Non-Destructive Measurement of Coating Adhesion - Project Update

Workshop and Mini-Symposium

Loughborough University, UK, 1 May 2001 - This Workshop/Symposium reported the findings of a three year study sponsored by the DTI under the ‘Materials Metrology Programme’ and supported by manufacturers and users of coatings to investigate non-destructive methods.

The workshop /symposium programme can be downloaded as an Acrobat (pdf) file.

Note: The following summary is provided for dissemination purposes to those who may have an interest in the work but not be directly involved in the projects. The summary contains references to project reports containing full experimental details, access to which is restricted to Industrial Advisory Group Members.

Project Summary

Adhesion is directly or indirectly implicated in many types of coating failure. Examples include blistering, slow delamination as in flaking and peeling, impact damage and corrosion. The latter is a good example of the difficulty in separating cause and effect. The jury is still out as to whether corrosion is the cause of adhesion failure, or alternatively is the first stage of corrosion a loss of adhesion under wet conditions?

In dealing with adhesion problems there have always been difficulties with terminology. Distinctions must be made between the fundamental aspects of adhesion at the molecular level, where forces of attraction can include hydrogen bonding, donor/acceptor interactions, dispersion and electrostatic forces, and thermodynamic aspects such as wetting and surface energy. At the technological level adhesion is usually appraised by destructive tests which measure some aspect of the force needed to separate two surfaces. Such tests have great difficulty in normalising the results since adhesive performance is modified by other factors including the mechanical properties of the coating system. Even a change in film thickness can be sufficient to alter the apparent locus of failure from adhesive to cohesive. Fracture mechanics is one way to address this problem, but at least 300 different types of destructive test have been listed. Since these will be in arbitrary or different units, comparisons can be very difficult.

Coatings differ from adhesives in several important ways. From the point of view of testing there is no convenient ‘handle' by which to detach the coating, and the act of attaching a suitable handle, e.g. by gluing on a stud, will itself alter the result. Coating behaviour is often governed by a strain, rather than stress regime. Further more the adhesive performance of coatings is not a fixed parameter but varies continually in response to service conditions such as temperature and humidity.

If adhesive performance is to be defined by force or work of separation, then ‘non-destructive adhesion testing' seems a contradiction of terms! However there are other possibilities for predicting performance which focus on what might be called ‘mediating factors' these will include various mechanical properties or possibly detecting some change such as the ingress of moisture at an interface. This broader interpretation of NDAT is consistent with other established ND techniques such as the use of ultrasonics to detect flaws in metal castings, an absence of flaws is then equated with good performance. On this basis there are a number of possibilities for predicting the adhesive performance of coatings from ND tests. Techniques include thermography, laser shearography, impedance spectroscopy, scanning acoustic microscopy, ultrasonics and acoustic emission.

In 1998 an industrial advisory group (IAG) of coatings users and manufacturers defined more clearly the types of problem where adhesion is involved, and the benefits of non-destructive testing. These were been analysed from the perspectives of quality control, product development, site testing and service life prediction. Each has different needs. Some techniques are non-destructive but not portable and hence might be suitable for laboratory work but not site inspection. The IAG cooperated in producing test panels exemplifying different levels of performance for various coating types. These were used for a preliminary evaluation of available techniques. Some differences were detected using laser shearography and thermography and were followed up using more fully characterised test panels. PRA also purchased acoustic emission apparatus can be used in a non-destructive mode by demonstrating an absence of damage at defined stress levels, A-E is also a useful bridge between destructive and non-destructive tests.

Project Objectives

  1. To review existing and novel non-destructive methods for testing adhesive bonds, consider their applicability to the different requirements of coatings systems, and their capability to predict performance.
  2. To evaluate the most promising non-destructive techniques using a range of coated systems at contrived upper and lower levels of performance when assessed by destructive methods. Analytical techniques will be used to characterise surfaces and detect locus of failure.
  3. To verify the predictive potential of one or more short listed methods under laboratory and/or field conditions using production samples.
  4. To prepare the outline for a draft Code of Practice on NDT methods for coatings.

The project was formally approved in September 1998 and completed in February 2001. A final report is available to members and other interested parties.

Project Conclusion

‘True’ adhesion is not readily measured non-destructively, there are however a number of techniques which can appraise factors which will either influence adhesive performance or be influenced by differences in adherence. There are no universal techniques for appraising the adhesion of coatings non-destructively and it is useful to draw a distinction between:-

Methods for detecting an existing failure (such as corrosion, blistering or localised delamination).

Methods of predicting future failure.

The former situation is more akin to established N-D techniques and there are a number of viable possibilities which must be considered on a case by case basis. The second requirement is much more difficult and is only likely to be met by relatively few techniques. A further distinction must be made between methods that can be applied under field conditions, and those that are only practicable in the laboratory. All of the methods studied need care in interpretation and would require considerable method development.

A promising technique of general applicability for predictive testing under both field and laboratory conditions is that of Electronic Speckle Pattern Shearing Interferometry (ESPSI) which effectively maps strain distribution on the surface of any coated object. Both ESPSI and Pulse Thermography can be used to detect existing failure even when this is not visible on the surface. Acoustic Emission techniques are not strictly speaking non-destructive but under specific conditions might be used as the basis of a ‘pass’ criteria in some types of inspection. Further work in this area is planned (from Autumn 2004).

For further details contact Jon Graystone or Richard Holman.