Adhesive bonding between CFRP and concrete is affected by both the flexural and tensile strength. The strength and failure mode are related to the following five parameters: (1)mechanical properties of CFRP materials; (2) adhesive
FIG. 4—Flexure-shear failure mode.
strength; (3) concrete strength; (4) concrete surface pretreatment; and (5) effective length of FRP bonding. Currently many different definitions of failure modes exist in CFRP research [23,24]. The research team defined five failure modes in this study listed below and shown in Fig. 4-8.
1. Flexure/shear failure in concrete: Diagonal crack initiated at the end of the CFRP on one end of the specimen; CFRP remains intact and fully attached to the concrete specimen (Fig. 4).
2. Substrate failure: Cohesive failure with rupture surface through concrete paste and aggregate. Concrete remains adhered to CFRP composite (Fig. 5).
3. Adhesive failure: Adhesive failure with rupture surface between CFRP and concrete surface. CFRP failure surface is clean or covered with thin layer of adhesive (Fig. 6).
4. Mixed failure mode: A combination of substrate and adhesive failure (Fig. 7).
5. Composite delamination: CFRP composite splits between laminations. Laminates remain adhered to concrete (Fig. 8).
FIG. 5—Substrate failure mode. (a) Flexure test. (b) Direct tension test.
(a) Flexure test (b) Direct tension
FIG. 6—Adhesive failure mode. (a) Flexure test. (b) Direct tension test.
In this study, Failure Mode 1 is undesirable because it does not identify the adhesive bond strength. The specimen geometry, concrete strength, and CFRP length were selected to ensure Failure Mode 2 through 5 . For three-point bending testing, Failure Mode 2 depends on the concrete strength and bond strength between CFRP composite and adhesive resins, while Failure Modes 3, 4, and 5 depend primarily on the bond strength which is a function of the adhesive strength. An ideal externally bonded composite is stronger than the substrate concrete throughout the life of the structure and minor flaws in the adhesive are not fatal to the system . The heterogeneities from a variety of materials involved in this interface complicate evaluating the bond strength. For externally bonded CFRP applications, the interface between CFRP and concrete substrate is often the weakest zone.
Control specimens were tested for flexural and tensile strength after curing for 14 days in air conditions. Additionally, a number of control specimens were air cured and tested at the conclusion of exposure testing. The investigators designed the specimens to fail in a substrate mode. Conditioned specimens tested later in the program failed by mixed-mode or adhesive failure indicating a loss of strength due to the accelerated aging protocol. A natural log curve was fit to each set of test results based on the 60, 180, 365, and 550 day test results;
(a) Fluxure test (b) Direct tension
FIG. 7—Mixed failure mode (a) Flexure test. (b) Direct tension test.
(a) Flexure test (b) Direct tension
FIG. 8—Composite delamination failure mode. (a) Flexure test. (b) Direct tension test.
this type of curve is consistent with the Arrhenius equation for strength degradation.
The three-point bending test and direct tension test results of CFRP System A versus time are shown in Fig. 9 and Fig. 10, respectively. Strength in flexure tests rapidly decreases during the first two months of exposure at elevated temperatures and then begins to stabilize after twelve months of exposure. Two months of exposure to different elevated temperature results in a reduction of tensile strength of 30 % to 50 % at different temperatures. With the exception of 30 ° C the same trend is observed in direct tension test results. After six months of exposure the 30° C test results decreased rapidly also.
When compared to the bending test results, direct tension tests show a larger variance. Three possible factors are attributed to this increased variation: test time, potential water exposure or damage during coring, and bond flaws
FIG. 9—Failure load ratio versus time for beam tests for CFRP System A.
FIG. 10—Failure load ratio versus time for direct tension tests for CFRP System A.
between adhesive and concrete surface. Direct tension tests were performed two days after the flexural bond tests. An automated coring and testing process may decrease the coefficient of variation of the direct tension test results. Flaws or voids between adhesive and concrete surface will have less of an effect on a flexural test and a larger effect on the direct tension test . Thus, a larger coefficient of variation would be anticipated from a tension test versus flexural tests.
The observed failure modes also change as the strength ratio decreases. Before exposure, a substrate failure is observed for both bending and direct tension tests. The failure modes change into adhesive or mixed-mode failure for bending tests and mixed failure for direct tension tests after exposure times of two months or longer. A 35 % decrease in strength was observed after submersion in water at 30°C and a 55 % strength degradation was observed after submersion at 60 ° C.
Test results of CFRP System B are shown in Fig. 11 and Fig. 12 for bending and direct tension tests, respectively. Exposure to a 60 ° C water bath results in a 35 % degradation of bending strength in the first two months that tends to stabilize after six months exposure to the same conditions. At higher temperatures, the flexural strength degradation is worse and stabilizes at a lower value. Direct tension test results show a similar degradation trend at the same temperature intervals. As observed in System A, test results of direct tension tests have a larger variation than flexural tests. The strength degradation is higher for flexural specimens than direct tension specimens for the same exposure conditions. As with CFRP System A, the failure modes change from substrate shear/interfacial failure mode to interfacial failure mode for flexural tests. Direct tension tests also change from substrate failure to interfacial or mixed failure modes.
Other investigators [8,15,26] report similar losses and suggest this loss can be partially recovered after drying. To evaluate this theory further, five beam specimens of Composite A and B Systems were removed from the 60 ° C water bath tanks at 16 months exposure, and placed in a dry environment for two months. After drying, these beam specimens were tested. Test results are presented by an open circle in Fig. 9 to Fig. 12. Compared to the specimens with
0 100 200 300 400 500 600
Exposure Time (days)
FIG. 12—Failure load ratio versus time for direct tension tests for CFRP System B.
so severe in this epoxy system that a power curve was used to keep strength ratio positive. The flexural beam strength of the plain concrete prisms with the saw cut and without CFRP was approximately 20 % of the control strength. Consequently, once the strength ratio of the CFRP flexural specimens drops below 20 % it is equivalent to zero additional strength from the CFRP. After six months of exposure, CFRP System C specimens at 50 ° C and 60 ° C lost nearly all the flexural strength due to FRP material. Specimens subjected to 30 ° C and 40 ° C exposure degrade 50-65 %. Twelve month exposure results indicate that CFRP System C lost all bond strength at 40 ° C and nearly all bond strength at 30 ° C. Direct tension test results show the same trend as flexural tests. Eighteen month tests were unnecessary since visual deterioration of CFRP material (Fig. 15) confirmed measured test results of 100 % loss of bond strength.
Using the curve fit data for each system, an extrapolation of the strength ratio for periods between one to five years is plotted in Fig. 16. Although direct tension test results are higher, there is more variation in these data. An appropriate strength ratio for System B based on flexural results would be between 0.55 and 0.6. System A would have a knock down factor between 0.35 and 0.4 based on this extrapolation. For practical purposes System C has no remaining strength after exposure.
Five main theories of adhesion contribute to bond strength: mechanical interlock; adsorption; diffusion; chemical bonds; and electrostatic forces . Mechanical interlock occurs when the liquid adhesive penetrates the roughened concrete surface and then solidifies to enhance bond strength by forcing material failure in the adhesive. The irregularities, concrete surface pores, or any rough surfaces improve the bond strength. Water absorption into the adhesive decreases epoxy stiffness and dramatically decreases bond strength between the epoxy matrix and concrete. This leads to a lower Tg, a lower modulus of elasticity, and an increase in the creep rate. All of the above factors accelerate the adhesive bond failure, and the adhesive becomes more sensitive to increases in temperature. Water uptake is a thermodynamic property that depends on the sensitivity of the material to absorption, inherent water content, and relative vapor pressure. Submerging the specimen provides the highest
500 1000 1500
FIG. 16—Extrapolation of data for time periods between one and five years.
vapor pressure and maximum absorption. Elevated temperature accelerates the absorption rate. The absorption has a direct impact on the chemical bond.
Three commercial CFRP external application systems are considered in this study to evaluate strength degradation of CFRP specimens submerged in elevated temperature water baths. The accelerated aging in this study directly results in a rapid strength reduction, but different composite systems exhibit different strength losses. In terms of the test results, the failure load ratio for direct tensile strength is generally higher than that of flexural tests, which is in agreement with other investigators . Although direct tension tests show higher variation than flexural tests, the degradation trend from direct tension tests is generally consistent with those observed in the flexural tests. Test results from the 18 month samples indicate that strength degradation of CFRP external application systems is closely related to water present at the bond line, exposure temperature, and inherent properties of the CFRP composite systems. No CFRP rupture failure or concrete shear failure was observed after 18 months exposure at the highest temperature for CFRP System A and B. For CFRP System C, composite delamination failure occurred, which indicates that the composite of CFRP System C experienced 100 % strength degradation and should not be used when exposed to moisture and elevated temperatures.
Based on the test results and above analysis, the following conclusions are drawn:
• An effective adhesive will develop the strength of the concrete substrate;
• After environmental exposure, bond between adhesive and concrete surface is often the weakest zone for externally applied CFRP;
• Water and temperature are two key variables that affect the bond performance of CFRP;
• Different composite systems exhibit different bond performance when subject to submersion in elevated temperature water baths, and the bond performance appears related to the composition of CFRP and adhesive materials;
• Selection of CFRP composite systems should be based on the specific service environment and mechanical and chemical properties of CFRP and adhesive materials;
• Using elevated temperature water baths as an accelerated aging method is conservative for CFRP composite applications in a wet environment, but may be too harsh for a completely dry environment; and
• Two months of submersion in elevated temperature water baths is a conservative predictor of accelerated aging.
Since the mechanical and chemical composition of commercial CFRP composite systems vary, additional data on the long-term bond performance of CFRP composite systems must be developed experimentally on a product-byproduct basis at various field conditions.
To fully understand accelerated aging of externally bonded CFRP applications without testing each resin individually, researchers must develop an understanding of mechanical and chemical bond properties of the CFRP system and the influence of water content and temperature on bond. Further research may provide a logical categorization of CFRP composite systems based on better defined mechanical and chemical properties of adhesive materials and different environments. Presently the proposed strength reduction factor for durability is the strength ratio of specimens submerged in water at 60 ° C for 60 days, which provides a lower bound durability strength reduction factor for CFRP applications. Investigators are looking into tests at different relative humidity ratios and wet/dry conditioning.