BOND CHARACTERISTICS OF STRENGTHENED AND RETROFITTED STEEL BY SMART CFRP TECHNIQUE

Transcription

1 BOND CHARACTERISTICS OF STRENGTHENED AND RETROFITTED STEEL BY SMART CFRP TECHNIQUE Chamila Batuwitage 1*, Sabrina Fawzia 1, Xumei Liu 1,David Thambiratnam 1 and Md Iftekharul Alam 1 1 School of Civil Engineering and Built Environment (CEBE), Queensland University of Technology, 2 George Street, Brisbane QLD * c.batuwitage@qut.edu.au ABSTRACT Usage of new smart materials in retrofitting of structures has become popular within last decade. Carbon fiber reinforced polymer (CFRP) has been widely used in retrofitting and strengthening of concrete structures and its usage in metallic structures is still in the developing stage. The variation of mechanical properties of CFRP and the consequent effects on strengthening and retrofitting CFRP systems are yet to be investigated under different loading and environmental conditions. This paper presents the results of CFRP strengthened and retrofitted corroded steel plate double strap joints under tension. An accelerated corrosion cell has been developed to accelerate the corrosion of the steel samples and CFRP strengthened samples. The results show a direct comparison of bond characteristics of CFRP strengthened and retrofitted steel double strap joints. KEYWORDS CFRP, accelerated corrosion, retrofitting, strengthening. INTRODUCTION A large proportion of existing steel infrastructures can suffer deterioration with time due to various effects. The number is increasing as more structures reach the end of their service life, which has been shortened by increased loading, material degradation, and structural fatigue. Galvanic corrosion is one of the prevalent issues with all steel members, and although extensive research has been conducted on the issue and several methods have been developed to reduce or eliminate the effects of corrosion, it remains a prevalent concern. Demolition and reconstruction of such infrastructure may involve a significant cost and structural rehabilitation is becoming the preferred solution. A popular, and increasingly common strengthening technique employed by engineers to rehabilitate structures is carbon fibre reinforced polymer (CFRP) systems. The process has been derived from a traditional retroactive strengthening of steel beams using steel plates. Several studies have been conducted on the performance of fibre composites bonded to concrete surfaces; however a comparatively lesser amount of research has been conducted on the application to steel surfaces. Studies carried out in early stage, have proven that the CFRP systems effectively enhance the structural performance of structures (Vatovac et al. 2002, Jiao et al. 2004,Colombi et al. 2006, Dawood et al. 2006,Photiou et al. 2006,Fawziaet al. 2007).Research was further extended to determine the bond characteristics between CFRP and steel because majority of the failure was observed at the CFRP-steel interface (Fawzia et al. 2005, 2006, Matta et al, 2005).As a result of further investigation, it is known that a major cause of failure in steel to CFRP joints is composite delamination (Nozaka et al. 2005) and design guidelines have been developed to assist the development of methodologies for applying CFRP to steel surfaces (Schnerch et al. 2007, Fernando et al. 2013). All of the above research focused on the performance and effectiveness of CFRP systems.researchers extended their scope to determine durability of CFRP/steel systems. Many studies have been conducted on the environmental durability of CFRP strengthened members. The effects of cold conditions (Kimet al. 2012, Humayunet al. 2015), warm conditions(gamage et al. 2009, Nguyen et al. 2013), freeze-thaw, warm-cold, and wet-dry cycles(smith et al. 2005, Kim et al. 2012, Nguyen et al. 2012), humidity(gamage et al. 2009, Nguyen et al. 2013), UV radiation and fire(smith et al. 2005)and submersion in water(zanni-deffarges and Shanahan 1995, Nguyen et al. 2012), have been extensively studied. Although, it is well known that carbon fibre can facilitate galvanic corrosion by acting as an electrical conductor(tavakkolizadeh and Saadatmanesh 2001), a comparatively lower number of studies have examined the effect of corrosion on joint strength. They investigated the galvanic corrosion between steel and a CFRP 190

2 sheet. It was concluded that galvanic action will only occur when the carbon fibre is in direct contact with the steel, and the galvanic corrosion rate is directly related to the epoxy coating thickness. A study by Nguyen et al. 2012(Nguyen et al. 2012)showed that the performance of steel/cfrp double-strap joints were significantly affected by the ingress of water into the adhesive layer and further suggested that corrosion may present itself between the adhesive and steel interface. One proposed method of resisting galvanic action which has been suggested by a number of sources (Mertz and Gillespie 1996, Gillespie and West 2002, Dawood and Rizkalla 2010)is to place an insulating material such as glass fibre between the carbon fibre sheets and the steel surface. It was found by Dawood and Rizkalla (Dawood and Rizkalla 2010) that the glass fibre sheet did not increase the long term durability of the bond. However, it enhanced the initial bond strength of the specimens. Literature clearly suggests that CFRP/steel systems are vulnerable to galvanic corrosion. Presence of sea water can accelerate the corrosion in steel. However, the behavior of CFRP/steel interface under accelerated corrosion is yet to be investigated. In this research, corrosion cell is developed to accelerate the corrosion of strengthened and retrofitted steel double strap joints specimens. This paper presents the results of an experimental study based on the strengthened and retrofitted CFRP/steel double strap joints subjected to galvanic corrosion. The joint capacity is experimentally determined by applying direct tensile loads and the results are compared and discussed. MATERIAL PROPERTIES The steel had a nominal elastic modulus of 200GPa, and an ultimate strength of 300MPa. Properties of the unidirectional CFRP, primer and the epoxy adhesive as provided by the manufacturer are listed in Table 1. Table 1: Material properties (Manufacturer provided) Property MBrace CF130 (CFRP) MBrace two-part epoxy adhesive MBrace two-part epoxy primer Elastic Modulus (MPa) >700 Tensile Strength (MPa) 3800 >50 >12 Compressive Strength (MPa) - >80 - Fibre Weight (g/m 2 ) Fibre Density (g/cm 3 ) Thickness (mm) Ultimate Strain Glass Transition Temp. ( O C) EXPERIMENTAL PROGRAMME Experimental programme was conducted in two phases. Phase 1 (Class R): exposure to accelerated corrosion before applying CFRP, Phase 2 (Class S): exposure to corrosion after strengthened with CFRP. Also two control specimens were prepared (Class C) for the comparison of results. Specimen labelling and experimental parameters are listed in Table 2. Fourteen steel-cfrp double strap specimens were tested to evaluate the joint capacity. Typical test specimen consists of two 6 mm thick 25 mm wide steel flat bars bond together using 25 mm width CFRP sheet. CFRP-steel joint consist of three layers of CFRP each side. Three CFRP layers were used to increase the joint stiffness. The bond configuration considered in the experimental programme is shown in Figure 1.Bond lengths considered in this experiment were 120 mm and 75 mm. shorter bond length was used to initiate the failure in the corroded end of the joint. To maintain consistency of the test specimens, a standard fabrication process was adopted. The samples were cured under room temperature for 10 days as recommended by the manufacturer to develop its full strength. Important steps in specimen preparation are illustrated in Figure 2. Retrofitted Specimens (Class R) Steel samples were exposed to accelerated corrosion prior to applying CFRP. Corrosion resistive priming layer was applied to prevent the corrosion of un-bonded and longer bond length. Shorter bond length was exposed to accelerated corrosion conditions to obtain failure in the shorter bond length side. After exposed to required period, the samples were removed from corrosion tank. The specimens were sand blasted prior to retrofit with CFRP. All the retrofitted specimens were fabricated simultaneously. Sand blasted steel specimens were cleaned with acetone and then aligned together with epoxy adhesive and left 24 hours. Then priming agent was applied using 191

3 a brush on the steel surface according to the manufacturer s recommendations and allowed it to dry for approximately 1 hour. The CFRP sheets were cut in to required dimensions and cleaned with methylated spirit to remove any residual dust. The adhesive was mixed thoroughly by hand and applied on primed steel surface. Then, the CFRP sheet was placed on and pressed in to the adhesive. A ribbed roller was used to impregnate adhesive properly in to CFRP layer. Another adhesive layer was applied on top of the existing CFRP layer and pressed again using the ribbed roller. Smooth round roller is then used to maintain even adhesive thickness throughout the sample. Second and third CFRP layers were bonded following the same procedure. Strengthened Specimens (Class S) Steel samples were sand blasted and then strengthened with CFRP. All the strengthened specimens were fabricated simultaneously. Sand blasted steel specimens were cleaned with acetone and then aligned together with epoxy adhesive and left 24 hours. Then priming agent was applied using a brush on the steel surface according to the manufacturer s recommendations and allowed it to dry for approximately 1 hour. Applying CFRP layers to the samples was similar to the procedure mentioned above in retrofitted specimens. Corrosion resistive priming layer was applied to prevent the corrosion of un-bonded and longer bond length. Shorter bond length was exposed to accelerated corrosion conditions to obtain failure in the shorter bond length side. After exposed to required period, the samples were removed from corrosion tank. Figure 1: Bond configuration of typical CFRP-steel double strap joint (a) Pairing of two steel parts (b) Applying ribbed roller on CFRP (c) Strengthened specimens Figure 2: Specimen preparation 192

4 Class C Class R Class S Specimen ID Table 2: Specimen matrix Mass loss ratio Exposure based to determine duration current (days) C1 0% C2 0% none R1A 5% 2 R1B 5% 2 R2A 15% 6 R2B 15% 6 R3A 10% 4 R3B 10% 4 S1 10% 4 S2 10% 4 S3 5% 2 S4 15% 6 S5 15% 6 S6 5% 2 Corrosion condition No corrosion Exposure to corrosion before strengthening Exposure to corrosion after strengthening Accelerated Corrosion Cell Setup The corrosion cell was developed to accelerate the corrosion process and shown in Figure 3. 5% NaCl solution was used as the electrolyte. DC power supply was set up and the negative terminal was connected to two stainless steel bars to act as cathodes. Specimens were connected to the positive terminal of the power supply to act as anodes. By inducing a current through the cell, the specimens lose mass in the form of ions to the cathodes through the solution. Each of the specimens was impressed with the same current (170mA) for different exposure durations to induce proportional quantities of mass loss. The current and exposure durations have been calculated by the method proposed in ASTM G (ASTM 2010). This method is based on the Faraday s law using Equation 1. (1) Where t is the time, I is the current, (g/mol) is the molar mass for iron and 96,487 (coulomb) is the Faraday's constant. Figure 3: Corrosion cell setup 193

5 RESULTS AND DISCUSSIONS All the double strap joints were tested under direct tension in an Instron Tension machine at a constant displacement rate of 1mm/min (Figure 4). The tests were carried out until delamination of the carbon fibre layers or a complete failure of strength observed. Figure 4: Test setup Ultimate Load It was observed that all the joints failed by debonding of CFRP from the steel surface. Class R specimens experienced debonding at an average load of 10.52kN, while the class C and class S specimens failed at an average load of 16.73kN and 16.59kN respectively. The ultimate load on each of the specimens is summarised in Table 4, and the load-displacement graphs for each specimen are given in Figures5 to Load (kn) Extension (mm) C1 C2 Figure 5: Load-displacement relationship for control specimens (Class C) 194

6 Load (kn) Load (kn) Extension (mm) R1A R1B R2A R2B R3A R3B Extension (mm) S1 S2 S3 S4 S5 S6 Figure 6: Load-displacement relationship for retrofitted specimens (Class R) Figure7: Load-displacement relationship for strengthened specimens (Class S) Initial slip can be observed in control specimens (Figure 5). Strengthened and retrofitted specimens did not show any slippage during the initial loading (Figures 6 and 7). Failure Mode Summary of each of the specimen s failure bond lengths and the resulting failure modes are presented in Table 3 and the typical failure of double strap joint is shown in Figure 8. All the joint failures were due to bond failure between steel surface and CFRP termed as debonding failure. Table 3: Failure modes of double strap joints ID Mass loss Ultimate load (kn) Bond length Side 1 Side 2 75mm 120mm 75mm 120mm Class C C1 0% Debonded Debonded Debonded - C2 0% Debonded Debonded - Class R R1A 5% Debonded - - Debonded R1B 5% Debonded Debonded - R2A 15% 9.76 Debonded Debonded Debonded - R2B 15% Debonded - Debonded - R3A 10% Debonded - Debonded R3B 10% Debonded - Debonded Class S S1 10% Debonded Debonded Debonded - S2 10% Debonded - Debonded - S3 5% Debonded - Debonded - S4 15% Debonded - Debonded - S5 15% Debonded Debonded Debonded S6 5% Debonded - Debonded Debonded It was observed that for a number of specimens (S2, R1B, and R2B) the failure occurred in two steps, with one side of the bond delaminating while the other maintained strength. Most of the specimens experienced initialisation of delamination process from the shorter bond length where the failure was designed to occur. 195

7 (a) Strengthened specimen (C) Un-corroded CFRP/steel interface of strengthened specimens (b) Retrofitted specimen Figure 8: Typical failure of a double strap joint Discussion Figure 6 illustrates the load vs deflection relationship for strengthened specimens of different corrosion mass loss. Specimen S2 showed the lowest joint capacity among all the strengthened specimens and it can be observed that all the other ultimate load values are in the same range. It is observed that a twisting occurred during the testing of specimen S2 and this might be the cause for lower ultimate load. However, the applied current and exposure duration wasn t able to produce any degradation within CFRP-steel interface. Inspecting the specimens after delamination showed that the class S specimens experienced almost no corrosion mass loss below the carbon fibre layers and underneath the painted steel (Figure 8). This helps to describe the maintained strength of the class S joints, as the bond area appeared comparatively undisturbed. It was observed that there was a significant disparity between the strength of the class C and class R joints; while the class S joints exhibited no significant loss of strength when compared to the class C joints. It was observed that the joints of class R experienced premature failure at around 60% of the original bond strength (as given by the average of class C specimens). Theoretically the ultimate load of retrofitted joints should be in the same range of control specimens. After analysing the geometry of the joint configuration, the reason causing for lower ultimate load can be described as follows. A significant amount of mass loss was noted on the shorter bond length, as designed. The corrosion was observed to be uniform, with very few pits were observed and those which were present had only penetrated to a shallow depth. As a result of the mass loss, when the two lengths of steel were adhered together a small level difference of approximately 1mm was created, with a larger discrepancy being present on the two R2 type specimens (Figure 9). This level difference may cause for premature failure, introducing a stress concentration at the location of the joint. CONCLUSIONS Figure9: Level difference resulting from corrosion mass loss This paper presents the findings of an experimental programme conducted to evaluate the bond behaviour of CFRP strengthened and retrofitted double strap joints under accelerated corrosion. Ultimate joint capacity of 196

8 each specimen was determined under direct tensile load and compared. Based on the results following conclusions can be drawn. 1. The applied accelerated corrosion on the strengthened double strap joints which were related to 5%, 10% and 15% mass loss respectively did not have any effect on the strengthened joint strength. Interface between CFRP and steel was not affected by the applied current and resulted in the same joint strength as control specimens. It may be necessary to increase the galvanic current or the duration to get sufficient ion penetration through CFRP layers to determine the bond characteristics between CFRP and steel under accelerated corrosion environment. 2. Retrofitted joints, which were corroded prior to strengthening, exhibited a significant loss of strength of 37% as an average. This loss of strength can be attributed to the different level of the steel surface to which the carbon fibre was bonded. The reduction in joint strength is due to the level difference of the two steel parts and not due to the corrosion effect. This level difference may lead to stress concentration near the joint hence resulted a joint failure under a lower load. It is recommended to have flat level for retrofitted joints to achieve the ultimate joint capacity in direct tension. 3. Strengthened joints exhibited no noticeable loss of strength, maintaining on average 99% of the original strength. This is mainly due to un-corroded steel surface beneath the CFRP layers. The surface of the steel which was designed to be corroded appeared to have experienced no corrosion under applied direct current and exposure duration considered under this experimental programme. ACKNOWLEDGMENTS The authors gratefully acknowledge the financial support provided by the Queensland University of Technology (QUT) to the first author. Authors specially acknowledge Mr. Mitchell McPherson and Mr. Junaid Hassan who are undergraduate students at QUT for their contribution during the experiments. REFERENCES ASTM G (2010). Standard practice for calculation of corrosion rates and related information from electrochemical measurements. Colombi, P., & Poggi, C. (2006). Strengthening of tensile steel members and bolted joints using adhesively bonded CFRP plates, Construction and Building Materials, 20(1-2), Dawood, M., Sumner, E., Rizkalla, S. H., & Schnerch, D. (2006). Strengthening steel bridges with new high modulus CFRP materials, Paper presented at the Proceedings of the 3rd International Conference on Bridge Maintenance, Safety and Management - Bridge Maintenance, Safety, Management, Life-Cycle Performance and Cost, Dawood, M. and Rizkalla, S. (2010). "Environmental durability of a CFRP system for strengthening steel structures", Construction and Building Materials, 24(9), Fawzia, S., Zhao, X. L., Al-Mahaidi, R., & Rizkalla, S. (2005). Bond characteristics between cfrp and steel plates in double strap joints, Advanced Steel Construction, 1(2), Fawzia, S., Al-Mahaidi, R., & Zhao, X. -. (2006). Experimental and finite element analysis of a double strap joint between steel plates and normal modulus CFRP, Composite Structures, 75(1-4), Fawzia, S., Al-Mahaidi, R., Zhao, X. L., & Rizkalla, S. (2007). Strengthening of circular hollow steel tubular sections using high modulus CFRP sheets, Construction and Building Materials, 21(4), Fernando, D., Teng, J. G., Yu. T and Zhao.X. L. (2013). "Preparation and Characterization of Steel Surfaces for Adhesive Bonding", Journal of Composites for Construction, 17(6), Gamage, J. C. P. H., Al-Mahaidi,R. and Wong,M. B. (2009). "Durability of CFRP-Strengthened concrete members under extreme temperature and humidity", Australian Journal of Structural Engineering, 9(2), 8. Gillespie, J. W. J. and West,T. D. (2002). Enhancement to the bond between advanced composite materials and steel for Bridge Rehabilitation, Masters Thesis, University of Delaware. Humayun, K. M., Fawzia, S., and Gamage, J. C. P. H. (2015). "Durability performance of carbon fibrereinforced polymer strengthened circular hollow steel members under cold weather".australian Journal of Structural Engineering, 15(4), Jiao, H., and Zhao, X. L. (2004). CFRP strengthened butt-welded very high strength (VHS) circular steel tubes. Thin-Walled Structures, 42(7),

9 Kim, Y. J., Hossain,M. and Yoshitake,I. (2012). "Cold region durability of a two-part epoxy adhesive in doublelap shear joints: Experiment and model development".construction and Building Materials, 36, Li, J., Yan,Y.,Zhang,T. and Liang,Z. (2015). "Experimental study of adhesively bonded CFRP joints subjected to tensile loads".international Journal of Adhesion and Adhesives, 57, Matta, F., Karbhari, V. M., and Vitaliani, R. (2005). Tensile response of steel/cfrp adhesive bonds for the rehabilitation of civil structures. Structural Engineering and Mechanics, 20(5), Mertz, D. R. and Gillespie,J. J. W. (1996). Rehabilitation of Steel Bridge Girders through the Application of Advanced Composite Materials. Nguyen, T. C., Bai,Y., Al-Mahaidi,R. and Zhao,X.L. (2012). "Time-dependent behaviour of steel/cfrp double strap joints subjected to combined thermal and mechanical loading".composite Structures, 94(5), Nguyen, T. C., Bai,Y., Al-Mahaidi,R. and Zhao,X.L. (2012). "Durability of steel/cfrp double strap joints exposed to sea water, cyclic temperature and humidity".composite Structures 94(5), Nguyen, T. C., Bai,Y., Al-Mahaidi,R. and Zhao,X.L. (2013). "Curing effects on steel/cfrp double strap joints under combined mechanical load, temperature and humidity".construction and Building Materials, 40, Nozaka, K., Shield,C. K. and J. F. Hajjar (2005). "Effective bond length of carbon-fiber-reinforced polymer strips bonded to fatigued steel bridge I-girders".Journal of Bridge Engineering, 10(2), Schnerch, D., Dawood,M., Rizkalla,S., Sumner,E. and K. Stanford (2006). "Bond Behavior of CFRP Strengthened Steel Structures".Advances in Structural Engineering, 9(6), Schnerch, D., Dawood,M., Rizkalla,S. and Sumner,E. (2007). "Proposed design guidelines for strengthening of steel bridges with FRP materials".construction and Building Materials, 21(5), Smith, S. T., Kaul,R., Ravindrarajah,R. S. and Otoom,O. M. (2005). Durability considerations for FRPstrengthened RC structures in the Australian environment. Proceedings of Australian Structural Engineering Conference, M. S. B. Dockrill. Centre for Built Infrastructure Research, Faculty of Engineering, University of Technology, Sydney, Australia. Tavakkolizadeh, M. and Saadatmanesh,H. (2001). "Galvanic Corrosion of Carbon and Steel in Aggressive Environments".Journal of Composites for Construction, 5(3), Vatovec, M., Kelley, P. L., Brainerd, M. L., & Kivela, J. B. (2002). Post strengthening of steel members with CFRP. Paper presented at the International SAMPE Symposium and Exhibition (Proceedings),, 47 II Zanni-Deffarges, M. P. and Shanahan,M. E. R. (1995). "Diffusion of water into an epoxy adhesive comparison between bulk behaviour and adhesive joints".international Journal of Adhesion and Adhesives, 15(3),