An Ultrasonic Method to Measure Thin Surface Coatings and Films Using A Superimposed Standing Wave

The thickness of a protective surface coating that is left on a substrate after a wear process is an indicator of how much ‘life’ is left in terms of the surface coating functionality. Monitoring and quantifying the surface coating thickness is crucial in the evaluation of the risk of failure and maintenance to acceptable risk levels.
Ultrasound methods can be used to measure a coating/film layer thickness that is out of direct reach. Most pulse-echo ultrasonic methods use a time-of-flight through the layer to determine its thickness. As the layer becomes thinner, the reflected echoes overlap, and deconvolution and resonance methods are used to resolve this in most cases. However, the deconvolution methods are generally complex and time consuming whereas the resonance methods are usually limited by a narrow transducer usable bandwidth in common practice. In this work, the use of an ultrasonic continuously repeated chirp longitudinal wave within a solid was proposed to extend the usable transducer bandwidth and to amplify the effect of a coating. The multiple reflections interfere within the solid to form a superimposed standing wave whose amplitude spectrum is highly sensitive to a coating thickness at the interface of interest.
Epoxy coatings in the thickness range 70μm to 350μm were successfully measured using a standing wave method and showed a good agreement with standalone profilometer measurements. The standing wave measurements showed minimum and maximum percentage change accuracies of 1.04% and 10%, respectively. The epoxy resonances obtained from the standing wave method showed a greater reduction in the resonant mode magnitude than those obtained from a pulse-echo method, signifying increased sensitivity due to the multiple reflections.
A standing wave was then modelled from single frequency wave interference principles to simulate the frequency response of a coating on aluminium substrate using ultrasound propagation equations. The experimental data and mathematical model results showed a strong correlation indicating that the model can be applied to simulate or to compare real coating frequency responses. The model can also be used to optimise the standing wave method.
The thickness of a single epoxy layer bonded on a steel substrate was measured at different temperatures using a standing wave method. A good correlation was observed between experimental results and a linear thermal expansion model for data in the temperature range 24°C to 40°C. Beyond this range up to 70°C the difference between the experimental data and model results increased. Identification of epoxy resonant frequencies became difficult beyond 70°C due to the onset of a glass transition state.
A standing wave method was then applied to investigate the lowest measurable polyimide layer thickness that was bonded on a steel substrate using a standing wave method. The highest and lowest thicknesses measured were 530μm and 42.49μm, respectively. The lowest measurable thickness was limited by the upper bandwidth limit. There was no upper limit on the highest thickness measurable by a standing wave method although for much thicker layers it is more convenient to use a time-of-flight method. The results between the polyimide thickness measured using a standing wave and pulsed wave methods showed a strong agreement. Moreover, the standing wave measurements showed greatest sensitivity in terms of the resonant mode magnitudes when compared with conventional pulsed wave methods. This is practically useful in improving the identification of the coating/film resonant modes. The standing wave method was simpler because it had fewer processing steps involved. The hardware to be used was also relatively cheap.
Finally, a measurement concept was developed that ultrasonically measured both viscosity and film thickness of lubricants and solvents spreading on a surface. Ultrasonic viscometry was performed using shear sensors and an acoustic matching layer that increased the viscosity measurement sensitivity. A standing wave method was used to measure the film thickness. Results showed that the liquid viscosity was proportional to the film thickness measured at the end of the experiment with the least viscous sample having the least film thickness and vice versa. The solvent viscosity was directly proportional to the carbon chain length of the constituents in the mixture. The results also showed a relatively rapid step-change reduction in the film thickness at certain points occurring particularly for solvents that were volatile. This indicated evaporation of the lighter constituents in the mixture. The proposed measurement concept could be optimised to further investigate the performance properties of lubricants and volatile solvents in real engineering components.