Post-graduate student Iermoliuk R.S.

Scientific supervisor Yerysh L.A.

Donetsk national university of economics and trade

named after M. Tugan-Baranovsky, Ukraine

The use of acoustic emission for testing

coatings properties

The use of acoustic emission to detect cracks in engineering structures under stress, such as North Sea oil platforms, high-pressure, vessels, aircraft wings, etc., has been commonplace for a number of years.

However, it is a novel application to study the acoustic emission of coatings under stress. This technique has proved to be of considerable use in monitoring and even, in some cases, predicting the durability of coatings during environmental testing, such as natural weathering, accelerated weathering, salt spray corrosion tests, etc.

In addition, it has proved very useful in evaluating the effects of formulation variables on the ultimate mechanical properties of coatings, and in evaluating these properties for the individual layers of a coating system, as well as elucidating the ways in which these properties interact in producing the total properties of the full system.

The technique is, in principle, extremely simple. Any sudden microscopic movement in a body, e.g. crack formation and propagation, may give rise to acoustic emission. For example, strain is concentrated at the growing tip of a crack. As this crack propagates, strain energy is released in two main forms: as thermal energy and as acoustic energy. The acoustic energy radiates as a deformation wave from the source, and is refracted and reflected by solid inclusions and interfaces until it reaches the surface of the body. Here the surface waves may be detected by sensitive detectors: usually piezoelectric or capacitive transducers. The amplified signal from the transducer is then analysed.

Familiar examples of acoustic emission, at frequencies and intensities audible to the human ear, are the cracking of ice on a pond or the creaking of the treads of wooden stairs under the weight of a human body.

The paint is coated onto one side of a metal foil strip, and this is then inserted in the jaws of a tensile tester; the transducer is attached and the sample stretched.

The noise emitted is analysed, and some noise characteristic is plotted as a function of total strain.

Although tensile testing is more usual, there is no reason why bending or any other form of deformation may not be used. The only essential is that there should be no spurious noise generated by slippage between the specimen and the instrument’s clamps. This is why, in practice, slow deformation rates are used. Apart from this source of noise, there is no need to shield the apparatus acoustically, narrow-band resonant piezoelectric devices (resonant frequency around 150 kHz) to be satisfactory for this purpose.

Good acoustic coupling between the transducer surface and the specimen is essential to maximize detector sensitivity: this is readily achieved by means of a thin connecting layer of silicone grease.

The methods of analysis available for characterizing the acoustic emission are numerous. Because of the simultaneous occurrence of many noise sources, often of different kinds, as well as the modification of the waveforms both by propagation through the body of the specimen and by the response characteristics of the detector itself, it is very difficult with acoustic emission from paint specimens to analyse the complex signal forms to obtain information about the original signal source.

There is also too little theory or experimental work with ‘model’ systems relating waveform characteristics to source mechanism. Thus complicated frequency analysis or amplitude analysis techniques are not generally useful, although amplitude analysis can be revealing if the failure mechanism changes drastically, for example, if there is a change from micro- to large-scale cracking at a particular strain value.

The manufacturers have concentrated on using simple analysis techniques, such as plots of ‘ring-down’ or event counts against total strain to characterize the coating, and have then used these on a comparative basis, for example, to monitor changes in the failure properties of coatings during weathering, to study the effects of changes in formulation or chemical structure on the ultimate mechanical properties of the coatings, etc. This simple utilization of the technique has proved extremely useful.

If the amplified voltage output of the transducer, corresponding to a single event, is idealized as a sinusoidal decay curve, then in ring-down counting, one count is registered every time the voltage rises above a threshold voltage (this is imposed to stop random electrical noise affecting the analysis). In event counting, a preset delay is imposed after the first count before another can be registered. By proper choice of the delay time, comparison of ring-down and event counts will give a crude estimate of the average amplitude of the signals. A cumulative plot of total counts against strain is then produced for each.

The use of acoustic emission for testing coatings properties is the perspective development in coating industry.

References:

1. Paints, coatings and solvents. Dieter Stoye; Werner Freitag (ed.). - 2. completely rev. ed. - Weinheim; Wiley-VCH. 1998.

2. Paint and surface Coatings. Theory and practice. Second edition. Editors: R. Lambourne and T. A. Strivens. Cambridge, England. 1999.