying levels (5, 10, and 15%). It was noted that each of the bars, including the stirrups suffered almost equal and uniform metal loss throughout the surface. Also the bars were individually weighed before and after the corrosion process. Fig. 3 shows the specimens subjected to corrosion and Table 3 gives actual percentage of mass loss. 3.1.3. Effect of concrete cover on the corrosion of reinforcements The fore mentioned preliminary study was also extended to RCC specimens to understand the effect of concrete resistance to the time calculated using Eq. (1) for corrosion. Specimens of 100 100 500 mm were prepared, as shown in Fig. 4, having two numbers of 8 mm diameter bars at the top and two numbers of 10 mm diameter bars at the bottom with 6 mm diameter
Table 2 Results of the tension test. Diameter of bars (mm)
Corrosion (%)
Yield strength (N/mm2)
Decrease in yield strength (%)
Ultimate strength (N/mm2)
Elongation (%)
10
0 5 10 15
308.13 285.31 256.36 228.89
0.00 7.50 16.90 25.70
451.93 413.12 387.97 315.94
18.20 18.00 11.50 08.30
8
0 5 10 15
410.31 359.02 348.76 307.73
0.00 12.40 14.90 24.90
461.59 410.31 379.53 359.02
21.10 20.60 13.92 09.07
6
0 5 10 15
499.26 445.77 427.94 410.11
0.00 10.80 14.20 17.80
552.75 517.09 481.43 463.59
17.30 16.83 11.40 08.65
Fig. 3. Bare bars in group.
Table 3 Time and mass loss for bare bars in group. Theoretical mass loss (%)
Current (A)
Time (s)
Actual mass loss (%)
5 10 15
8 8 8
15097.10 30194.19 45291.29
4.94 10.22 15.12
Fig. 4. Accelerated corrosion in rebars embedded in concrete specimens.
Fig. 2. Accelerated corrosion on bare bars (experimental setup).
stirrups with a concrete mix of M 20 grade. Before casting initial weight of the individual bars were measured. The specimens were then subjected to accelerated corrosion in order to achieve varying
S. Jayasree et al. / Construction and Building Materials 121 (2016) 92–99
levels of corrosion (5, 10, and 15%) in reinforcements. After the corrosion process, the specimens were demolished and weight of the cleaned corroded reinforcement was taken to find out the actual mass loss and which are shown in Table 4. It may be seen from the Table 4 that the actual mass loss was less when compared to the theoretical mass loss. In order to compensate this, the calculated time was suitably modified by a factor k as given below.
Theoretical mass loss k¼ Actual mass loss
ð2Þ
By modifying Eq. (1), the time required to provide a certain level of mass loss for reinforcements embedded in concrete is given by:
t ðsecÞ ¼
kmzF IM
ð3Þ
The average value of k obtained for specimens subjected to various corrosion levels is also given in Table 4. Since the values are approximately same, an average value of 1.65 was considered for future study. 3.2. Flexural behaviour of RCC beams Flexural behaviour of RCC beams with corroded reinforcements (corroded beams) were compared with that of beams with noncorroded reinforcements (control beams). Also the study was extended to ferrocement retrofitted RCC beams with corroded reinforcements (5%, 10% and 15%) and the behaviour was compared with that of control beam and corresponding non retrofitted corroded beams. 3.2.1. Specimen details Fig. 5 shows the geometry and reinforcement details of the specimens. The overall dimension of the specimens is 100 150 1000 mm. All the beams were designed as under reinforced sections [22]. The reinforcement consists of two numbers of 10 mm diameter bars at the tension face and two numbers 8 mm diameter bars at the compression face. Stirrups consists of 6 mm diameter bars at a spacing of 90 mm. Electrical strain gauges were attached to the reinforcement. A total number of 21 specimens were prepared in steel moulds, using M20 grade concrete. After 24 h of casting, the specimens were demoulded and kept immersed in water for 28 days. Three beams were designated as control specimens (B0). The remaining 18 beams were subjected to varying levels of 5, 10
and 15% of accelerated corrosion. Three beams in each group corresponding to 5, 10 and 15% of corrosion are designated as B5, B10 and B15. They were kept for studying the properties of corroded beams and the remaining specimens were kept for retrofitting and designated as RB5, RB10 and RB15(three in each group). 3.2.2. Accelerated corrosion Specimens were kept in a stainless steel chamber which serves as cathode and one of the 10 mm bar to which a terminal already connected before casting was serving as anode. The electrolytic medium in the corrosion cell was 4% NaCl solution. When the reinforcing steel was used as anode, there were chances of loss of connection between the electrical supply and the rebar due to loss of metal at the terminal. In order to avoid this, internal threading was done on the rebar and the terminal was rigidly connected to the rebar by means of screws as shown in Fig. 6. The end of the rebar, which houses the terminal, was protected by heat shrink tubing as shown in Fig. 7. Fig. 8 shows the electrical connection to the reinforcement. The time required for different corrosion levels were computed using Eq. (2). A constant current at 2.24 A was applied between the terminals. In order to keep the specimens, during corrosion process, masonry tanks of size 200 250 1200 mm were used as shown in Fig. 9. The stainless steel cathode was placed inside the tanks. Small wooden blocks were placed under the beams in order to separate the specimen from cathode to facilitate free flow of electrons through the electrolyte. Constant current was applied to three specimens simultaneously, by means of three different power sources. The setup was left undisturbed, until the time required for a particular level of corrosion was attained. Then the beams were taken out and kept ready for testing in a dry environment. 3.2.3. Testing All the beams were tested under 2 point loading in a Universal Testing Machine of 1000 kN capacity and the experimental setup is
Table 4 Time and mass loss in steel embedded concrete specimen. Theoretical mass loss (%)
Time (s)
Initial weight (Wi)
Final weight (Wf)
Actual corrosion (%)
Value of k
5 10 15
64395.00 124388.80 185582.40
988.03 988.06 981.04
957.56 928.19 892.68
3.08 6.06 9.01
1.62 1.65 1.665
95
Fig. 5. Specimen details.
Fig. 6. Terminals connected to the rebars by screws.
96
S. Jayasree et al. / Construction and Building Materials 121 (2016) 92–99
Fig. 10. Testing of beam.
Fig. 7. Terminal housed in heat shrink tubing.
load increments of 2.5 kN. In addition to this width of cracks was measured using a crack detection microscope of 50 magnification. Three beams from each group were tested to investigate the flexural properties for different degrees of corrosion levels. 3.3. Rehabilitation of specimens using ferrocement jacketing After accelerated corrosion, beams designated as RB5, RB10 and RB15 were preloaded to 70% of their ultimate load capacities, prior to the retrofitting process. Fig. 11 shows the preloaded specimens exhibiting flexural cracks ready for retrofitting. A suitable epoxy was used for developing bond between concrete surface and wire mesh. For ferrocement retrofitting, two layers of wire mesh 6 22 SWG (volume fraction 1.2%) of tensile strength 862.3 N/mm2 were wrapped around the beams, as shown in Fig. 12. The matrix consists of cement sand mortar of 1:2 ratio by weight and the compressive strength of mortar is 36 N/mm2. After retrofitting, the beams were cured under water for 14 days. All the beams were then tested under two point loading in the same way as that of corroded beams.
Fig. 8. Electrical terminal connected reinforcement.
4. Results and discussion 4.1. Behaviour of specimens under load
Fig. 9. Induction of accelerated corrosion to the beams.
shown in Fig. 10. The span of the beam was kept as 900 mm. Two LVDT’s (Linear Variable Differential Transformer) were employed in the compression and tensile regions to measure the deformations. Dial gauges were placed at midpoint of the span and under the loading points in the soffit of the beam. Since electrical strain gauges were damaged during accelerated corrosion the strains could not be monitored. For control specimens the electrical strain gauge readings were taken. All the measurements were made at
During testing, a large number of multiple fine cracks were observed in the control specimens. As the level of corrosion increased, wider cracks were noticed. Cracks observed parallel to main reinforcements in corroded beams were absent in the case of retrofitted beams. Also spalling of cover concrete were observed for the corroded specimens. Crack pattern of retrofitted beams, were almost similar to that of control beams. Figs. 13–16 show the crack patterns of corroded and retrofitted beams. Propagation of cracks and corresponding crack width of specimens for each load interval is shown in Fig. 17. The cracks developed in retrofitted beams, are finer than those in corroded beams. Also large number of closely spaced finer cracks were observed in the retrofitted specimens. For 15% corroded specimens, the maximum deflection was observed at the load point and not in the middle, may be due to the formation of localised corrosion. Moreover, spalling of cover was noticed in corroded beams during testing. 4.2. First crack load and ultimate load The first crack load was obtained from load deflection graph and it is the load point at which the curve deviates from linearity. The ultimate load was obtained from the experimental tests.
S. Jayasree et al. / Construction and Building Materials 121 (2016) 92–99
97
Fig. 11. Pre-loaded specimens exhibiting flexural cracks.
Fig. 12. Wrapping of wire mesh (U-Wrap).
Fig. 14. RB5 compared with B5 and B0.
Fig. 13. Corroded beams with varying levels of corrosion.
Fig. 15. RB10 compared with B10 and B0.
Both values are shown in Table 5. There was a decrease in first crack load as well as ultimate load, as the level of corrosion increased. It can be observed that due to ferrocement jacketing with two layers of wire mesh, the ultimate load carrying capacities were completely restored except for RB15, which had a higher degree of corrosion (15%). This indicates that the number of layers of wire mesh provided may not be adequate and further tests are required with more number of layers to have effective retrofitting.
4.3. Load deflection behaviour Fig. 18 shows the load – deflection behaviour of corroded and retrofitted beams. The curve is linear up to the formation of first crack load and then it became nonlinear with multiple cracks. Due to experimental limitations, the load deflection could be traced only up to 80% of the post peak loading in the descending
98
S. Jayasree et al. / Construction and Building Materials 121 (2016) 92–99
portion of the curve. It may be noted that the corroded specimens show high values of deflection at lower loads, compared to that given by control specimens. In the case of retrofitted beams, it may be noted that the deflections were less, compared to the control specimens. The area under the load deflection curve indicates the energy absorption capacity (EAC) and the values were calculated for all the specimens and are shown in Table 5. It can be seen that the energy absorption capacity of corroded beams were less compared to that of control beams. This property may be due to the loss of ductility of reinforcement due to corrosion. For RB15 specimens, the calculated value of EAC was less when compared with B0. This may be due to the fact that the two layer ferrocement retrofitting could not restore the ultimate load to the level of control beams. The stiffness was obtained from the slope of the tangent drawn to the load deflection curve and are given in Table 5. There was a decrease in stiffness, with an increase in level of corrosion and it may be noted from the table that there is a significant reduction in stiffness as the level of corrosion increases. Also it may be noted from Table 5 that the stiffness of retrofitted specimens were higher than those of control specimens.
Fig. 16. RB15 compared with B15 and B0.
B0
70
RB5
RB10
RB15
4.4. Moment curvature relationship
LOAD (KN)
60
Fig. 19 shows the comparison of moment curvature relationship of corroded and retrofitted beams. The curve is linear up to first crack moment and beyond which, the curve shifts from linearity. When it reaches yield moment, the curves becomes more or less flat. When steel yields, a large increase in curvature occurs with a small change in moment.
50 40 30 20 10 0
0
0.2
0.4
0.6
0.8
1
CRACK WIDTH (MM)
4.5. Ductility index Displacement ductility was calculated as the ratio of the displacement at ultimate load to the displacement at yield load and the values are given in Table 6. The large increase in curvature, before collapse of the beam is an indication of ductile failure of
Fig. 17. Crack width propagation in corroded beams.
Property
B0
B5
B10
B15
RB5
RB10
RB15
First crack load Ultimate load Yield load EAC (kN mm) % variation Stiffness (kN/mm) Stiffness variation Toughness % variation
11.5 57.0 45.0 107.0 1.0 30.0 1.0 38.9 1.0
11.0 52.3 37.0 79.0 26.0 27.2 9.3 28.7 26.2
10.0 44.8 36.5 63.0 41.0 24.3 19.0 26.1 32.9
37.0 31.0 22.6 46.0 59.0 20.6 31.3 24.4 37.4
13.0 66.3 37.0 118.0 10.3 50.6 68.7 70.1 80.0
12.0 62.3 31.0 113.0 5.6 45.4 51.3 69.7 78.9
9.5 52.0 22.6 80.0 25.3 36.0 20.0 66.0 69.5
12
B0
10
B5
8
B10
6
B15
4
RB5
2
RB10
Moment (kNm)
Table 5 Comparison of engineering properties of corroded and retrofitted beams.
0
RB15
0
0.01
0.02
0.03
0.04
0.05
Curvature (M-1) 80
LOAD (kN)
B5
60 50
B10
40
B15
30
RB5
20
RB10
10 0
Fig. 19. Comparison of Moment-Curvature plots.
B0
70
RB15
0
2
4 6 8 DEFLECTION (mm)
10
Fig. 18. Comparison of Load-Deflection plots.
12
Table 6 Comparison of ductility index. Specimen
B0 B5 B10 B15 RB5 RB10 RB15
Displacement ductility
Curvature ductility
Absolute
Relative
Absolute
Relative
2.41 4.32 5.11 7.57 2.10 2.855 3.71
1.00 1.79 2.12 3.10 0.87 1.18 1.54
2.98 5.44 5.90 9.07 3.41 6.13 7.40
1.00 1.83 1.98 3.04 1.15 2.05 2.48
S. Jayasree et al. / Construction and Building Materials 121 (2016) 92–99
beam. Curvature ductility was also calculated as the ratio of curvature at ultimate load to that of curvature at yield load. As the percentage of corrosion is increased, the value of calculated ductility factor increased, and this may be attributed to the large deflections recorded in the corroded beams, due to loss of bond between steel and concrete, and spalling of cover concrete due to corrosion. Table 6 shows the displacement ductility and curvature ductility of both corroded and retrofitted beams. 5. Conclusions Most of the countries in the world have long coastal lines. Structures located along the coastal line are subjected to different degrees or levels of corrosion. Structures located along the coastal line are subjected to different degrees or levels of corrosion. These structures have to be audited to assess the amount of corrosion that have already occurred and suitable rehabilitative measures have to be arrived at. Considering these aspects the present investigation is carried out and the main aim of this investigation is to study the type of rehabilitative procedure to be adopted in the case of structures subjected to partial corrosion. The following conclusions were arrived based on the present experimental investigation. &
&
&
&
&
&
The impressed current technique can be used for accelerated corrosion in concrete specimens by modifying Faradays’ law. As the corrosion rate increases, the ultimate stress and yield stress of rebars were found to decrease considerably. At 15% level of corrosion, width of cracks due to corrosion surpassed the allowable crack width as per IS 456:2000. Mechanical properties such as ultimate load carrying capacity and stiffness were found to decrease as percentage loss of mass due to corrosion in reinforcement increases. As the percentage of corrosion increases, the deflections observed in the corroded specimens were greater than that of control specimens due to loss of bond between steel and concrete. The deflection of the corroded beams were found to reduce considerably due to ferrocement retrofitting. Ferrocement retrofitting with 1.2% volume fraction of mesh reinforcement improved the ultimate load carrying capacity of corroded beams (B5 & B10) significantly higher than that of control specimens. On the other hand, RB15 specimens with same volume fraction of mesh reinforcement improved only 42.5% of ultimate load carrying capacity of B15. This indicates that ferrocement retrofitting with 1.2% volume fraction of mesh reinforcement is inadequate for higher corrosion levels and hence additional layer of mesh reinforcement may be required for restoring the ultimate load carrying capacity.
Acknowledgements The authors would like to thank the Kerala State Council for Science Technology and Environment (KSCSTE) for the financial
99
assistance for the purchase of equipments required for this work and the Mar Baselios College of Engineering and Technology, Trivandrum and National Institute of Technology Calicut for supporting the work. References [1] D.V. Val, L. Chernin, M.G. Stewart, Experimental and numerical investigation of corrosion-induced cover cracking in reinforced concrete structures, J. Struct. Eng. 135 (2009) 376–385. [2] D.V. Val, Deterioration of strength of RC beams due to corrosion and its influence on beam reliability, J. Struct. Eng. (2007) 1297–1306. [3] L. Yu, R. François, V.H. Dang, V. L’Hostis, R. Gagné, Structural performance of RC beams damaged by natural corrosion under sustained loading in a chloride environment, Eng. Struct. 96 (2015) 30–40, http://dx.doi.org/10.1016/j. engstruct.2015.04.001. [4] Y. Ou, H. Chen, Cyclic behavior of reinforced concrete beams with corroded transverse steel reinforcement, J. Struct. Eng. (2014) 1–10, http://dx.doi.org/ 10.1061/(ASCE)ST.1943-541X.0000932. [5] W. Zhu, R. François, D. Coronelli, D. Cleland, Effect of corrosion of reinforcement on the mechanical behaviour of highly corroded RC beams, Eng. Struct. 56 (2013) 544–554, http://dx.doi.org/10.1016/j. engstruct.2013.04.017. [6] V.H. Dang, R. François, Influence of long-term corrosion in chloride environment on mechanical behaviour of RC beam, Eng. Struct. 48 (2013) 558–568, http://dx.doi.org/10.1016/j.engstruct.2012.09.021. [7] Y. Du, M. Cullen, C. Li, Structural performance of RC beams under simultaneous loading and reinforcement corrosion, Constr. Build. Mater. 38 (2013) 472–481, http://dx.doi.org/10.1016/j.conbuildmat.2012.08.010. [8] N. Okude, M. Kunieda, T. Shiotani, Flexural failure behavior of RC beams with rebar corrosion and damage evaluation by acoustic emission, J. Acoustic Emission 27 (2009) 263–271. [9] C.A. Apostolopoulos, V.G. Papadakis, Consequences of steel corrosion on the ductility properties of reinforcement bar, Constr. Build. Mater. 22 (2008) 2316–2324, http://dx.doi.org/10.1016/j.conbuildmat.2007.10.006. [10] V.M. Shinde, J.P. Bhusari, Response of ferrocement confinement on behavior of concrete short column, IOSR J. Mech. Civil Eng. (1989) 24–27. [11] S.U. Khan, S.F.A. Rafeeqi, T. Ayub, Strengthening of RC beams in flexure using ferrocement, IJST Trans. Civil Eng. 37 (2013) 353–365. [12] R. Al-hammoud, K. Soudki, T.H. Topper, Fatigue flexural behavior of corroded reinforced concrete beams repaired with CFRP sheets, J. Compos. Constr. 15 (1) (2011) 42–51, http://dx.doi.org/10.1061/(ASCE)CC.1943-5614.0000144. [13] M.J. Shannag, Bending behavior of ferrocement plates in sodium and magnesium sulfates solutions, Cem. Concr. Compos. 30 (2008) 597–602, http://dx.doi.org/10.1016/j.cemconcomp.2008.03.003. [14] B. Kondraivendhan, B. Pradhan, Effect of ferrocement confinement on behavior of concrete, Constr. Build. Mater. 23 (2009) 1218–1222, http://dx.doi.org/ 10.1016/j.conbuildmat.2008.08.004. [15] S. Ahmad, Techniques for inducing accelerated corrosion of steel in concrete, Arab. J. Sci. Eng. 34 (5) (2009) 95–104. [16] T.A. El Maaddawy, K.A. Soudki, Effectiveness of impressed current technique to simulate corrosion of steel reinforcement, Concrete (2003) 41–47. [17] A. Meda, S. Mostosi, Z. Rinaldi, P. Riva, Experimental evaluation of the corrosion influence on the cyclic behaviour of RC columns, Eng. Struct. 76 (2014) 112–123, http://dx.doi.org/10.1016/j.engstruct.2014.06.043. [18] IS 1489 (Part 1) – 1991 (Reaffirmed 2005), Specification for Portland Pozzolana Cement, Bureau of Indian Standards, New Delhi, India, 1991. [19] IS 383 – 1970 (reaffirmed 2002), Specifications for Coarse and Fine Aggregate From Natural Sources for Concrete, Bureau of Indian Standards, New Delhi, India, 2002. [20] IS 10262-2009, Concrete Mix Proportioning – Guidelines, Bureau of Indian Standards, New Delhi, India, 2009. [21] ASTM G1-03, Standard Practice for Preparing, Cleaning, and Evaluation Corrosion Test Specimens, American Society for Testing and Materials Standard Practice, Philadelphia, Pennsylvania, 2004. [22] IS 456 – 2000 (Reaffirmed 2005), Plain and Reinforced Concrete – Code of Practice, Bureau of Indian Standards, New Delhi, India, 2000.