eXtended Finite Element Method for fracture characterization of adhesive joints in pure mode I

• Type: Article in International Scientific Journal
• Date: 2011

Authors

Campilho, RDSG (Author) Other; Banea, MD (Author) FEUP; Chaves, FJP (Author) Other; da Silva, LFM (Author) FEUP

Journal

Title: Computational Materials Science

Vol. 50

Pages: 1543-1549

ISSN: 0927-0256

Publisher: Elsevier

Indexing

ISI Web of Knowledge - 66 Citations

Scopus - 86 Citations

Scientific classification

FOS: Engineering and technology > Materials engineering

Other information

Authenticus ID: P-002-V9R

DOI: 10.1016/j.commatsci.2010.12.012

Abstract (EN): Adhesive-bonding for the unions in structures is gaining momentum over welding, riveting and fastening. It is vital for the design of bonded structures the availability of reliable damage models to predict their fracture behaviour. Cohesive Zone Models (CZM's) have been extensively used in the past, taking advantage of traction-separation laws between stresses and relative displacements to simulate crack growth along predefined paths. The eXtended Finite Element Method (XFEM) is a recent improvement of the Finite Element Method (FEM) that relies on damage laws based on the bulk strength of materials for damage initiation and strain for the assessment of failure, rather than the tipping tractions and tensile/shear relative displacements used for the CZM's. Compared to the FEM, XFEM excels in allowing discontinuities to grow within bulk solids along an arbitrary path. This work aims to assess the viability of the XFEM to predict the fracture behaviour of a thin layer of adhesive between stiff and compliant adherends. To build the XFEM damage laws, the fracture toughness in pure mode I (G(Ic)) and tensile cohesive strength (sigma(0)(n)) of the two adhesives were initially determined by Double-Cantilever Beam (DCB) and bulk tensile tests, respectively. Particular emphasis was given to G(Ic), as this is the most influent parameter for the numerical predictions. The numerical simulations of the DCB tests with the proposed laws matched accurately the experimental load-displacement (P-delta) curves, which validated the analysis procedure. The accuracy of the data reduction methods for calculation of G(Ic) was checked by comparison of the input values in the simulations with the results of G(Ic) calculations, with good results.

Language: English

Type (Professor's evaluation): Scientific

No. of pages: 7