Modifying CFRP–concrete bond characteristics from pull-out testing

Abstract

Most available models for the prediction of bond characteristics between carbon fibre reinforced polymer (CFRP) composites and concrete are based on data from tests on pull-out specimens, but the use of such characteristics in the analysis and design of reinforced concrete and CFRP strengthened or repaired beams could yield inaccurate estimations. The present work aimed to develop empirical models to generate modification factors for bond characteristics as obtained on small-size pull-out specimens. For this, an experimental programme was designed and conducted to relate bond characteristics from pull-out tests to those from concrete beam bond specimens. CFRP plates or sheets were bonded to both types of specimens at length and width ratios of 1/3, 2/3 and 1. The beams were tested under four-point loading with load measurements acquired against CFRP elongation and its free-end slippage, whereas pull-out specimens were tested for bond stress against free-end slippage. Using statistical modelling, the bond characteristics from both types of specimens were correlated to obtain modification factors in terms of the geometric properties of CFRP composites. The findings indicate that bond length and width ratios have respectively significant and insignificant impact on bond characteristics, regardless of the bond type specimen employed. The modification factors reveal that pull-out specimens tend to overestimate bond strength yet underestimate bond slippage at failure.