Influence of temperature on concrete beams strengthened in flexure with CFRP

Abstract

The increasingly faster changing demands to existing buildings and ongoing deterioration of buildings and infrastructure have increased the need to strengthen existing structures. One of developments during the last two decades is the use of externally bonded Carbon Fiber Reinforced Polymer (CFRP) reinforcement to strengthen existing concrete structures. Failure of CFRP strengthened concrete structures is generally initiated by debonding of the CFRP reinforcement from the concrete surface. It can be expected that the debonding is affected by temperature, due to the significant difference in the coefficient of thermal expansion between concrete (?c ˜ 10 × 10-6 /°C) and CFRP (?f ˜ -1 × 10-6 /°C in the fiber direction) and due to the change in material properties at elevated temperatures, especially those of the adhesive. So far, only a limited amount of research has been carried out into the effect of temperature on the debonding behavior of externally bonded CFRP. Moreover, the available research has mainly been carried out with small scale test setups, while full scale CFRP strengthened structures could be affected by temperature in a different way. In this research project, the effect of temperature on the CFRP strengthening of concrete structures has been investigated both with small scale bond tests and with full scale beams, strengthened in flexure. Experimental results have been verified by numerical simulations of the tests by means of finite element analyses. First the effect of temperature was investigated with small scale bond tests, for which two different types of test setups were used; the double-lap shear test and the three-point bending test. With both test setups, the capacity of the joint initially increased with increasing temperatures up to the glass transition temperature of the adhesive (Tg = 62°C). Above this temperature, the type of failure changed from cracking in the concrete adjacent to the concrete-adhesive interface, leaving a small layer of concrete remaining attached to the adhesive, to failure exactly in between the concrete and the adhesive. This was accompanied by a significantly reduced, but also scattering bond strength. The results of the numerical simulations confirmed the experimental results and showed that the increasing failure load with increasing temperature, up to the glass transition temperature, was mainly related to the difference in coefficient of thermal expansion between concrete and CFRP. This can be explained with the development of thermal shear stresses that are mainly concentrated at the plate-ends. These shear stresses acted at elevated temperature in the opposite direction as the shear stresses due to loading. Other observed effects of temperature were a reduced Young’s modulus and creep of the adhesive, especially close to and above the glass transition temperature of the adhesive. Both effects caused a decrease in the peak in thermal shear stresses close to the plate-end, but did not have a significant effect on the failure load. Additionally to the small scale bond tests, an experimental test program was set up to investigate the influence of temperature on full scale beams that were strengthened in flexure with externally bonded CFRP reinforcement. Four different beam configurations were investigated, each at three different temperatures, 20°C, 50°C and 70°C. Test results showed that the type of bond failure and the capacity of the beams that were tested at 50°C were not significantly affected by the temperature increase. At 70°C, the type of failure did not change or changed only partly, from failure in the concrete to failure exactly in the concrete-adhesive interface. This can be explained by the temperature cycle that was applied during heating of the beam to 70°C. A temperature cycle increases the glass transition temperature of the adhesive. Hence, the load capacity was not significantly affected at 70°C, except for the beam with a relatively short laminate length. This beam was designed to fail after debonding in the end anchorage zone. It turned out that the beams where failure initiated after cracking exactly at the plate-end or debonding close to the plate-end were more sensitive to the effects of temperature compared to the beams that failed after debonding further away from the plate-end. This can be explained by the fact that most effects of temperature, like the development of thermal stresses and the lower Young’s modulus and creep of the adhesive at elevated temperatures, mainly affect the (shear) stress distributions close to the plate-end, and not significantly further away from the plate-end. Finite element analyses of the full scale tests confirmed the findings of the experiments and were able to simulate the experiments both qualitatively and quantitatively. The performed full scale experiments and nonlinear numerical analyses, that can be regarded to be unique, provided a good insight in the effects of temperature on the strengthening of concrete structures with externally bonded CFRP, but also provided insight in the debonding behavior in general. The most important conclusion of the research is that the influence of temperature can safely be neglected up to about 10°C below the glass transition temperature of the adhesive. CFRP strengthened concrete structures should not be exposed to higher temperatures, as the capacity can suddenly drop above the glass transition temperature. Higher temperatures can be allowed by applying an adhesive with a higher glass transition temperature.