STRUCTURAL BEHAVIOUR OF REINFORCED CONCRETE SQUARE COLUMNS DAMAGED BY REBAR CORROSION

Transcription

1 60 Internationa Journa on Design and Manufacturing Technoogies, Vo.2, No.1, Juy 2008 Abstract STRUCTURAL BEHAVIOUR OF REINFORCED CONCRETE SQUARE COLUMNS DAMAGED BY REBAR CORROSION 1 2 Usha Rani M., Subramanian K. 1 Department of Civi Engineering, St.Joseph Coege of Engineering and Technoogy, Dar es Saaam, Tanzania. 2 Department of Civi Engineering, Coimbatore Institute of Technoogy, Coimbatore, India 1 Emai: murrajv2000@yahoo.com Reinforced cement concrete is one of the most versatie materias used in the construction industry wordwide. It is expected that the reinforced concrete structures wi serve for a ong period of time. But there are many instances of premature faiure of R.C. structures raising doubts about its durabiity. Most of these faiures are attributed to the corrosion of reinforcement embedded in concrete. Corrosion causes gradua deterioration of structures and consequenty their expected service ife drasticay gets reduced. In some cases, sudden coapse of concrete structures due to corrosion have been reported. A great variety of concrete structures such as industria structures, hydrauic structures, marine structures, storage tanks, sewage treatment pants, highway bridges, parking garages, buidings etc., experience the damaging effects of corrosion. Corrosion attack of concrete structures takes pace whenever an aggressive medium is present. The ow durabiity of corroding concrete structures has raised great concern because of the cost associated with repacing and maintaining the existing infrastructure and the potentia hazard to the pubic if the probem is not corrected. Key words: Stee Corrosion, Coumn, Axia behaviour, Energy absorption capacity, Weight Loss. I. INTRODUCTION Corrosion of stee embedded in concrete is a compex probem. Understanding the phenomenon is important. This understanding can ony be gained by proper experimentation on concrete structura member. RC coumns, the main structura member of RC structures, usuay sustain axia forces of dead oads and ive oads. When the structures are exposed to corrosive environment, premature faiure of RC coumns due to corrosion eads to utimate coapse of the compete structure. Particuary, with the extensive use of de-icing sat in cod weather regions, key bridge components such as decks and piers are vunerabe to stee reinforcement corrosion. Design inadequacies and substandard detaiing practices in the coumns are often compounded by the effects of reinforcement corrosion owing to carbonation or choride exposure in marine environment. W i t h t h i s background it is essentia to study the infuence of stee corrosion on the strength of an RC coumn in order to ensure its structura performance. In recent years, considerabe research has been focused on the investigation on the behavior of beams, sabs in corrosive environment by using various protective coating and inhibitors by researchers. Reativey ess work has been reported on RC coumns subjected to stee corrosion. Hence the main objective of this research is to investigate the behavior of RC corroded coumns under axia oad. II. REVIEW OF LITERATURE Pantazopouou et a (2001) discussed the structura behaviour of reinforced concrete coumns with various reinforcement configuration subjected to acceerated corrosion condition in the aboratory. Fifty specimens of diameter 150 mm and 300 mm height were subjected to 6V power suppy in the acceerated corrosion process. It was concuded that the oad carrying capacity and axia strain are reduced due to corrosion. Tastani et a (2004) studied the structura behaviour of corrosion-damaged coumns with sub-standard detaiing of reinforcement comprising of 21 prismatic specimens. The specimens were subjected to 6V power suppy in the acceerated corrosion process for a period of 93 days. Their study concuded that the corroded coumns reached ower strain at faiure eve and the atera ties had undergone severe corrosion oss and were entirey ineffective in restraining eastic bucking and decines the axia oad. Morris et a (2004) presented a study which was based on a correation of eectrochemica parameters such as corrosion potentia and current density together with concrete resistivity and choride concentration data. Cyindrica reinforced concrete test specimens were cast with four numbers of 10 mm diameter and exposed to sea shore and immersed in sea water. The performance of four different concrete mix proportioning was investigated over a period of 1000 days of exposure. Their study concuded that when the eectrica resistivity of concrete increases from 2 to 100 kw cm the vaue of choride threshod increases from 0.44 to 2.32% reative to the weight of cement Baim et a (2001) carried out experimenta study on RC beams subjected to simutaneous oad and stee corrosion. Reinforcement was acceerated by impressing a DC current on the main tension stee whie the oad was appied through a static spring-oaded support frame. Their study concuded that depending on the extent of

2 Usha Rani et a : Structura Behaviour of Reinforced Concrete quare corrosion on the mode of faiure, the utimate bending strength of the beams reduces up to 30%. In most of the previous studies, attempts have been made by different authors on corrosion study on reinforced concrete beams and sabs by using various protective coating to stee, inhibitors, service ife prediction mode, Non- destructive tests to study the eectrochemica behaviour, and corrosion monitoring methods, and reativey ess work has been reported on RC coumns subjected to stee corrosion. Hence, there is a ot of scope for studying the effect of corrosion on the strength and ductiity of reinforced concrete coumns. III. EXPERIMENTAL PROGRAMME A. Materias Used and their Properties Ordinary Portand cement (OPC) 53 grade confirming to IS 12269(1987) obtained from a singe source having a specific gravity of 3.15 was used in this study. Locay avaiabe river sand which fas in zone III of IS , and with a specific gravity of 2.55, fineness moduus of 2.73 were used as fine aggregate. The crushed bue granite having a specific gravity of 2.69 were used as coarse aggregate. B. Mix Proportions The concrete mix was proportioned to target a mean strength of 30 Mpa. The procedure for proportioning of ingredients for the mix was based on Indian standards IS: A mix proportion was arrived at to target the above strength. To find the workabiity of fresh concrete, concrete sampe were taken from various batches of mixed concrete and tested in sump cone apparatus. It gave satisfactory sump vaues ranging from 80mm to 150 mm. The design mix proportion is furnished in Tabe 1. Tabe 1. Mix Proportions of the Concrete C. Detais of Test Specimen Totay forty eight coumns were cast, three specimens for each Size. The coumns were reinforced with 4 Nos of 10 mm diameter as main reinforcement and atera ties of 6mm diameter was used. The specimen names composed of three terms is shown in Tabe 2. Tabe 2. Detais of Test Specimen Square Coumn The first term refers to the first size referred as S1. The second term UC as uncorroded specimens and C as Corroded specimen. The third term L refers to Loading condition of corrosion process. The coumns S3UC had a midde test region 750 mm ong and two enarged capitas at their ends. The capita of height 125 mm was provided at both ends of the coumn to distribute the oad eveny. To faciitate connection to the power suppy and haf ce measurement, sma hoe of 3mm diameter was dried in the ongitudina reinforcement. Through this hoe insuated eectrica wire was sodered. In order to prevent the connection from the i effects of choride the sodered portion was covered thoroughy with seaing compound. The wire was hed in pace so that it exited from the top face of the coumn. D. Method ofacceerating Reinforcement Corrosion Gavanostatic method was to acceerate the corrosion of the reinforcement. This method invoved connecting the positive termina of a DC power suppy to the main reinforcement thus forcing it to become anodic. The negative eectrode was then connected to a stainess stee pate paced in the eectroytic soution and the cathodic/anode surface area ratio of 3.7 were adopted. The scae mode coumns were paced in fibre reinforced pastic tank were subjected to a current of 5V.Two third portions of the coumns were immersed in 3 % NaC soution. Fig.1 shows the acceerated corrosion process of the specimen.

3 62 Internationa Journa on Design and Manufacturing Technoogies, Vo.2, No.1, Juy 2008 Fig. 1. Acceerated Corrosion Process E.Sustained Load Corrosion Process A oad of 45kN was appied to the specimens. The oading was exerted by keeping two 2.5 cm thick stee pates one at the top and the other at the bottom of the specimens and the whoe assemby is tightened with threaded rod and nuts which was controed by a torque wrench. The oad decreases as the corrosion proceeds (Creep effects) with increase in time. To maintain the oads constant throughout the duration of the test, continuous monitoring and adjustment of nuts were required. The stee pates, nuts and bots were pated with Zinc coating to avoid the gavanostatic corrosion. F. Instrumentation and Testing Procedure A specimens were oaded unti faiure under axia compression in a testing frame. The top of the coumn specimens were provided with a stee box of coumn size to ensure uniform distribution of the oad on the specimen. Longitudina strains were manuay using DEMEC gauge with 150mm engths having a east count of 0.002mm. Peets were gued either on the concrete surface. Peets were fixed at mid height of the coumn for strain measurements in the ongitudina direction. Strains were noted down for every 100kN increment of oad. Test set up for the coumns is shown in Fig. 2 and 3. Fig. 3. Test Set - up - of Coumn S3UC IV. RESULTS AND DISCUSSION The corrosion activity of coumn specimens are evauated by destructive tests. In the Destructive test the effect of corrosion on utimate oad, axia strain, energy absorption capacity and weight oss of stee of a the specimens were studied. A. Effect of Corrosion on Axia Behaviour The effect of corrosion on the axia oad carrying capacity, energy absorption capacity, and ductiity of reinforced coumns are examined and discussed in detai in the foowing section. The comparisons are presented in Fig. 4 to 6. A the specimens were tested under monotonic oading. The first set incudes S1UC, S1C, S1UC served as contro coumns. A the corroded specimens were subjected to thirty days of corrosion. The contro specimen faied due to shear and this faiure was indicated by incined cracks appearing on the sides of the coumn. Fig. 2. Test Set - up - of Coumn S4UC Fig. 4. Comparison of utimate oad between corroded and uncorroded coumn S series specimens

4 Usha Rani et a : Structura Behaviour of Reinforced Concrete quare Fig. 5. Comparison of axia strain between corroded and uncorroded coumn S series specimens Fig. 6. Comparison of energy absorption capacity between corroded and uncorroded coumn S series specimens The specimen attained the peak oad of 825 kn. The axia stains were measured at eighty percentage of oading and energy absorption capacity was cacuated for the area under the stress-strain curve. Specimen S1C reached an utimate oad of 747 kn which is 10 % ess than SIUC. This can be ascribed to the deterioration of the bond at the stee to concrete interface caused by corrosion and reduction in the cross sectiona area of the reinforcing bars. This is supported by the observation of a strain in about 1000 µ (micro strain) which is 24 % ess than the virgin specimens. The energy absorption capacity is reduced by 15%. The second set specimens incudes S2UC, S2C, S2LUC and S2LC. S2UC attained the maximum oad of 734 kn foowed in sequence by S2LUC, S2C and S2LC. Specimen S2C reached an utimate strength of 642 kn, which is 1.14 times ess than the conventiona coumn. The axia strain decreases by a factor of 1.33 and energy absorption decreases by 1.36 times than the uncorroded coumn. The utimate oad observed in the specimen S2LUC was 653 kn. which is 1.12 times ess than the conventiona specimen. The axia stain and energy absorption are observed as 1.2 times ess than S2UC. Specimen S2LC attained an utimate oad of 481kN which is 25% ess than that of S2C and 27% ess than S2LUC. The axia strain in S2LC decreases by 39% when compared to S2LUC and 32% when compared to S2C. At the same time 42% and 35% variations were observed in the energy absorption capacity. Sustained oad expedited the initiation of reinforcement corrosion in the specimens by cracks in the concrete. These cracks might connect fow paths faciitating water and ion penetration to the surface of reinforcing stee and further acceerate the corrosion of reinforcement. Once the corrosion progresses, it eads to further surface cracking, deamination and debonding, resuting in reduction of the utimate capacity and axia rigidity. The third set of specimens incude S3UC, S3C, S3LUC and S3LC. Specimen S3C sustains an utimate strength of 623 kn, which is 11% ess than the estimated virgin strength of 696 kn. The axia strain and energy absorption capacity decreases by 24% than the conventiona specimen. The utimate oad taken by the specimen S3LUC was 626 kn, which is 10% ess than the conventiona specimen. The axia stain and energy absorption are observed as 14 % and 18% ess than S3UC. Specimen S3LC attained an utimate oad of 488 kn which is 22% ess than that of S3C and S3LUC. Notabe percentage variation was observed in axia strain. The vaue of 27% was noted when compared to S3LUC and 17% when compared to S2C. At the same time 20% and 14 % variations were observed in the energy absorption capacity. Specimen S4C reached an utimate oad of 616 kn which is 8 % ess than the S4UC. The observation of strain is about 745 µ (micro strain) which is 20 % ess than the virgin specimens. The energy absorption capacity is reduced to 9% when compared to S4UC. The contro coumn faied due to bucking and this faiure was indicated by a ump of concrete spaing from the sides of the coumn at approximatey mid height at the utimate oad eve. The specimen attained the peak oad of 672 kn.specimen S5LC reached an utimate oad of 429 kn which is 36 % ess than the S45LUC. The observation of strain is about 245 µ (micro strain) which is 70 % ess than the virgin specimens. The energy absorption capacity is reduced to 83% when compared to S5LUC. B. Stress-Strain Response Typica Stress- Strain curves for uncorroded and corroded specimens are shown in Fig. 7 to 9. Each curve is an average of the stress-strain curves for the three identica specimens with the same eve of corrosion. The axia rigidity of the coumn decreases due to corrosion. The above behaviour can be reated to the deveopment of cracks which may affect the faiure mode and utimate strength of the specimen through the significant oss of bond between concrete and reinforcement.

5 64 Internationa Journa on Design and Manufacturing Technoogies, Vo.2, No.1, Juy 2008 Fig. 7. Axia stress Vs axia strain for S1 and S4 series specimens Fig. 10. Faiure Mode of S2C Specimen Fig. 8. Axia stress Vs axia strain for S2 and S3 series specimens Fig. 11. Faiure Mode of S1UC Specimen Fig. 9. Axia stress Vs axia strain for S5 series specimen C. Faiure Mode of Square Coumn In corroded coumn, the faiure started in the ower corroded portion and extended over the entire specimen and disintegration of the concrete cover and core especiay in the ower haf of the coumns. The spaing of concrete cover was caused due to corrosion cracks and it occurs a over the ength. This premature mode of faiure was accompanied by simutaneous bucking of ongitudina reinforcement. In the case of convention coumn, the fine irreguar cracks were noticed in the surface of the coumns. These cracks get widened with increase in oad eve and the concrete was crushed as the ongitudina reinforcement had yieded. Fig. 10 to 15 show the faiure mode. Fig. 12. Faiure Mode of S3C Specimen Fig. 13. Faiure Mode of S2UC Specimen

6 Usha Rani et a : Structura Behaviour of Reinforced Concrete quare Fig. 14. Faiure Mode of S4C Specimen Fig. 15. Faiure Mode of S3UC Specimen D. Measurement of Weight Loss The oss of weight of the rebars were cacuated by the gravimetric method. Fig. 16 shows the percentage weight oss of the S series specimens. Fig. 17 shows the corroded rebars. Fig. 16. Comparison of weight oss - S Series Specimens Fig. 17. Corroded rebars V. CONCLUSION The foowing concusions are arrived at based on this study: The oad carrying capacity, ductiity and energy absorption capacity of the conventiona coumn increases with the increase in the cross section of the coumn (S1UC and S2UC). The oad carrying capacity, ductiity and energy absorption capacity of the conventiona short (S1UC) coumn increases with respect to ong coumn (S4UC) of same cross section. The percentage decrease in oad carrying capacity, ductiity and energy absorption capacity of the corroded coumn decreases with the increase in the cover concrete (S1C and S2C). The percentage decrease in oad carrying capacity, ductiity and energy absorption capacity of the corroded short coumn increases when compared to the percentage decrease in ong coumn of the same cover (S1C and S4C). The percentage oss of weight in stee increases with decrease in cover and the percentage decrease in oad carrying capacity of the corroded coumn increases with increase in percentage weight oss of stee. The percentage oss of weight in stee and oad carrying capacity of the corroded coumn increases with decrease in surface areas of stee embedded. The sustained oad and acceerated corrosion procedure

7 66 Internationa Journa on Design and Manufacturing Technoogies, Vo.2, No.1, Juy 2008 REFERENCES [1] ASTM G1-67 (1971), 'Recommended Practice for Preparing Ceaning and Evauating Corrosion Test Specimens', American Society for Testing and Materias, Pennsyvania. [2] Baim Y., Reid J.C. and Kemp A.R. (2001), 'Defection of RC Beams under Simutaneous Load and Stee Corrosion', Magazine of Concrete Research, Vo.53, No.3, pp [3] IS: , 'Specifications for Coarse and Fine Aggregates from Natura Sources for Concrete', Bureau of Indian Standards, New Dehi. [4] IS: , Part III, 'Indian Standard Methods of Test for Aggregates for Concrete', Bureau of Indian Standards, New Dehi. [5] IS: , 'Pain and Reinforced Concrete - Code of Practice', Bureau of Indian Standards, New Dehi. [6] IS: , 'Methods of Tests for Strength of Concrete', Bureau of Indian Standards, New Dehi. [7] IS: , 'Recommended Guideine for Concrete Mix Design', Bureau of Indian Standards, New Dehi. [8] IS: , 'Specifications for 53 Grade Ordinary Portand Cement, Bureau of Indian Standards, New Dehi. [9] Morris W., Vico A. and Vazquez M., (2004), 'Choride Induced Corrosion of Reinforcing Stee Evauated by Concrete Resistivity Measurements' Eectrochimica Acta, Vo.49, No. 25, pp [10] Pantazopouou S.J., Bonacci J.F., Sheikh S., Thomas M.D.A. and Hearn N., (2001), 'Repair of Corrosion - Damaged Coumns with FRP Wraps', Journa of Composites for Construction, Vo.5, No.1, pp [11] Tastani S.P. and Pantazopouou S.J. (2004), 'Experimenta Evauation of FRP Jackets in Upgrading RC Corroded Coumns with Substandard Detaiing', Engineering Structures, Vo.26, No. 6, pp Ms. M. Usha Rani is a ecturer in the Department of Civi Engineering at St.Joseph Coege of Engineering and Technoogy, Dar es Saaam,Tanzania and submitted her Ph.D Thesis to Anna University Chennai. She is a postgraduate from Coimbatore Institute of Technoogy, Taminadu.