Rational

Simultaneous detection of two orthogonal components of the surface displacement is possible using two-wave mixing (TWM) combined with multi-channel detection. If ultrasounds are generated through the sample, multi components of the surface displacement can be measured using a single probe beam: the out-of-plane and in-plane components.

Optical Design

In a TWM based system the in-plane information is carried out by the speckles that are scattered away from the specular reflection. An interferometer with a large collecting aperture collects speckles corresponding to different angle of incidence.

An important feature of the Tempo systems is that multiple beams can be independently processed inside the same crystal without running into a cross-talk issue. This feature is used to realize a multiplexed interferometer for simultaneous detection of multiple beams corresponding to the observation at different viewing angles of the same illuminated point. The optical layout of the Tempo 1D is modified, such that the front collection optic is imaged on the detector and the single-element detector is replaced by a linear detector array.

Each element of the detector array corresponds to a small area of the back-scattered light entering through the front collecting optic and thus corresponds to light backscattered along well-defined incidence angles. Processing of the interference signals, as a function of the backscattered angle, simultaneously yields in-plane and out-of-plane displacements. The multi-detector is a linear-array consisting of 16 elements. The collected light is focused on the linear array using a cylindrical lens.

Backscattered light may not be uniformly distributed as it is affected by surface roughness. Therefore, each channel must be normalized before calculating the in-plane and out-of-plane components. This is achieved using automatic gain controlled (AGC) amplifiers which serve to monitor a low frequency calibration signal generated by an internal mirror mounted on a piezo-electric transducer in the path of the reference beam.

Following amplitude normalization, the signals are processed by pairs having the same incidence angles. For each pair, the normalized signals are added to generate the elementary out-of-plane component and their subtraction yields the elementary in-plane component.

However, the contribution in the final in-plane signal of each elementary in-plane component depends on the angle. A weight is thus applied, depending on the angular contribution of the in-plane for the elementary in-plane and for the out-of-plane components.

A through transmission experiment was carried out. Generation was achieved in a thermoelastic regime with a Nd:YAG pulsed laser. The generation laser beam was focused along a line with a cylindrical lens. The detection was carried out on the opposite side of the sample with a bandwidth between 20 kHz and 20 MHz. The sample was a 12.7 mm thick aluminum plate. The Tempo 2D scanned along a 50 mm line. In a first experiment (Figure 1.), we used a sample free of defects (no blind holes), while in the second (Figure 2.), two seperate blind holes were introduced.

The B-scan results for both in-plane and out-of-plane displacements were recorded for both samples (Figure 3. and Figure 4.). The direct P-wave (P) and S-wave (S) arrivals at their respective arrival time 2 μs and 4 μs are clearly visible, along with multiple reflected and reflected/converted waves. As expected, the thermoelastic source generates a stronger S than P-wave. For these measurements, the calibration coefficient is 100mV/nm for both in-plane and out-of-plane outputs. After 12μs, some reflections from the sample edges are also visible on both in-plane and out-of-plane measurements. Moreover, as expected, the in-plane B-scan signal is asymmetrical around the epicenter, whereas the out-of-plane B-scan signal is symmetrical around the epicenter. In the second experiment, where two deffects where introduced, reflection / diffraction from the two defects are visible on both signals, but they are clearly identifiable on the in-plane B-scan.

A through transmission experiment was performed. The sample is a 1 mm thick aluminum sheet. The generation was carried out with a line-focused pulsed laser beam. The measured in-plane direction is along the propagation direction. Figure 3-A illustrates low-frequency Lamb waves detected with the Tempo 2D after propagation along 50 mm. The detection bandwidth is [20kHz – 20MHz]. The signals were low-pass filtered with a cut-off frequency set at 1MHz, in order to only visualize the lower frequencies for the first set of measurements.

Figure 3-B describes the result of this experiment. The Lamb wave mode A0 is clearly visible after 16μs on both graphs. The S0 mode is clearly present on the in-plane signal, but hardly distinguishable on the out-of-plane. The graph enhances the fact that the symmetrical S0 mode has a stronger in-plane component which propagates faster than the asymmetrical A0 mode. Moreover, we can clearly see that the asymmetrical A0 carries more energy than the S0 mode. Finally, the A0 mode presents a phase shift of 90° between the in-plane and out-of-plane components, which justifies its asymmetrical property.

A second set of measurements is shown on Figure 4, with detection at the epicenter. Here the data are high-pass filtered, showing only the frequency above 1MHz.

Strong resonances are detected. Some Lamb modes exhibit an anomalous behavior at frequencies where the group velocity vanishes while the phase velocity remains finite. The zero-group velocity (ZGV) leads to sharp continuous resonance and ringing effect. Figure 4 shows the fast Fourier transform computed on the first 50μs of the signals. The spectrum of the out-of-plane signal (Figure 4-A) and the in-plane signal (Figure 4-B) clearly show the resonance of the S1 mode and of the A2 mode as described by Clorennec et al in their paper Laser Impulse Generation and Interferometer Detection of Zero Group Velocity Lamb Mode Resonance (Appl. Phys. Lett. 89, 2006). The resonance of the A2 Lamb mode corresponds to the thickness shear resonance (F_2.d=3.V_s/2), where VS is the shear wave velocity and d is the thickness.

Measurements were carried out using 45° and 70° angle beam transducers, generating compressional waves with an angle relative to the surface of inspection. The interferometer is positioned in such a way that the outgoing laser beam is normal to the transducer surface of inspection, as shown here.

When propagating through a specimen, the ultrasonic waves carry information about the inner structure. Similarly, when propagating along a surface, the information about the surface quality and surface coatings can be extracted.

Compressional waves can easily go through kissing bonds between two layers. The out-of-plane information therefore becomes insufficient to determine the quality of mechanical bonding. Due to motion perpendicular to the direction of propagation, the coupling of shear waves does not occur if there is no mechanical contact. The in-plane information is therefore used to determine the quality of bonding between two layers