Cyclic performance of adhesively bonded joints using the Distinct Element Method: Damage and parametric analysis

Research output: Contribution to journalArticlepeer-review

8 Citations (Scopus)

Abstract

Adhesively bonded joints have been widely used by engineers to solve problems in different industries. Despite recent studies which have helped with the design and application of these joints, there are still several uncertainties that justify empirical implementation in practice. One main aspect is the rigorous understanding of the performance of these joints when subjected to cyclic loading. Although the monotonic performance of adhesively bonded joints is well known, the cyclic behaviour raises several issues. Therefore, a numerical strategy based on the Distinct Element Method (DEM) was implemented in this work to mitigate the lack of knowledge about the cyclic behaviour between two materials adhesively bonded to each other. For that purpose, the single-lap pull-push test was modelled and due to the diversity of the existing cohesive models, four different bond-slip relationships and two unloading paths (with ductile or “cleavage unloading”) were considered. The results suggest that if the bond-slip relationship has an elastic stage, then the interfacial bond degradation is delayed. On the other hand, the interfacial damage of the joints with the number of cycles increases rapidly in those cases where the bond-slip relationships have no elastic stage and, at the same time, the slip accumulation is allowed. In addition, the results of two different tests available in the literature were implemented and fairly reproduced by the DEM.

Original languageEnglish
Article number107468
JournalComposites Part B: Engineering
Volume178
DOIs
Publication statusPublished - 1 Dec 2019

Keywords

  • Bond
  • Cyclic loading
  • Damage
  • Distinct Element Method
  • Numerical simulations

Fingerprint

Dive into the research topics of 'Cyclic performance of adhesively bonded joints using the Distinct Element Method: Damage and parametric analysis'. Together they form a unique fingerprint.

Cite this