< back to main site

Publications

Hyperelastic modelling of flexible adhesives.

Crocker, L E; Duncan, B C; Hughes, R G; Urquhart, J M (1999) Hyperelastic modelling of flexible adhesives. NPL Report. CMMT(A)183

[img] Text
CMMT183.pdf

Download (1MB)

Abstract

Commercial finite element analysis (FEA) software packages are used extensively for the prediction of stress and strain distributions in adhesive joints. The accuracy of any prediction will depend on the validity of the material model employed and the reliability of the test data input into the chosen model. The aim of this work was to carry out an initial investigation into a variety of material models available in the Finite Element package ABAQUS that may be suitable for predicting the behaviour of flexible adhesives. The material models studied were the simple models; Elastic and Elastic-Plastic, the hyperelastic models; Mooney-Rivlin, Ogden and the Hyperfoam model. Three adhesive systems were studied; 3M DP609, Evode M70 and PPGY5000.
Mechanical testing of bulk adhesive samples was carried out to generate material property data, e.g. Young’s modulus, Poisson’s ratio. ABAQUS requires test data under uniaxial strain, biaxial strain and planar strain conditions to fully implement the various material models. ABAQUS uses these data to produce the model coefficients required to run the analyses. These Finite Element (FE) predictions have been compared to experimental lap joint test data. The applicability of these models was investigated for each adhesive. Problems with obtaining sufficiently accurate input data and materials with the same states of cure in the joint and bulk specimens led to poor agreement between the FE predictions and the lap joint tests. The Elastic model and the von Mises Elastic-Plastic models appeared inadequate for characterising the flexible adhesives. The hyperelastic models seemed to offer more promise.
A failure criterion based on the energy to failure has been investigated. Elastic strain energies predicted from the FE analyses have been compared to the energy to break obtained from uniaxial testing of bulk samples. However, the problems with the FE predictions prevent a full evaluation from being made. This work will continue in the extension project PAJex2 where the issues arising from this work will be addressed.

Item Type: Report/Guide (NPL Report)
NPL Report No.: CMMT(A)183
Subjects: Advanced Materials
Advanced Materials > Adhesives
Last Modified: 02 Feb 2018 13:18
URI: http://eprintspublications.npl.co.uk/id/eprint/1158

Actions (login required)

View Item View Item