Reliability analysis of adhesively bonded CFRP-to-steel double lap shear joint with thin outer adherends. (15th June 2017)

• Record Type: Journal Article
• Authors: Liu, Meng; Dawood, Mina
• Is Part Of: Construction & building materials. Volume 141(2017)
• Page Start: 52
• Page End: 63
• Publication Date: 2017-06-15
• Subjects: CFRP carbon fiber reinforced polymer -- DLS double lap-shear -- FORM First-Order Reliability Method -- FEA Finite Element Analysis -- MCS Monte-Carlo Simulation -- COV coefficient of variation -- LSE least-square estimation -- A Araldite 420 -- S Sikadur 30 -- TT Tyfo TC -- TS Tyfo S -- MB Mbrace Saturant -- AG angle-grinding -- SB sand-blasting -- WL wet lay-up -- PL pultruded laminate -- SA steel-adhesive interface debonding failure -- C cohesive failure; Adhesive -- Surface preparation -- Double lap-shear -- Model uncertainty -- Reliability -- Bond strength -- Ductility; Building materials -- Periodicals; 624.18
• DOI: 10.1016/j.conbuildmat.2017.02.113
• Languages: English
• ISSNs: 0950-0618

Abstract

Highlights: Compiled database of CFRP/steel bonded double lap joints that failed by debonding. Quantified model uncertainty for the Hart-Smith model for thin outer adherends. Calculated reliability indices, resistance factors for multiple joint configurations. Demonstrated the importance of each joint parameter to the debonding variability. Abstract: This paper presents the details of a reliability-based analysis of bonded double-lap shear (DLS) joints between steel and carbon fiber reinforced polymer (CFRP) composites. A comprehensive database of experimental results of CFRP-to-steel DLS joints is compiled and a probabilistic analysis of the data is conducted. The compiled experimental results are compared with the bond strengths predicted by the Hart-Smith model for thin adherends and the model uncertainty is characterized, for five popular structural epoxy adhesives and two types of surface preparation techniques. Considering the mechanical and geometrical uncertainties of constituent materials, two reliability-based approaches, First-Order Reliability Method (FORM) and Monte-Carlo Simulation (MCS), are used to calculate the resistance factor at a target reliability index of 3.5. It is found that these two approaches agree well and the resistance factor varies with adhesives, surface preparation techniques, and CFRP types. The importance vector of random variables reveals that the adhesive shear ductility is the most influential material property in determining the reliability index of the bonded joints.