Laser based Ultrasonic Testing

Christopher Kuhlmey, winter semester 2017/18

Laser-based ultrasonic testing is a non-destructive, non-contact material testing method, which allows for the measurement of various material properties such as Young’s modulus and density properties and the detection of discontinuities like cracks, delamination, voids, etc. Furthermore, it is possible to determine geometrical properties like the thickness or the layer structure of a specimen. Laser-based ultrasonics, which is closely related to conventional non-destructive ultrasonic testing methods (see also: Ultraschall-Durchschallung,Ultrasonic Pulse-Echo Method,Ultraschall-Schweißnahtprüfung), was developed due to some disadvantages of said techniques.

Motivation

Conventional ultrasonic testing techniques use piezoelectric transducers coupled to the surface of a structure to both generate ultrasonic elastic waves in the material and detect the ultrasonic waves reflected from geometrical or material discontinuities inside the structure. One major shortcoming of this technique is that the transducers need to be properly coupled to the material to ensure optimal energy transmission towards and from the material to enable a certain level of sensitivity. Since this method requires a fluid coupling between the transducer and the surface for most applications and the piezoelectric transducers cease to function when a certain temperature (Curie temperature) is exceeded, conventional ultrasonics is not suitable for elevated temperatures or harsh environments. The couplant also needs to be applied before the measurement, hence conventional ultrasonic testing doesn’t lend itself for the fast scanning of large areas. Since transducers emit and receive vibrations from their whole bottom surface and are thus required to be normally orientated towards the structure with only a few degrees of tolerance, inspection of curved surfaces may be problematic.^[1] The typical size of the transducers ranges from roughly 0,65 cm to 2,5 cm, which can be disadvantageous for the placement on curved, complex geometries. Moreover, every transducer attached to the structure generates a surface load and can therefore have an influence on the measured results.^[2]

Fundamentals

Ultrasonic Waves

The basic principle of ultrasonic testing methods is the investigation of the propagation of ultrasonic elastic waves throughout a medium, induced to its surface (see also: Elastic waves, Body and surface waves). Longitudinal (also compressional or P-wave) and shear waves (also transverse or S-wave) are the two types of bulk waves that can propagate through a volume, usually in a spherical-symmetric manner away from the source. Due to interaction of these two waves near the surface, two types of surface waves develop, namely Rayleigh- and Love-waves, although the latter only exists in layered mediums. The characteristic property of P-waves is the oscillation of particles in direction of wave propagation, while a particle oscillation perpendicular to the direction of propagation is specific for S-waves.^[3] The relationship between a wave’s frequency $//<![CDATA[ \begin{array}{l}f[Hz]\end{array} //]]>$, its wavelength $//<![CDATA[ \begin{array}{l}\lambda [m]\end{array} //]]>$ and the acoustic velocity $//<![CDATA[ \begin{array}{l}v[m]\end{array} //]]>$ within a specific medium is connected by the following equation:^[3]

$//<![CDATA[ \begin{array}{l}f=\frac{v}{\lambda}\qquad\end{array} //]]>$

Moreover, the respective wave velocities are associated with the elastic parameters of a material, and therefore the acoustic velocities of P-, S-waves and Rayleigh in a homogeneous isotropic infinite medium can be described as:

$//<![CDATA[ \begin{array}{l}v_p=\sqrt{\frac{E(1-\sigma)}{\rho(1+\sigma)(1-2\sigma)}}\end{array} //]]>$

$//<![CDATA[ \begin{array}{l}v_s=\sqrt{\frac{E}{2\rho(1+2\sigma)}}\end{array} //]]>$

$//<![CDATA[ \begin{array}{l}v_R=\frac{0,87+1,12\sigma}{1+\sigma}\sqrt{\frac{E}{2\rho(1+2\sigma)}}\end{array} //]]>$

Usually:

$//<![CDATA[ \begin{array}{l}v_p > v_s > v_r\end{array} //]]>$

With:

$//<![CDATA[ \begin{array}{l}v_p\end{array} //]]>$...acoustic velocity of a P-wave $//<![CDATA[ \begin{array}{l}[m/s]\end{array} //]]>$

$//<![CDATA[ \begin{array}{l}v_s\end{array} //]]>$...acoustic velocity of a S-wave $//<![CDATA[ \begin{array}{l}[m/s]\end{array} //]]>$

$//<![CDATA[ \begin{array}{l}v_r\end{array} //]]>$...acoustic velocity of a Rayleigh wave $//<![CDATA[ \begin{array}{l}[m/s]\end{array} //]]>$

$//<![CDATA[ \begin{array}{l}E\end{array} //]]>$...Young's modulus $//<![CDATA[ \begin{array}{l}[N/m^2]\end{array} //]]>$

$//<![CDATA[ \begin{array}{l}\sigma\end{array} //]]>$...Poisson ratio $//<![CDATA[ \begin{array}{l}[-]\end{array} //]]>$

$//<![CDATA[ \begin{array}{l}\rho\end{array} //]]>$...density $//<![CDATA[ \begin{array}{l}[kg/m^3]\end{array} //]]>$

$//<![CDATA[ \begin{array}{l}\lambda\end{array} //]]>$...wavelength $//<![CDATA[ \begin{array}{l}[m]\end{array} //]]>$

It is evident, that propagation mechanics of bulk waves are closely connected to the elastic parameters of a medium, which is why Young’s modulus and the Poisson ratio can be determined with knowledge of $//<![CDATA[ \begin{array}{l}v_p\end{array} //]]>$ or $//<![CDATA[ \begin{array}{l}v_s\end{array} //]]>$. ^[3][4] Elastic waves will be refracted or reflected when hitting an interface (geometrical boundary, defects, etc.) in a ratio that correlates with the difference in impedance of the two mediums. The impedance Z of a medium is the product of its density $//<![CDATA[ \begin{array}{l}\rho\end{array} //]]>$ and the acoustic wave velocity $//<![CDATA[ \begin{array}{l}v\end{array} //]]>$. The reflection coefficient of an elastic wave between two mediums with a normal incidence towards the interface is described by:

$//<![CDATA[ \begin{array}{l}R_{1,2}=\frac{Z_2-Z_1}{Z_2+Z_1}=\frac{\rho_2 v_2-\rho_1 v_1}{\rho_2 v_2+\rho_1 v_1}\end{array} //]]>$

The reflected energy at an interface is $//<![CDATA[ \begin{array}{l}R^2\end{array} //]]>$ and, as shown above, is determined by the difference of impedance of two mediums. Elastic waves are also subject to attenuation due to scattering at inhomogeneities and energy absorption by inelastic processes and inner friction, as well as a decrease in intensity due to the geometrical spreading in three dimensions. While both effects increase with the travel path of the wave, attenuation and scattering are furthermore determined by an attenuation coefficient, which in return is dependent on the material and the wave frequency. The attenuation coefficient (and thus attenuation and scattering) increases with higher frequencies, and therefore penetration depth decreases. However, defects within the material can only reflect an elastic wave if they’re larger than $//<![CDATA[ \begin{array}{l}\lambda/2\end{array} //]]>$, which can make it difficult to achieve a sufficient resolution while maintaining a reasonable penetration depth, especially for highly heterogeneous materials.^[5]

Doppler effect

When a source emits a harmonic wave, i.e. sinusoidal wave with a known frequency $//<![CDATA[ \begin{array}{l}f_0\end{array} //]]>$, while moving relatively towards or away from an observer with a relative velocity $//<![CDATA[ \begin{array}{l}\nu\end{array} //]]>$ (in this case a laser reflected from the vibrating surface of a specimen), a frequency shift will be measurable at the point of the observer.^[6] The shifted frequency $//<![CDATA[ \begin{array}{l}f_1\end{array} //]]>$ is described as follows:

$//<![CDATA[ \begin{array}{l}f_1=f_0\sqrt{\frac{c+\nu}{c-\nu}}\qquad\end{array} //]]>$ or $//<![CDATA[ \begin{array}{l}\qquad\ f_1=f_0+\Delta f\end{array} //]]>$

With:

$//<![CDATA[ \begin{array}{l}f_0\end{array} //]]>$...original frequency $//<![CDATA[ \begin{array}{l}[Hz]\end{array} //]]>$

$//<![CDATA[ \begin{array}{l}f_1\end{array} //]]>$...shifted frequency $//<![CDATA[ \begin{array}{l}[Hz]\end{array} //]]>$

$//<![CDATA[ \begin{array}{l}\Delta f\end{array} //]]>$...frequency shift $//<![CDATA[ \begin{array}{l}[Hz]\end{array} //]]>$

$//<![CDATA[ \begin{array}{l}c\end{array} //]]>$...acoustic velocity or speed of light (depending on medium) $//<![CDATA[ \begin{array}{l}[m/s]\end{array} //]]>$

$//<![CDATA[ \begin{array}{l}\nu\end{array} //]]>$...relative velocity between source and observer $//<![CDATA[ \begin{array}{l}[m/s]\end{array} //]]>$

$//<![CDATA[ \begin{array}{l}\nu\end{array} //]]>$ is positive when the source moves towards the observer, and negative if it moves away.^[7] This means the observer will perceive a higher frequency $//<![CDATA[ \begin{array}{l}f_1\end{array} //]]>$ compared to $//<![CDATA[ \begin{array}{l}f_0\end{array} //]]>$ when the source moves towards him and a lower frequency if the source moves away. Since $//<![CDATA[ \begin{array}{l}f_0\end{array} //]]>$, $//<![CDATA[ \begin{array}{l}f_1\end{array} //]]>$ and $//<![CDATA[ \begin{array}{l}c\end{array} //]]>$ are known or can be easily measured, the Doppler effect is well suited for determining the relative velocity between an object and an observer. In case the relative velocity between source and observer is very small compared to the speed of light, $//<![CDATA[ \begin{array}{l}\Delta f\end{array} //]]>$ is determined by:^[8]

$//<![CDATA[ \begin{array}{l}\Delta f=\frac{2\nu}{\lambda}\end{array} //]]>$

Laser Based Ultrasonics

Laser based generation of ultrasound

The most distinct feature of laser-based ultrasonic testing is the contactless generation of elastic ultrasonic waves within a structure by a high-energy laser pulse (usually 1-100 ns long). The laser light is partially absorbed by the material within a small volume near the surface (ideally the top 10-100 microns). Depending on the power density of the laser pulse two possible mechanism can ensue, either within the thermoelastic regime or the ablation regime.^[9][10][11] In case of a lower power-density pulse the absorbed energy leads to a local transient surface heating, which causes a local thermoelastic expansion and generates principal stresses parallel to the surface. The quickly developing strains radiate energy as elastic ultrasonic longitudinal, shear and Rayleigh surface waves that propagate through the structure or along the surface (Rayleigh).^[11][12] The angular dependence of amplitudes of P-waves generated in the thermoelastic regime is depicted in Figure 1, which shows a plot of equation [1]. Experiments have confirmed a good match between the theoretical values and reality. ^[13]

$//<![CDATA[ \begin{array}{l}u_{r,t}\propto\frac{\sin \theta \sin 2 \theta\sqrt{k^2-\sin^2 \theta}}{(k^2-2\sin^2 \theta)^2+4\sin^2\theta\sqrt{1-\sin^2 \theta}\sqrt{k^2-\sin^2 \theta}}\qquad\end{array} //]]>$ [1]^[13]

$//<![CDATA[ \begin{array}{l}u_{r,t}\end{array} //]]>$...angular dependence of P-wave amplitudes in the thermoelastic

$//<![CDATA[ \begin{array}{l}k\end{array} //]]>$...velocity ratio $//<![CDATA[ \begin{array}{l}v_p/v_s\end{array} //]]>$

$//<![CDATA[ \begin{array}{l}\theta\end{array} //]]>$...direction between surface normal and considered angle of P-wave

What is striking is that the P-waves with the biggest amplitude propagate in an angle of 67° away from the source, while the amplitude of P-waves at 0° is zero. Consequently, there are no strong (reflected) P-waves to be expected directly opposite or directly beside the generation point. This mechanism is entirely non-destructive. The penetration depth of the laser light and the amount of energy absorbed corresponds to the absorption characteristics of the material at the wavelength of the laser. This in turn has an influence on the amplitudes of the generated ultrasonic waves. ^[2] If the source of ultrasonic waves lies beneath the surface, either due to a bigger penetration depth of the laser light or a thin, transparent surface coating (paint, thin liquid film, etc.), there is a constraining effect of the material above it, causing an emission of strong longitudinal waves perpendicular to the surface, comparable to the wavefront generated within the ablation regime.

A higher laser power density (for common metals $//<![CDATA[ \begin{array}{l}~10^7[W/cm^2]\end{array} //]]>$) leads to the melting and vaporization of a small amount of surface material (ablation regime). The recoil effect resulting from the material ejection produces principal stresses normal to the surface and strong longitudinal waves travelling perpendicularly to the surface, which can be seen in the angular dependence for P-waves shown in Figure 2, which is a plot of equation [2].^[11][13]

$//<![CDATA[ \begin{array}{l}u_{r,a}\propto\frac{2k^2\cos \theta(k^2-2\sin^2 \theta)}{(k^2-2\sin^2 theta)^2+4\sin^2\theta\sqrt{1-\sin^2 \theta}\sqrt{k^2-\sin^2 \theta}}\qquad\end{array} //]]>$ [2]^[13]

$//<![CDATA[ \begin{array}{l}u_{r,a}\end{array} //]]>$...angular dependence of P-wave amplitudes in the ablation regime

Shear waves and Rayleigh surface waves are generated as well within the ablation regime. Although this mechanism isn’t entirely non-destructive, the usual surface damage of a few microns is often accepted or even insignificant for coarse materials like concrete. The typical absorption characteristics of different materials necessitate the selection of the laser system type with the best fitting wavelength since a high absorption ratio is desired. For most metals, concrete, asphalt and composite materials a Nd:YAG laser with a wavelength of 1.064 µm is used, whereas the wavelength of 10.600 µm of a CO2 Laser corresponds best with ceramics and glasses. ^{[2][11][14][15]} In practice the distance from which the pulsed laser can be launched ranges from a few centimetres to a few meters, allowing for the usage in harsh environments, at high temperatures or when the inspected part is not directly accessible.^[16] Since technically the laser pulse isn’t the source of the ultrasonic waves, but the material itself, the incidence of the laser pulse doesn’t necessarily need to be normal to the surface. Nevertheless, the amplitudes of the ultrasonic waves decrease with the incidence of the generating laser from a normal to surface incidence to near grazing incidence, due to the diminishing energy density at generation point. The energy density at generation point decreases by $//<![CDATA[ \begin{array}{l}\cos \theta\end{array} //]]>$ ($//<![CDATA[ \begin{array}{l}\theta\end{array} //]]>$ angle between surface normal and generating laser beam).^[17] Another favourable characteristic of laser induced ultrasonic pulses is their wide frequency range which extend from zero frequency to more than 50 MHz.^[11]

Laser based detection of ultrasound

As a consequence of P- and S- waves reaching the surface of the specimen, ultrasonic vibrations with very small amplitudes of the magnitude of 0.1 to few nanometers occur on the surface. These displacements have a normal component that can be measured by interferometric techniques (in plane displacements can be detected as well). ^[11][18]Therefore a second high-power continuous wave or significantly longer (in respect to the generation pulse) laser pulse (for most applications 10-100 μs long) by a Nd:YAG laser with a wavelength of 1.064 µm is directed towards the surface of the specimen (not necessarily the same surface as for the generation laser; depending on desired results) in order to capture all relevant echoes. The requirement for a high-power laser is especially given when inspecting highly absorbing surfaces and/ or the standoff is relatively high, as well as for improvement of sensitivity, which is proportional to power.^[16] The detection laser is scattered from the specimen’s surface, with the scattered light being subject to a phase shift due to the surface displacement and a frequency shift due to the high velocity of the vibration (Doppler effect). This scattered beam has a random distribution of intensity and phase, which is a so called optical speckle that corresponds to the roughness of the surface. There are several approaches to measure these phase or frequency modulations, reference beam interferometry being one of the most common.^[18]

Reference beam interferometry

As depicted in fig. 3 the detection laser is split within a typical Michelson-interferometer at a beam splitter to derive a reference beam, while the other part of the laser is directed towards the specimen’s surface. There it is modulated as described above and reflected into the interferometer, while the reference beam is reflected from a stationary mirror. When the reference beam and the inspection beam are recombined, an interference pattern is generated depending on the relative phase of both beams. The path length of the signal beam changes direct proportionality with the surface displacement.^[19] With this method however, there is a sign ambiguity, meaning it is not possible to retrieve information on the directivity of the surface displacements from the output signal. In many cases in ultrasonic testing this information isn’t required, since the mere time of arrival of an ultrasonic signal to the surface is of interest. Directional discrimination can be achieved with heterodyne interferometers, whose essential difference to the method described above is a frequency offset of the reference beam of usually 40 MHz by a frequency shifter (e.g. Bragg cell).^[20] When there’s no surface movement the output signal essentially is this shifted frequency (also called beat frequency). In case the surface moves towards the interferometer the beat frequency increases and vice versa. This method enables a distinct identification of surface displacement and velocity.^[21] The speckle affected reflected signal beam can be made sufficiently uniform by focusing the incoming probe beam to the surface. This is necessary because the relatively small phase shift induced by the surface vibration is almost canceled out by the random phase distribution in the speckled beam. The focusing of the probe beam on the surface of highly absorbing materials may however cause surface damage. The result can still be variable since the intensity of the collected speckle is very variable.^[22] Moreover, this technique enables point-like detection of displacements and has flat response over a wide bandwidth (10 kHz-100 MHz).^[11][23] At higher amplitude levels (regarding change of path length; usually larger than a few microns) the response of the interferometer ceases to be linear, which is why it’s necessary to reduce the system’s sensitivity to low frequency, relatively high amplitude background vibrations. This is achieved by electromagnetic or piezoelectric displacement devices, that change the optical path length of the reference beam by moving the mirror.^[24]

Fabry-Perot

This type of interferometry makes use of so called confocal Fabry-Perot setup. This is essentially a type of optical resonator consisting of two identical concave mirrors with a cavity in between them, which equals the radius of the curvature of the mirrors. A ray confined in the cavity reflects multiple times until it retraces its path, and therefore this type of system has a high throughput. Consequently, frequency or phase modulated waves are demodulated by the Fabry-Perot resulting in an amplitude modulation of the wave which can be detected and interpreted as a certain movement of the surface. Characteristic for this method is its low sensitivity to low frequencies (generally below 2 MHz) which can be considered as a weakness, whereas its practically non-existent sensitivity to very low frequencies is advantageous for a use in industrial environments (eliminates problems regarding background vibrations). Another favorable property is the fairly large detection spot, which is a result of the confocal mirror directing light from wide angles and which is especially helpful with respect to optical speckles. ^[16]

Advantages and Disadvantages

Advantages

• can be operated remotely
• applicable for geometrically complex geometries
• useable in harsh environments and under elevated temperatures
• generation of broadband (zero frequency to 50 MHz) ultrasonic pulses
• simultaneous generation of P-, S- and surface waves
• reproducible process
• absolute calibration standard for laser interferometry
• allows fast scanning of large areas

^[2][23]

Disadvantages

• expensive compared to conventional ultrasonic methods
• generally less sensitive than piezoelectric based ultrasonics
• laser safety precautions necessary
• generation efficiency dependent on material absorption properties

^[2]

Literature

• Blouin, A., Lévesque, D., Néron, C., Drolet, D., Monchalin, J.-P.: Improved resolution and signal-to-noise ratio in laser-ultrasonics by SAFT processing. Optics Express, (1998) 2:13.
• Blodgett, D.W., Baldwin, K.C.: Laser-Based Ultrasonics: Applications at APL. John Hopkins APL Technical Digest, (2005) 26:1.
• Große, C. U.: Einführung in die Zerstörungsfreie Prüfung im Ingenieurwesen. Grundlagen und Anwendungsbeispiele. Version 2017-10-25-Skript_ZfP_70.
• Wang, H.-C., Fleming, S., Lee, Y.-C.: A Remote, Non-destructive Laser Ultrasonic Material Evaluation System with Simplified Optical Fibre Interferometer Detection. J. Nondestruct. Eval.,(2009) 28.
• Günther, H.: Grenzgeschwindigkeiten und ihre Paradoxa. Springer, Wiesbaden (1996).
• Scruby, C.B., Drain, L.E.: Laser Ultrasonics. Techniques and applications. Adam Hilger, Birstol (1990).
• Osterkamp, M.: What is LaserUT. PaR Systems. 28.02.2018.
• Monchalin, J.- P., Aussel, J.- D.: Ultrasonic Velocity and Attenuation Determination by Laser-Ultrasonics. J. Nondestruct. Eval., (1990) 9:4.
• Jacobs, L.J., Whitcomb, R.W.: Laser Generation and Detection of Ultrasound in Concrete. J. Nondestruct. Eval., (1997) 16:2.
• Nowland Hopko, S., Ume, I. C.: Laser Ultrasonics: Simultaneous Generation by Means of Thermoelastic Expansion and Material Ablation. J. Nondestruct. Eval., (1999) 18:3.
• Brocklehurst, F.K., Moss, B.C., Smith, R.L., Tasker, C.G.: Evaluation of the Use of Laser Ultrasonics for the Rapid, Noncontact Inspection of Concrete and Asphalt. National Research Council, Washington, DC(1992).
• Monchalin, J.- P.: Non contact generation and detection of ultrasound with lasers. Industrial Materials Institute - NRC, Boucherville, Quebec, Canada. 28.02.2018.
• Monchalin, JP., Néron, C., Bussière, J.F. et al.: Laser-Ultrasonics: From the Laboratory to the Shop Floor. Advanced Performance Materials, (1998) 5:7.
• Lindamood, L.: Laser Ultrasonic Testing: An EnLIGHTening Advancement in Nondestructive Evaluation. EWI (2017). 28.02.2018.
• Laser-Doppler Vibrometrie. Polytec. 28.02.2018.

References

1. Blouin, A., Lévesque, D., Néron, C., Drolet, D., Monchalin, J.-P.: Improved resolution and signal-to-noise ratio in laser-ultrasonics by SAFT processing. Optics Express, (1998) 2:13, p. 532.
2. Blodgett, D.W., Baldwin, K.C.: Laser-Based Ultrasonics: Applications at APL. John Hopkins APL Technical Digest, (2005) 26:1, p.36-37.
3. Große, C. U.: Einführung in die Zerstörungsfreie Prüfung im Ingenieurwesen. Grundlagen und Anwendungsbeispiele. Version 2017-10-25-Skript_ZfP_70, p.37-43.
4. Wang, H.-C., Fleming, S., Lee, Y.-C.: A Remote, Non-destructive Laser Ultrasonic Material Evaluation System with Simplified Optical Fibre Interferometer Detection. J. Nondestruct. Eval., (2009) 28, p. 76.
5. Große, C. U.: Einführung in die Zerstörungsfreie Prüfung im Ingenieurwesen. Grundlagen und Anwendungsbeispiele. Version 2017-10-25-Skript_ZfP_70, p. 50-54.
6. Günther, H.: Grenzgeschwindigkeiten und ihre Paradoxa. Springer, Wiesbaden (1996), p.234.
7. Günther, H.: Grenzgeschwindigkeiten und ihre Paradoxa. Springer, Wiesbaden (1996), p. 249.
8. Scruby, C.B., Drain, L.E.: Laser Ultrasonics. Techniques and applications. Adam Hilger, Birstol (1990), p. 84.
9. Osterkamp, M.: What is LaserUT. https://www.par.com/files/9913/6640/3159/WhatisLaserUT.pdf, 01.03.2018., p. 9.
10. Scruby, C.B., Drain, L.E.: Laser Ultrasonics. Techniques and applications. Adam Hilger, Birstol (1990), p. 16.
11. Monchalin, J.- P., Aussel, J.- D.: Ultrasonic Velocity and Attenuation Determination by Laser-Ultrasonics. J. Nondestruct. Eval., (1990) 9:4, p. 212.
12. Jacobs, L.J., Whitcomb, R.W.: Laser Generation and Detection of Ultrasound in Concrete. J. Nondestruct. Eval., (1997) 16:2, p. 59.
13. Nowland Hopko, S., Ume, I. C.: Laser Ultrasonics: Simultaneous Generation by Means of Thermoelastic Expansion and Material Ablation. J. Nondestruct. Eval., (1999) 18:3, p. 92.
14. Brocklehurst, F.K., Moss, B.C., Smith, R.L., Tasker, C.G.: Evaluation of the Use of Laser Ultrasonics for the Rapid, Noncontact Inspection of Concrete and Asphalt. National Research Council, Washington, DC(1992). p. 2.
15. Brocklehurst, F.K., Moss, B.C., Smith, R.L., Tasker, C.G.: Evaluation of the Use of Laser Ultrasonics for the Rapid, Noncontact Inspection of Concrete and Asphalt. National Research Council, Washington, DC(1992). p. 4.
16. Monchalin, J.- P.: Non contact generation and detection of ultrasound with lasers. Industrial Materials Institute - NRC, Boucherville, Quebec, Canada. p. 2-3. 28.02.2018.
17. Monchalin, JP., Néron, C., Bussière, J.F. et al.: Laser-Ultrasonics: From the Laboratory to the Shop Floor. Advanced Performance Materials, (1998) 5:7, p. 8
18. Monchalin, JP., Néron, C., Bussière, J.F. et al.: Laser-Ultrasonics: From the Laboratory to the Shop Floor. Advanced Performance Materials, (1998) 5:7, p. 10-11.
19. Scruby, C.B.,Drain, L.E.: Laser Ultrasonics. Techniques and Applications. Adam Hilger, Birstol (1990), p.76-78.
20. Scruby, C.B., Drain, L.E.: Laser Ultrasonics. Techniques and Applications. Adam Hilger, Birstol (1990), p. 111-113.
21. Laser-Doppler Vibrometrie. Polytec. 28.02.2018.
22. Monchalin, JP., Néron, C., Bussière, J.F. et al.: Laser-Ultrasonics: From the Laboratory to the Shop Floor. Advanced Performance Materials, (1998) 5:7, p.12.
23. Lindamood, L.: Laser Ultrasonic Testing: An EnLIGHTening Advancement in Nondestructive Evaluation. EWI (2017). p. 2. 28.02.2018.
24. Scruby, C.B., Drain, L.E.: Laser Ultrasonics. Techniques and Applications. Adam Hilger, Birstol (1990), p.105-107.