Key words

laser; electronic speckle pattern interferometry (ESPI); digital shearography; scanning photogrammetry; laser Doppler vibrometry (LDV); laser-ultrasound; safety

12.1 Introduction

Since the debut of light amplification by stimulated emission of radiation (LASER) in the 1960s, laser-based sensing techniques have found expanding application and growing influence in science and engineering, as well as in academia and industry. The noncontact nature of using coherent electromagnetic waves (monochromatic radiation) in detecting the characteristics of a remote target in a laser-based sensing is enabled by the high energy levels at which laser beams are generated. This high energy level characteristic of laser-based sensing improves the signal-to-noise ratio (SNR) in many engineering inspection problems, still creating new applications for laser in the twenty-first century.

Laser-based sensing can be destructive (e.g., electric speckle pattern interferometry (ESPI)) or nondestructive (e.g., laser Doppler vibrometry (LDV)), depending on the purpose of investigation. It is one of the noncontact inspection/evaluation/testing (NDI/E/T) methods applied to assess and monitor the performance/condition of engineering materials and structures. While the theoretical foundation of laser-based sensing was laid down by A. Einstein in 1917 (Einstein, 1917), its practical use and wide application were not possible until W.E. Lamb and R.C. Retherford demonstrated the mechanism of stimulated emission in hydrogen spectra in 1947, the principle of microwave amplification by stimulated emission of radiation (MASER) developed by J.R. Gordon, H.J. Zeiger, and C.H. Townes in 1955, the method of optical pumping theoretically by A. Kastler in 1950 and experimentally by J. Brossel, A. Kastler, and J. Winter in 1952 (Brossel et al., 1952; Kastler, 1957), the experimental work on the use of ruby for lasers, and the MASER principle for the amplification of light proposed by A.L. Schawlow and C.H. Townes in 1958. G. Gould coined the acronym LASER in 1959, while Schawlow and Townes proposed ‘optical maser’ for the same concept in 1957.

Early developments of lasers include the first solid-state laser (made of synthetic ruby crystal) built by T.H. Maiman at the Hughes Research Labs in 1960, and the first gas laser (made of helium–neon) built by D.R. Herriott, A.M. Javan, and W.R. Bennett at the Bell Telephone Labs in 1960. A.M. Prokorov and N. Basov also made major contributions in the theoretical development of lasers.

While widely used in precision manufacturing at the present time, lasers have been applied in the sensing, inspection, and measurement applications in science and engineering. Since lasers can range in power from the harmless level (> 1 mW) (e.g., laser pointer) to the dangerous level (1.3 PW or 1.3 × 10¹⁵ W) (e.g., the laser developed by the Lawrence Livermore National Laboratory (LLNL)), they are suitable for the distant inspection and remote sensing of physical parameters such as displacement/deformation and velocity in hostile environments (e.g., high temperature) for a wide variety of applications. Furthermore, lasers have also been used to ignite and/or drive a nuclear fusion reaction. The US National Ignition Facility at the LLNL is a representative example.

When choosing laser-based sensing techniques for a particular application, energy level (watts or W for continuous wave (CW) laser and millejoules or mJ for pulsed laser), laser carrier wavelength (nm), laser resolution (e.g., spot size = μm, angular resolution), laser pulse length (ns) (for pulsed lasers), and pulse interval (ns) (for pulsed lasers) of a selected laser system are usually evaluated. CW or pulsed lasers can be selected, depending on the required sampling resolution for time-dependent parameters such as velocity and dynamic strain. In general, the advantages of laser techniques include (i) noncontact nature of measurement, (ii) flexible range of measurement, (iii) multi-dimensional measurement, and (iv) real-time assessment. On the other hand, the disadvantages of laser techniques include safety considerations and data interpretation.

In this chapter, the principles of lasers are first introduced in a concise manner. Current laser-based sensing techniques including laser interferometry or ESPI, laser digital shearography, laser scanning photogrammetry, LDV, and laser-ultrasound, and other techniques are also described. Safety measures and standards in the use of laser-based sensing techniques are also provided. Finally, research issues associated with laser-based sensing are discussed in Section 12.11.

12.2 Laser principles

A laser is a device that generates an intense light beam of highly monochromatic radiation (coherent electromagnetic waves in space and time) through externally amplifying a light source inside an optical resonator/cavity. In other words, it is an optical oscillator, consisting of an amplifying medium placed inside an optical resonator. Standing waves are excited within the medium to amplify the input light beam in order to produce an intense, highly monochromatic output beam. Theoretically, this is achieved by stimulated or induced emission of radiation by atomic systems. In the following, key principles in the formation of laser are addressed.

12.2.1 Stimulated emission and thermal radiation

As A. Einstein first introduced in 1917, radiation of atomic systems can be generated by the concept of stimulated (induced) emission. When an active, amplifying medium is subject to an intense burst of broadband (incoherent) light, some of the input light energy is absorbed by the atoms of the medium (or the atoms in the medium are excited) such that the energy level of these atoms is increased to a higher level. After achieving the desired energy level, these atoms are released to return to their unexcited state (ground state, lower energy level), accompanied by the emission of light waves (photons).

Selection of a laser amplifying medium is made based upon the requirement that the excited (stimulated) atoms in the amplifying medium can pause or stay in an energy level (metastable state) before returning to their ground state. Light emission of the same input light wavelength in the medium can occur during this process. In other words, stimulated atoms start emitting single wavelength (coherent) light. Figure 12.1 shows three energy states in laser.

Consider a quantized atomic system in two different energy states: E₁ and E₂ with the numbers of atoms N₁ and N₂, respectively. Assuming that E₂ is higher than E₁ and the system is in equilibrium with thermal radiation at a given temperature T, the relationship between N₁ and N₂ can be determined by Boltzmann’s equation:

$\frac{{\mathit{N}}_2}{{\mathit{N}}_1}=\frac{exp\left(-{\mathit{E}}_2/\mathit{kT}\right)}{exp\left(-{\mathit{E}}_1/\mathit{kT}\right)}$ [12.1]

[12.1]

where k = R/N_A is the Boltzmann constant (k = 1.380648 × 10^− 23 J/K), R the gas constant (= 8.314462 J/mol/K), N_A the Avogadro constant (= 6.022141 × 10²³ mol^− 1), and T the absolute temperature (K).

Emission of a photon occurs when an atom in E₂ decays to E₁. Figure 12.2 conceptually describes the energy absorption and emission of atoms. The frequency of emitted/radiated photons (light wave), f, is determined by:

$\mathit{v}=\frac{\left({\mathit{E}}_2-{\mathit{E}}_1\right)}{\mathit{h}}$ [12.2]

[12.2]

where h is the Planck constant (= 6.6260069 × 10^− 34 J-s). Since the emission of photons by an atom can be created spontaneously or by stimulation, the energy equilibrium between emission and absorption is:

${\mathit{N}}_2{\mathit{A}}_{21}+{\mathit{N}}_2{\mathit{B}}_{21}{\mathit{u}}_{\mathit{v}}={\mathit{N}}_1{\mathit{B}}_{12}{\mathit{u}}_{\mathit{v}}$ [12.3]

[12.3]

where A₂₁ is the probability per unit time for spontaneous emission, B₂₁ the proportional constant for stimulated emission, B₁₂ the proportional constant for stimulated absorption, and u_v the energy density of the radiation frequency v. By applying the Planck radiation formula, the energy density of the radiation frequency can be found by:

${\mathit{u}}_{\mathit{v}}=\left(\frac{{\mathit{A}}_{21}}{{\mathit{B}}_{21}}\right)\frac{1}{\left(exp\left(\mathit{hv}/\mathit{KT}\right)-1\right)}$ [12.4]

[12.4]

The ratio of stimulated emission rate to spontaneous emission rate, R_ss, is determined by:

${\mathit{R}}_{\mathit{ss}}=\frac{\left({\mathit{B}}_{21}{\mathit{u}}_{\mathit{v}}\right)}{{\mathit{A}}_{21}}=\frac{1}{\left(exp\left(\mathit{hv}/\mathit{KT}\right)-1\right)}$ [12.5]

[12.5]

When the value of kT becomes much greater than that of hv, R_ss is significantly increased, as shown in Fig. 12.3.

12.3 Laser interferometry or electronic speckle pattern interferometry

Laser interferometry or ESPI, also known as computer-aided speckle pattern interferometry (CASPI), television holography, and video holography, is the electronic processing of speckle patterns produced by a laser interferometer consisting of two laser beams (using a beam splitter) generated from the same laser source. Interferometers can be categorized into amplitude-splitting (e.g., Michelson, Mach-Zehner (Zipser and Franke, 2004), Fabry–Pérot, Fizeau) and wavefront-splitting (e.g., Fresnel bi-prism, Rayleigh, Lloyd’s mirror).

When a target with a perfectly-smooth surface is illuminated by laser, no speckles will be generated and an amplitude-splitting interferometer such as Michelson interferometer can be used. However, when a rough surface is encountered, a speckle pattern (irregular fringe pattern) will be revealed by the use of a coherent light source such as a laser. In ESPI, the laser interferometer generates a primary beam split into two beams: one is the object beam, which passes through a beam expander and is used to illuminate the target, and the other is the reference beam, which is directed to the image recording device or sensor, typically a charge-coupled device (CCD). The returned beam reflected from the target surface is also recorded by the CCD. When the beam path difference between the object beam and the reference beam is within the coherent length of the primary laser beam, the two beams will interfere and produce a speckle pattern. The purpose of a frame grabber is to capture and store these speckle images for post-processing and data interpretation. Further information on speckle interferometry can be found at Jones and Wykes (1989).

Measurement schemes for in-plane and out-of-plane deformation and velocity are shown in Figs 12.4 and 12.5, respectively.

12.3.1 Laser holographic interferometry

Laser holographic interferometry is a technique combining laser interferometry and holography. The principle of holography is based on the fact that, via simultaneously recording a reference light (laser) beam (split by a beam splitter) and a returned, scattered light beam from the surface of a target on a holographic plate, reconstruction of an exact replica of the target can be achieved. The image recorded on the holographic plate is called a hologram of the target, which carries both the amplitude and phase of the returned, scattered beam from the target. The first exposure of the holographic plate by the reference beam results in a reference hologram, while the second exposure by the returned beam imposes all changes of the target to the reference hologram. The result is a fringe pattern of the target, containing all information about the change that occurred between the first and the second exposures. In laser holographic interferometry, both a reference beam and a returned beam (scattered from the target) are required. Additional information on holographic imaging can be found at Erf (1974) and Vest (1979). Further information on laser interferometry can be found at Chu (2007) and Hariharan (2007).

Twitto et al. (1996) used an He–Ne laser to illuminate loaded brass plate specimens and a piezoelectric (PZT) crystal controlled mirror for the holographic imaging of the specimen. The reflected laser beam from the specimen was measured by a CCD sensor and the PZT mirror. Phase shifting was included in the development of their work. Figure 12.6 illustrates the experimental set-up of a holographic interferometry utilizing phase shift.

12.4 Laser digital shearography

Laser digital shearography is an interferometric technique developed to overcome the limitations of holography, due to the advantages including (i) removing requirement for a reference light beam, (ii) reduced coherence length, and (iii) direct measurement of surface strain (Hung et al., 2007). The principle of image shearing is based on a light beam passing through a doubly refractive prism and splitting into two angularly separated beams. Figure 12.7 shows the principle of laser digital shearography. Through the image-shearing device, two nonparallel light beams scattered from two different object points become nearly collinear, and the resulting low spatial frequency of the interference fringes can be resolved with a low resolution image sensor such as a CCD. The two sheared wavefronts transmitted by the two axes of the image's hearing device, however, are orthogonally polarized. Therefore, they will not interfere with each other. To enable interference, a polarizer with its polarization axis oriented at an azimuth of 45° is required, as shown in Fig. 12.7. Since the illuminated object surface is generally optically rough, interference of the two sheared wavefronts will result in a speckle pattern embedded in the shearographic image. This speckle pattern image of the specimen before loading can be represented by Hung and Ho (2005):

$\mathit{I}-{\mathit{I}}_0\left(1+\mathit{\mu }cos\mathit{\varphi }\right)$ [12.6]

[12.6]

where I is the speckle image intensity at the image plane of the CCD sensor before loading, I₀ the intensity of the laterally sheared image, μ the amplitude of modulation of the speckle pattern, and ϕ the random phase angle. The speckle pattern image of the specimens after loading can be expressed by:

${\mathit{I}}^{\prime }={\mathit{I}}_0\left[1+\mathit{\mu }cos\left(\mathit{\varphi }+\text{Δ}\right)\right]$ [12.7]

[12.7]

where I′ is the speckle image intensity at the image plane of the CCD sensor after loading and Δ the phase angle change due to surface deformation. To remove the random phase angle change, the difference between I and Ι′ is used:

${\mathit{I}}_{\mathit{d}}=2{\mathit{I}}_0\left[\mathit{\mu }sin\right]\left(\mathit{\varphi }+\frac{\text{Δ}}{2}\right)sin\frac{\text{Δ}}{2}$ [12.8]

[12.8]

where I_d is evaluated at the pixel level and used for relating the level of loading and surface deformation. Digital shearography using multiple wavelengths has also been proposed by Kastle et al. (1999). A comparison between digital shearography and ESPI can be found at Findeis and Gryzagoridis (2004).

12.5 Laser scanning photogrammetry

Laser scanning is a remote inspection/sensing method in which laser beams are steered/directed along a line (2D) (Fig. 12.8) or within an area (3D) for distance measurement. Laser scanners operate in either (i) stationary mode or (ii) mobile mode. In stationary mode, laser scanners are typically installed on a tripod, while in mobile mode, laser scanners can be handheld, vehicle-mounted, robot-mounted, or aircraft-mounted. Laser scanning photogrammetry is an integrated technique combining laser scanning and digital photogrammetry. Digital photogrammetry is based on the processing of images and produces digital terrain models (DTM), digital surface models (DSM), orthoimages, and 2D/3D reconstruction and classification of objects for mapping and visualization purposes (Baltsavias, 1999). The main differences between laser scanning and digital photogrammetry are summarized in Table 12.2.

Key components of laser scanning photogrammetry include: (i) a laser scanner, (ii) a digital camera, (iii) global positioning system (GPS) (when the position of the laser scanner is changed), (iv) a total station (optional), (v) a portable computer, and (vi) data processing software for 3D modeling and data registration.

El-Omari and Moselhi (2007, 2008) applied a 3D laser scanning system (Riegl LPM 100 VHS LADAR) and photogrammetry (including software, computer, and a digital camera) to generate point cloud data of a construction site and several boxes for image reconstruction and 3D modeling. Scanning angular resolution varied from 0.015° to 0.75°, and the inspection distance ranged from 1 to 50 m. Image resolution (or the distance between any two points in the point cloud image) depended on (i) the distance between the target and the laser scanner, and (ii) the angular resolution of the laser scanner. Inspection time depended on the required image resolution and the spatial range of inspection (size of the target). For example, with a distance of 4.3 m, horizontal angular range of 21°, vertical angular range of 18°, and an angular resolution of 0.015°, the total inspection time was 58 min, while the total inspection time could be reduced to 15 s in the case of 0.75° angular resolution (El-Omari and Moselhi, 2008).

Rodriguez et al. (2008) developed a portable laser photogrammetric sensor by combining a laser distance meter (ranger) (Leica Zinder Distro Plus) and a digital camera. The laser distance meter could measure distance ranging from 0.2 to 200 m at a resolution of 1.5 mm. Sturzenegger and Stead (2009) adopted laser scanning photogrammetry in a photogrammetric rock cut survey and used a 3D laser scanner (Optech ILRIS-3D), a digital camera, and data processing software. They scanned few terrestrial targets at distances ranging from 29 to 163 m. Armesto et al. (2009) applied an analog semi-metric camera, a laser total station, and a monoscopic digital photogrammetric station for the close-range (< 100 m) evaluation of timber roof construction. 3D finite element modeling was also adopted to perform static and dynamic analyses of the timber roof structure. In other damage detection problems such as surface potholes in asphalt pavement structures, 3D laser scanning technology was applied in literature (Koch and Brilakis, 2011).

12.6 Laser Doppler vibrometry

A laser Doppler vibrometer is a passive, nondestructive, dynamic interferometer in which the beam emerging from the laser is divided into two: one of which is used as a transmitting beam and the other as a reference beam for coherent detection.

The transmitting beam illuminates the surface of a target (which can be a solid structure, or a particle in air or in fluid), while the reference beam is combined with the returned beam from the target at the detector for coherent detection. The frequency shift between the returned beam and the reference beam is related to the instantaneous velocity of target, due to the Doppler effect. This frequency shift is the Doppler frequency shift. The basic laser Doppler vibrometer is a single point vibrometer. With the addition of a video camera and two orthogonal mirrors, scanning LDV systems can be made to allow large surface measurement with high spatial resolution. LDV is a mature technique for the nondestructive measurement of velocities of macroscopic objects (e.g., engineering structures), fluids (e.g., fluid flow structure), and vibration noise.

Depending on the type of velocity to be measured, LDV can be categorized into translational, rotational, and torsional LDV (Fig. 12.9). In translational LDV, the direction of measured motion is in parallel to the laser beam direction, while in rotational and torsional LDV, the direction of measured motion is perpendicular to the laser beam direction. LDV has been applied in the measurements of translational, rotational (Bell and Rothberg, 2000), and torsional (Halliwell, 1996; Drew and Stone, 1997; Oberholster et al., 2012) velocities in literature.

The principle of LDV is briefly described as follows. Consider the rotational velocity measurement problem using LDV, in which the laser beam is configured on the plane perpendicular to the axis of rotation. The target position x can be assumed to be:

$\mathit{x}={\mathit{x}}_{0}cos\left({\mathit{\omega }}_{\mathit{t}}\mathit{t}\right)$ [12.9]

[12.9]

where x₀ is the maximum amplitude of x,ω_t the angular velocity of rotation, and t the time variable. The rotational velocity υ is found by:

$\mathit{\upsilon }=\frac{\mathit{dx}}{\mathit{dt}}=-{\mathit{x}}_0{\mathit{\omega }}_{\mathit{t}}sin\left({\mathit{\omega }}_{\mathit{t}}\mathit{t}\right)$ [12.10]

[12.10]

When measuring the laser beam returned from a moving target, the optical frequency is shifted due to the Doppler effect. This Doppler frequency shift f_d is given by:

${\mathit{f}}_{\mathit{d}}=2\mathit{\upsilon }cos\frac{\mathit{\alpha }}{\mathit{\lambda }}$ [12.11]

[12.11]

where α is the intersection angle between the laser beam and the target velocity direction, and λ the optical wavelength. Substituting Equation [12.10] into Equation [12.11] leads to:

${\mathit{f}}_{\mathit{d}}=-2{\mathit{x}}_0{\mathit{\omega }}_{\mathit{t}}sin\left({\mathit{\omega }}_{\mathit{t}}\mathit{t}\right)cos\frac{\mathit{\alpha }}{\mathit{\lambda }}$ [12.12]

[12.12]

The intensity of measured returned laser beam I_v can be expressed as:

${\mathit{I}}_{\mathit{v}}={\mathit{I}}_{\mathit{v}0}cos⌊2\mathit{\pi f}+\mathit{\varphi }⌋={\mathit{I}}_{\mathit{v}0}cos\left[2\mathit{\upsilon }cos\frac{\mathit{\alpha }}{\mathit{\lambda }}+{\mathit{f}}_0\right]+\mathit{\varphi }$ [12.13]

[12.13]

where I_v0 is the maximum intensity of the returned beam, f= f_d+ f₀ the total frequency shift, f₀ a fixed requency shift used to improve SNR, and φ the random phase angle. In this result, the rotational velocity υ is related to the measured intensity of the returned laser beam I_v.

For basic LDV systems, the lead time for changing one point position to another can be less than 10^− 3 s. Typical velocity range for LDV is up to 11 000 rpm or 1000 rad/s. The frequency range in rotational velocity measurement problems is from 1 Hz to 10 kHz, at a calibration accuracy of approximately 0.5% (Castellini et al., 2006). For out-of-plane translational velocity measurements, the frequency range is from 0 Hz to 1 MHz (Olson et al., 2010).

Further information on the theory of rotational vibration measurements using LDV can be found at Bell and Rothberg (2000) and Rothberg and Tirabassi (2012).

LDV has been applied in various fields including modal vibrational analysis, rotating structures, on-line quality control, structural damage detection, assessment of microelectromechanical systems (MEMS) (DeCoster et al., 2008), micro-optical-electromechanical systems (MOEMS), and nanoelectromechanical systems (NEMS),material characterization (Huang et al. 2007), biomedical applications, and condition assessment in civil engineering (Castellini et al., 2006).

Pascal et al. (1993) applied a laser Doppler vibrometer (Polytec OFV 1000) to the characterization of vibrating plates.

Castellini and Santolini (1998) applied a laser Doppler vibrometer (Polytec OFV 3000) to measure the vibration of a naval propeller in water. Mallet et al. (2004) used LDV and Lamb waves on two rectangular plates for crack detection. Vanlanduit et al. (2004) investigated the microphony problem of a mask of a picture tube in a television using LDV. Damljanovic and Weaver (2005) applied a laser Doppler vibrometer (Polytec PDV 100) of 50 μm spot size and 0.05 μm/s resolution to the vibration measurement problem of rail specimens. Hong et al. (2007) applied an scanning laser Doppler vibrometer (SLDV) (Polytec LSV PSV 3000) to determine the piezoelectric coefficients and elastic constant of a piezoelectric thin film on silicon substrate cantilevers. The SLDV was used to measure the displacement and first/fundamental resonant frequency of thin film specimens. Prazenica et al. (2007) used LDV in the surface inspection of artwork and frescoes. Hapsara and Iliescu (2008) applied LDV to biomedical sensing by inspecting a bovine cortical tibia in vitro using a Polytec PSV 400 system. Olson et al. (2010) used LDV (Polytec PSV 400 3D-M) and Lamb waves to inspect 7075-T6 aluminum sheets. Oberholster and Heyns (2011) applied LDV (Polytec PSV 300) to characterize a multi-blade test rotor in operation.

Limitations of LDV include (i) measurement resolution (e.g., range-dependent and frequency-dependent resolution of analog and digital velocity decoders), (ii) measurement uncertainty (e.g., in the optical components of LDV and in the analog signal processor), (iii) direction uncertainty (e.g., misalignment of laser beam path with respect to the out-of-plane velocity direction), (iv) LDV standoff distance (e.g., surface reflection-dependent performance of LDV), (v) SLDV spatial resolution (e.g., minimum spot size of the laser beam), (vi) SNR (e.g., poor SNR on measuring low diffusive surfaces due to speckle effects), and (vii) environmental measurement conditions (e.g., dust, temperature, and pressure) (Castellini et al., 2006). Signal denoising and filtering to improve SNR of LDV can be achieved by hardware design or by software processing. Both frequency-domain (e.g., Fourier analysis) and time-domain (e.g., wavelets and multiresolution analysis (Prazenica et al., 2007)) filtering techniques can be applied.

Also, data compression and evaluation are needed when dealing with large amounts of data. Techniques such as principal component analysis (PCA) and singular value decomposition (SVD) can be used to evaluate the quality of data and to reduce the data size for interpretation (Vanlanduit et al., 2004). In addition, the use of auxiliary computational models in LDV, 3D modeling tools such as SolidWork® can be used for data visualization. Finite element (FE) models can also be used to complement the experimentally obtained LDV data (Olson et al., 2010).

12.7 Laser-ultrasound

Laser-ultrasound (also known as laser-ultrasonics, laser-based ultrasound) is an active, nondestructive, dynamic, interferometric, sensing method capable of revealing the subsurface condition of a target. Different from LDV, in laser-ultrasound, surface/subsurface ultrasonic vibration of a target is generated by one laser and the response is measured either by another laser (Amziane et al., 2012), or by other sensors (Park et al., 2009). In the case of using laser for both excitation and measurement, an interferometer is required to couple the two lasers. In laser-ultrasound the material surface serves as the emitting ultrasonic transducer. The ultrasonic source is produced either by a thermoelastic mechanism or material ablation. Detection is based on the Doppler effect that produces a frequency or phase modulation in the returned laser beam. Demodulation of laser signals is performed by the interferometer that is insensitive to speckle (confocal Fabry–Pérot, or photorefractive or Michelson interferometer). Laser-ultrasound is particularly useful for detecting flaws in targets with complex geometry and measurements on hot surfaces.

In laser-ultrasound techniques, ultrasonic waves or mechanical waves including body waves and surface waves are used to characterize a target. The governing equation of mechanical wave propagation in media is the Navier equation of motion. For homogeneous and isotropic materials (Malvern, 1969),

$\left({\mathit{\lambda }}_{\mathit{L}}+{\mathit{\mu }}_{\mathit{L}}\right)\nabla \left(\nabla \cdot \mathit{u}\right)+{\mathit{\mu }}_{\mathit{L}}{\nabla }^2\mathit{x}+\mathit{\rho F}=\mathit{\rho }\frac{{\partial }^2\mathit{u}}{\partial {\mathit{t}}^2}$ [12.14]

[12.14]

where λ_L= Eυ/(1 + υ)(1–2υ) and μ_L= E/2(1 + υ) are the Lamé constants, ∇ the Laplace operator or Laplacian, u= u(x_i; i = 1, 2, 3) the displacement (position) field of a unit cell, ρ the density of the unit cell, and F the body force of the unit cell per unit mass. The traction boundary condition is needed for solving the observable displacement field at the boundary.

${\mathit{\lambda }}_{\mathit{L}}\left(\nabla \cdot \mathit{u}\right)\stackrel{⌢}{\mathit{n}}+{\mathit{\mu }}_{\mathit{L}}\left(\frac{\partial {\mathit{u}}_{\mathit{i}}}{\partial {\mathit{x}}_{\mathit{j}}}+\frac{\partial {\mathit{u}}_{\mathit{j}}}{\partial {\mathit{x}}_{\mathit{i}}}\right)\cdot \stackrel{⌢}{\mathit{n}}={\mathit{F}}_{\mathit{BC}}$ [12.15]

[12.15]

where $\stackrel{⌢}{\mathit{n}}$ is the normal vector at the boundary and F_BC the prescribed force function acting on the boundary. Displacement fields at the boundary in different conditions (e.g., different mechanical properties: λ_L, μ_L) are measured in order to retrieve the variation in the mechanical properties of materials. Velocity fields ∂u/∂t and acceleration fields ∂²u/∂t² can be determined either analytically when displacement fields u are obtained by evaluating the Navier equation, or numerically when displacement fields are measured over a period of time. The Navier equation is also the general form of the governing equation of mechanical waves.

Depending on the boundary condition of the medium, different modes of mechanical waves can be accommodated by the medium. Body waves including compressional waves (or dilational waves, pressure waves or P-waves) and distortional waves (or equivoluminal waves, rotational waves, or shear or S-waves) can occur in an infinite medium or in a medium without finite boundaries. Surface waves including Rayleigh waves and Love waves can occur in a semi-infinite medium or in a medium with one finite boundary. Lamb waves can occur in thin plate and shells (special finite media). Table 12.3 lists the properties of selected mechanical waves. In Table 12.3, the direction of particle movement is determined with respect to the direction of wave propagation. It is obvious that changes in the mechanical properties (E, v, ρ) of a target are reflected in the variation of these wave velocities. The presence of anomalies on/inside a target, on the other hand, can also be reflected in the change of wave velocity, owing to the scattering of waves when encountering the anomalies. The phase and group velocities of Love waves (Jardetzky et al., 1957) can be numerically evaluated, while the phase and group velocities of Lamb waves can be obtained by solving the Rayleigh-Lamb equation (Viktorov, 1967). Further information about laser-ultrasound can be found at Scruby and Drain (1990).

Table 12.3

Types of mechanical waves

Class	Particle movement	Wave velocity	Name
Body wave	Parallel	$\mathit{v}=\sqrt{\frac{\mathit{E}\left(1-\mathit{\upsilon }\right)}{\mathit{\rho }\left(1+\mathit{\upsilon }\right)\left(1-2\mathit{\upsilon }\right)}}$	P-wave
	Orthogonal	$\mathit{v}=\sqrt{\frac{\mathit{E}}{2\mathit{\rho }\left(1+\mathit{\upsilon }\right)}}$	S-wave
Surface wave	Elliptical orbit, symmetrical mode	$\mathit{v}\cong \frac{0.87+1.2\mathit{\upsilon }}{1+\mathit{\upsilon }}\sqrt{\frac{\mathit{E}}{2\mathit{\rho }\left(1+\mathit{\upsilon }\right)}}$	Rayleigh wave
	Horizontally polarized, shear mode		Love wave
	Orthogonally symmetric and asymmetric	${\mathit{v}}_{\mathit{p}}=\frac{{\mathit{v}}_{\mathit{g}}}{sin\mathit{\phi }}$	Lamb wave
		where vp = phase velocity, vg = group velocity, φ = angle of incidence of the incident longitudinal wave

Pedrini et al. (2002) combined 3D digital holography with LDV (a pulsed Q-switched ruby laser) and a high-resolution CCD (Pulnix TM-1001) camera to reconstruct the dynamic profile of specimens with curved surfaces. Markov et al. (2003) proposed a whole field laser Doppler vibrometer, and demonstrated the performance of the system on a clamped, circular thin plate. The thin plate was remotely excited by a pulsed acoustical exciter, which generates a brief, broadband shock wave that stimulates the vibration of the plate. Huber et al. (2005) developed laser-acoustic microscopy using an Nd:YAG laser to generate 800 MHz (and higher) longitudinal mode ultrasound in several polymer films. Longo et al. (2006) used a Panametrics 10 MHz transducer to generate surface acoustic waves (SAW) on a steel beam and a laser vibrometer (Polytec PSV 300) for measuring the SAW response of the specimen. Kim et al. (2006) used a pulsed Nd:YAG laser to generate focused ultrasonic (Lamb) waves on an aluminum plate and a Michelson interferometer (Fiberpro MI7000) for measuring the response. Zhang et al. (2006) applied a pulsed Nd:YAG laser to generate Lamb waves on paper samples and a CW photorefractive laser interferometer for measurement. The ultrasonic wave response was used to determine the Young’s modulus, shear rigidity, and thickness of paper samples. In the aerospace industry, Kalms et al. (2008) applied a pulsed CO₂ laser to generate ultrasounds in semicrystalline polyphenylene sulfide thermoplastic composites (a material used in modern aircrafts such as Airbus A380) and a macropulse YAG laser for detection. In the nuclear industry, Liu and Yang (2008) applied a pulsed Nd:YAG laser, an interferometer (Polytec OFV 511), and an oscilloscope to characterize the cladding condition on hydrogen-charged Zircaloy tubes that are used in nuclear reactors. Material properties such as Young's modulus and Poisson's ratio were identified from the experiment.

Yu and Haupt (2010) proposed a distant acoustic-laser nondestructive evaluation (NDE) technique utilizing a high powered standoff parametric acoustic array (PAA) and LDV for the detection of debonding and delamination in multi-layer fiber reinforced polymer (FRP) wrapped-concrete systems. In their development, a focused sound beam is generated by the PAA such that the beam diameter can be controlled to directly and locally excite the surface area (where damages/defects are embedded) in the FRP-concrete systems, from a distance exceeding 30 m. Flaws and structural damage embedded underneath the externally bonded FRP plates or wrapped FRP sheets will be detected through the acoustically excited vibration launched by the system (Fig. 12.10). Surface dynamic signatures (Rayleigh waves) of damaged and intact regions are collected by laser Doppler vibrometer and analyzed for damage detection. The laser vibrometer beam size provides us the opportunity to obtain spatial resolution on the order of a millimeter. In this development, the PAA has two sources of sound: one source is high frequency, which is generated directly from one or more high-frequency transducers (2–7 kHz) and the other is low frequency, which is generated from nonlinear effects in the volume of air in front of the transducer (500 Hz–2 kHz). The PAA source can provide a practical means to deliver the necessary level of acoustic power in air to the void underneath the FRP sheet while minimizing system size and weight, reducing the sound level imposed on personnel close to the source. To be able to detect voids of small size between FRP and concrete, the use of PAA source exhibits potential advantages on this particular application.

Amziane et al. (2012) investigated the crack detection problem in a composite spherical particle consisting of four spherical layers, using a laser homodyne interferometer with 532 nm wavelength, 8 mm focal length, 0.5 numerical aperture, and 1 μm spot resolution.

Advantages of laser-ultrasound techniques include (i) noncontact and remote nature, (ii) applicability in hostile environments, (iii) rapid scanning capability, (iv) applicability on geometrically complex targets, (v) ability to generate ultrasonic point source both temporally and spatially, (vi) broadband inspection (in pulsed laser), (vii) high resolution, (viii) ability to generate both surface and bulk acoustic waves, and (ix) ability to generate various ultrasonic surface wavefronts.

Disadvantages of laser-ultrasound techniques include (i) being relatively expensive, (ii) lower sensitivity (iii) material-dependent ultrasound generation efficiency (e.g., material's optical absorption properties), and (iv) laser safety requirements.

12.8 Other laser-based techniques

Laser diffraction grating: Kurita and Ma (1998) used an He–Ne (helium–neon) laser beam (wavelength = 0.6328 μm) to illuminate the surface of a structural steel specimen in order to determine surface strain (ε_x) using a laser diffraction grating method. The intensity distribution of the diffracted laser beam was measured with a plasma-coupled device (PCD) sensor. The strain was calculated from the shift of the locations of the first-order laser beam spots measured with the PCD sensor. They found that neither the incident angle of laser nor the deformation of the grating affect the measured surface strain. Figure 12.11 schematically shows the experimental setup of the method.

Laser-acoustic-emission (laser-AE) technique: Enoki and Nishinoiri (2005) proposed an He–Ne laser-AE system to measure high-temperature (~ 773 K) AE signals during sintering of alumina ceramics and thermal spying of alumina powder on steel substrate to overcome the difficulties associated with conventional AE sensors, which typically do not perform well above 600 K.

Laser infrared photothermal radiometry (laser IR-PTR): Mandelis (2001) developed a laser IR-PTR using an argon-ion laser modulated by an acousto-optic modulator to direct the laser beam onto the specimen surface. The laser beam is broadened with a diffuser to attain a larger spot size for 1D analysis. The emitted IR radiation from the specimen surface is collected and focused onto the detector using two off-axis paraboloidal mirrors. Collected signals (amplitude and phase) are pre-amplified, sent to a lock-in amplifier, and recorded over a range of laser modulation frequencies for data analysis. Experimental work was reported on the surface roughness measurement of thermally processed carbon steels, thermal-barrier coatings, and industrial silicon wafers. Figure 12.12 shows the inspection scheme of laser IR-PTR.

Laser photoacoustic spectroscopy: van Neste et al. (2008) used a pulsed quantum cascade laser (QCL) to illuminate the surface of stainless steel specimens coated with varying amounts of chemical residue and a quartz crystal tuning fork (QCTF) to detect the scattered laser beam in developing a laser photoacoustic spectroscopy system.

Wireless power transmission using laser: Park et al. (2012) inspected aluminum plates using an Nd:YAG laser as power source to remotely excite ultrasonic transducers used for sensing, achieved by converting modulated optical power into electrical power.

12.9 Civil infrastructure applications

In civil engineering, laser-based sensing has shown its potential and practical value in a number of inspection and monitoring problems in bridges, buildings, tunnels, railways, roadways, and material characterization, due to its advantages in terms of (i) capability to access difficult to reach locations, (ii) non-invasive nature of interrogation, (iii) remote inspection, and (iv) relatively quick configuration. Selected applications are summarized as follows.

12.10 Laser safety

In laser-based sensing, potential hazard to the eye and skin from laser radiation are possible, depending on the laser wavelength, exposure duration, and viewing conditions. Specific safety measures must be obeyed, and safety procedures must be followed to avoid potential hazards, including:

Laser systems are categorized into five classes:

Exposure times or exposure limits (EL) are used to quantitatively evaluate the level of potential hazards imposed to human inspectors, from one nanosecond (ns) to 8 h and for wavelengths between 180 nm in the ultraviolet (UV) spectrum (at the end of the ultraviolet C band; 100–280 nm) to 1 mm in the extreme end of IR spectrum (at the edge of microwave spectrum).

For example, the maximum permissible exposure (MPE) for a CO₂ laser with wavelength 10.6 mm is 100 mW/cm² for 10 s to 8 h for limited area exposure and 10 mW/cm² for > 10 s for whole-body exposure. The MPE for a He–Ne laser with wavelength 632.8 nm is 3.2 mW/cm² for 0.1 s and 1.0 mW/cm² for 10 s. Other MPEs for various lasers can be found in ANSI Z136.1.

When using Class 3B and Class 4 lasers, eye protection is required wherever the exposure above the MPE of the laser occurs. ANSI Z136.1 provides general guidance for the use of eye protection devices.

12.11 Conclusion

Laser-based sensing has been a promising technology since its debut in 1960s and has been widely adopted in a variety of material characterization, damage/defect detection, and condition assessment applications in science and engineering. Owing to the flexibility of its energy level, the laser can be used as a tool for passive sensing or a source for active sensing, as well as either a nondestructive technique or a destructive technique. The integration of laser measurement with other imaging methodologies such as holography and tomography further enables laser-based sensing to be applied to the inspection problems that were previously considered difficult or impossible using laser and other sensing methods. This improvement on data visualization for laser-based sensing has fundamentally changed the face of laser-based sensing from a frequency-domain technique (spectroscopy) to a space domain technique. Although laser-based sensing basically measures the surface response of a target, it can be used to interrogate the subsurface region of a target when the optical energy carried by laser beams is converted to the mechanical energy to enable acoustic/ultrasonic waves propagating inward towards the target. When the surface response of a target can be successfully correlated to the subsurface condition of the target with the aid of theoretical and computational modeling, laser-based sensing can be used as an effective approach in some applications.

When applied in civil engineering applications, laser-based sensing is capable of accessing difficult to reach locations, non-invasive, remote, and relatively quick to set up in the field configuration. With all the promising technical capabilities possessed by the laser-based sensing techniques, laser safety remains a necessary concern in laboratory and in practice. Care must be taken when using high-power laser systems on large-size engineering structures (e.g., bridges, buildings, tunnels, railroads) for long inspection. Trade-offs between the safety and SNR (performance) of selected laser systems need to be assessed in the planning stage.