Dynamic fracture mechanics of FRP-concrete interface

External bonding of fibre reinforced polymer (FRP) has been widely used as an effective technique for strengthening reinforced concrete (RC) structures in quasi-static and dynamic scenarios. To strengthen RC structures, FRP is usually made by several layers and bonded to the tension zone subjected t...

Full description

Saved in:
Bibliographic Details
Main Author: Li, Gen
Other Authors: Fung Tat Ching
Format: Thesis-Doctor of Philosophy
Language:English
Published: Nanyang Technological University 2020
Subjects:
Online Access:https://hdl.handle.net/10356/144307
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:External bonding of fibre reinforced polymer (FRP) has been widely used as an effective technique for strengthening reinforced concrete (RC) structures in quasi-static and dynamic scenarios. To strengthen RC structures, FRP is usually made by several layers and bonded to the tension zone subjected to a tension load. Due to the inconsistent deformation in concrete and FRP laminate, relative slip and corresponding shear stress is created between the concrete and FRP and the bond of the interface eventually fails by shear debonding. However, tensile fracture failure is also noticed in many tests. Although the quasi-static shear bond-slip response has already been extensively investigated to aid the FRP strengthening design, studies on their dynamic behaviour are rather limited, especially those under high loading rates. To shed light on the FRP bond-slip behaviour under high loading rates (above 800 mm/s), the author proposed a novel experimental method for high loading-rate impact tests using a modified Split Hopkinson Pressure Bar (SHPB) set-up. Dynamic enhancing effect on ultimate load, shear bond stress and shear fracture energy could be quantified for a wide range of loading rates varying from 0.02 mm/s to 2150 mm/s. The effect of parameters on the bond behaviour was better understood from a detailed parametric study. A novel continuum damage model was proposed in this research to study the dynamic fracture mechanics of FRP-concrete interface in shear debonding. Model analyses and regression test data were conducted to formulate the relationship between model coefficients and material parameters of tested specimens in literature. It was found that the dynamic enhancing effect on interfacial bond properties became more significant when a specimen has a larger FRP load capacity or a lower concrete load capacity. The proposed damage model was verified through comparing numerical predictions of load-displacement curve and FRP strain distribution along the bond length with published test results. Apart from the studies on the shear debonding, this research consisted of two sets of experiments to study tensile fracture of FRP-concrete interface and to obtain the principal interfacial properties of such fracture. Direct tension tests were used to determine the tensile bond strength and notched three-point bending tests were used to determine the tensile fracture energy. Digital image correlation measurements were used to characterise the fracture process. It was found that the two bond properties were close to those of plain concrete in the quasi-static regime and showed significant dynamic enhancing effect at a loading rate of 20 mm/s. Dynamic increase factor (DIF) equations for the two bond properties are provided to predict the interfacial response of FRP strengthened RC structures under dynamic loads. Through previous experimental investigations, it was found that FRP-concrete bond interface contains diverse traction-separation performance in mode I and mode II fracture. To continuously describe the fracture behaviour from mode I to mode II under a dynamic loading, current study proposed a coupled dynamic cohesive zone model (CZM) to analyse FRP-concrete mixed-mode separation. The model was applied in finite element analysis (FEA) of various dynamic tests with FRP-concrete interface bond, e.g. single-lap shear tests, three-point bending tests, FRP strengthened RC beams subjected to dynamic loading and an FRP strengthened RC wall under blast load. From the comparison with test results, the model was shown to be reliable and accurate in simulating the behaviour of FRP-concrete mixed-mode separation.