The effects of elevated temperature on the bond behaviour of sand coated GFRP bars to concrete – Definition of bond vs. slip relations

• Autor(a) principal: Rosa, I. C.
• Data de Publicação: 2019
• Outros Autores: Firmo, J. P., Correia, J. R., Barros, Joaquim A. O.
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/1822/71733
• topic: A. Glass fibres; B. Adhesion; B. High-temperature properties; D. Mechanical testing; Glass fibres; Adhesion; High-temperature properties; Mechanical testing; Engenharia e Tecnologia::Engenharia Civil; Science & Technology

The use of glass fibre reinforced polymer (GFRP) bars as internal reinforcement of concrete structures has been growing, mainly due to the advantages they present over steel reinforcement, namely their low weight, high strength and corrosion resistance. However, at moderately elevated temperatures, especially when approaching the glass transition temperature (Tg) of the polymeric matrix (usually between 65 and 150 °C), the stiffness, strength and bond properties of these rebars are known to be significantly degraded. The first part of this paper presents an experimental investigation comprising tensile and pull-out tests on sand coated GFRP rebars at elevated temperatures under steady-state conditions; the tensile tests were carried out up to 300 °C, whereas the pull-out tests were performed up to 140 °C (measured at the GFRP-concrete interface); two embedment lengths of the rebars were considered. The obtained results confirmed that the stiffness and strength of the GFRP-concrete interface are significantly reduced with temperature increase, especially when the Tg of the GFRP rebars is approached and exceeded. In the second part of the paper, analytical bond vs. slip relations for the GFRP-concrete interface are proposed for each of the tested temperatures; the defining parameters of these local laws were calibrated with the experimental data from the pull-out tests. Moreover, the accuracy of two empirical (relaxation) models in predicting the GFRP-concrete bond strength reduction with temperature was also assessed