Duncan, B C; Arranz, E; Crocker, L E; Urquhart, J M (2004) Tests for strength of adhesion. NPL Report. MATC(A)161

Text MATC161.pdf Download (1MB)

Abstract

The design of adhesively bonded structures requires accurate material property data. These data are often best obtained from bulk test specimens. Such data can give an indication of the cohesive strength of the materials but designers also need to consider the strengths of the interfaces between the adhesive and substrates. Information on interface strengths is normally obtained from adhesives joint tests, such as lap shear or T-peel. However, the complexity of stress distributions in such joints leads to difficulties in obtaining quantitative interface strengths that are applicable to other loading configurations.
As part of a DTI funded Measurements for Materials Systems project, Interfacial Adhesion Strength, a number of alternative test methods for adhesion strength – pull-off, profiled butt joint, pull-out and 3-point bend – have been studied as alternative methods for quantifying adhesion strength. These test methods have been evaluated, in experimental and Finite Element (FE) studies, for their ability to quantify the strength of adhesion between adhesive and adherend. Techniques, such as pull-off, pull-out and three-point bend tests, are superficially easy to perform and interpret using analytical formulations. In principle, these tests might be suitable for large scale screening of adhesives and surface treatments. Work is continuing to evaluate the effectiveness of the pull-off and bend methods for durability assessment.
The tests’ suitability for providing design data on adhesion strengths required evaluation. In this report, both analytical interpretations of experimental measurements and FE studies have been undertaken. The results presented indicate that the simple test methods are able to distinguish between ‘good’ and ‘bad’ adhesion but are relatively insensitive to small differences in the quality of the surfaces. Care must be taken when evaluating results. The failure stresses calculated through analytical methods are often considerably lower than those predicted in the FE stress predictions. Therefore, calculated adhesion strengths will be conservative and likely to be far lower than the bulk material strengths. Only in the profiled butt joint test are the analytical average stresses and the FE predicted peak stresses comparable. However, this test is time consuming in both specimen preparation and performance, and requires special alignment fixtures for bonding and testing specimens.
FE analyses suggest that residual stress, generated by cure shrinkage or differential thermal expansion during cooling from elevated cure temperatures, may be a significant factor in interpreting failure.

Item Type:	Report/Guide (NPL Report)
NPL Report No.:	MATC(A)161
Subjects:	Advanced Materials Advanced Materials > Adhesives
Last Modified:	02 Feb 2018 13:16
URI:	http://eprintspublications.npl.co.uk/id/eprint/2885

Actions (login required)

View Item