Guidelines for selecting the dimensions of adhesively bonded end-loaded split joints: An approach based on numerical cohesive zone length

• Published in: Engineering fracture mechanics 2018-11, Vol.203, p.250-265
• Main Authors: Toolabi, M., Blackman, B.R.K.
• Format: Article
• Language: English
• Subjects: Adhesion tests; Adhesive bonding; Adhesive joints; Adhesively bonded joints; Bonded joints; Computer simulation; Crack propagation; Crack tips; Cracks; End-loaded split test specimen; Finite element analysis; Finite element method; Fracture mechanics; Fracture toughness; Guidelines; Mathematical analysis; Mode II fracture; Numerical cohesive zone length; Steady state; Substrates
• ISSN: 0013-7944; 1873-7315
• DOI: 10.1016/j.engfracmech.2018.05.019

•Cohesive zone modelling (CZM) is adopted in this work.•The CZM is used to determine the length of fracture process zone (LFPZ).•A user element subroutine is implemented in Abaqus v6.14.•A formulation for LFPZ in an end-loaded split (ELS) joint specimen is proposed.•Results from the expression, correlate well with those from numerical results.•Guidelines are developed for selecting the dimensions of an ELS test specimen. Recent attempts to measure the mode II fracture resistance of structural adhesive joints employing highly toughened adhesives have been hindered by the formation of extensive damage zones ahead of the crack tip. This can lead to strongly rising resistance curves being generated and to the steady-state not being attained. In this paper, a cohesive zone element formulation has been implemented in Abaqus v6.14 via a user element subroutine to determine the length of fracture process zone. It has been demonstrated that the existing analytical expressions cannot predict the length of fracture process zone accurately; hence a formulation for the length of the fracture process zone in an end-loaded split (ELS) joint specimen has been derived using a numerical parametric method. The results obtained from the proposed formulation correlate very well with those from numerical results. The results have allowed guidelines to be developed for selecting the dimensions of an ELS test specimen, comprising fibre-composite substrates bonded with a toughened adhesive, which ensure the attainment of the steady-state and hence a plateau in crack growth resistance curve is ensured. This removes a significant obstacle in the definition of a successful standard method test.