Material Property/State Characterization by Laser Speckle Photometry

Size: px

Start display at page:

Download "Material Property/State Characterization by Laser Speckle Photometry"

Louise Lawrence
6 years ago
Views:

1 18th World Conference on Nondestructive Testing, April 2012, Durban, South Africa Material Property/State Characterization by Laser Speckle Photometry Ulana CIKALOVA, Beatrice BENDJUS, and Juergen SCHREIBER Fraunhofer Institute for Nondestructive Testing, IZFP Dresden Branch, Dresden, Germany; Abstract: Variations of material conditions have to be tested continuously. For this goal, an optical method to determine a parameter that correlates to the coefficient of temperature diffusivity, called Laser Speckle Photometry (LSP), was developed. Thermal diffusivity is a property of conductive materials that can by be used for the characterization of micro-structural material properties, such as hardness, porosity, and/or the fatigued state. Laser Speckle Photometry is based on the detection and analysis of thermally activated characteristic speckle dynamics with high temporal resolution. The parameter of the thermal diffusivity, called speckle diffusivity, is determined using a correlation function from the pixel intensity of the speckle image variations during the thermal heating or cooling process. A direct correlation between the speckle diffusivity and material properties/state was found. The results of several investigations of porous metallic and ceramic materials are presented. In addition, results of successful evaluations of the damage state of construction steel are also presented. It is shown that the laser speckle photometry is a suitable approach for nondestructive characterization and monitoring of materials properties. Keywords: Speckle, Thermal Diffusivity, Nondestructive Testing, Condition Monitoring, Porosity, Hardness, Stress, Fatigue, Fractal Dimension 1. Introduction There is a need for industry to improve the quality control of their products by introducing new measurement systems/techniques to allow for nondestructive, contactless measurements as an alternative to presently used destructive inspection methods. In contrast, novel mathematical tools and the fast progress into hardware development allow the detection and processing of timeresolved optical changes and, therefore, opens the opportunity to develop new methods that are based on the processing of such time-resolved optical changes. The local short thermal excitation caused the thermal deformation. This thermal deformations cause non-stationary sample surface changes, which can be detected in optical 2-D images by high-speed CCD cameras. The evaluation of images taken this way provides information not only about the thermal sample deformation, but also about the thermal diffusivity of the material. Available literature presents a large number of possibilities for a non-contact determination of hardness [1-4], mainly by infrared radiometry and other methods where the relationship between the thermal diffusivity and material properties is used. Most of these methods apply to bulk, isotopic, homogeneous materials. These measurements [5] present the evaluation of material porosity of ceramic materials using photo-thermal methods, especially the mirage effect and thermo-reflectance microscopy in a similar way. The authors of [6] - describe the so-called Flash Method based on the excitation of a non-stationary thermal field with the flash pulse of laser or other pulse source for the registration of the time depended thermal radiation intensity. The analysis of this dependence allows determination of the thermal diffusivity, heat capacity, and thermal conductivity parameters. However, other photo-thermal methods can also be used for the characterization of material properties. The authors in [7] determined the thermal diffusion parameter by measuring the average speed of the speckle-field shift and distortion at the specimen surface. Their approach consisted of correlation functions describing the degree of

2 change in the speckle-image of same elements on the surface during the heating or cooling process. This approach has been applied for metal specimens and thin layer ceramic materials modified with the nano-diamonds. Numerous example patents present the speckle photometry method for deformation and elastic or plastic strain measurements. Using this process, the shifting of single speckles provide information about looking deformation quantity. [8-11]. Our department for Testing and Diagnosis Methods at the Fraunhofer Institute for Nondestructive Testing, Dresden Branch (IZFP-D) developed in the last year an innovative measurement technique to detect the variability of material properties, called Laser Speckle Photometry (LSP) [12]. This method was used to examine non-optical transparent homogenous and heterogeneous metals, primarily for the evaluation of hardness and porosity. The corresponding evaluation algorithms are the basis for the correlation functions and to determine the parameter specklediffusivity by thermo diffusivity laws, and the thermal diffusion equation, respectively. In [13], it was found that a direct correlation existed between the speckle-diffusivity and internal material properties, such as porosity of metal cellular material. The thermal excitation in this case was done using flame and induction heating. Problems with theses types of heating were: random disturbances such as thermal convection during flame heating and displacement of the samples at magnetic field of induction heating, which had an influence on the measurement process of the speckles movement. The evaluation algorithm first computed the differential auto-correlation function; then the temperature in the thermal diffusion equation was replaced by the timeresolved correlation function. The computing process was time-consuming. Considering the problems discussed, flame and induction heating were replaced in subsequent experiments by pulse-laser heating. Thereby, the influence of thermal convection and sample displacement in the magnetic field was eliminated. Additionally, the evaluation algorithm was simplified as the thermal diffusivity of the material was determined directly from the modified correlation functions. This paper shows the results of the correlation between hardness and porosity of materials and the speckle thermal diffusivity. Moreover, the damage state of steel was successfully determined from the slope of the same correlation function. 2. Experiments 2.1 Experimental setup The experimental setup and the process chart of the LSP are shown in Figure 1. For the thermal excitation of the sample surface, a pulsed heating source (Nd: YAG laser pulse KLS246 of LASAG LLBK45) was used; laser properties: wavelength, 1064nm; pulse energy, 50J; pulse width 20ms. A measurement point and temperature detection of the sample at the measurement point was placed at a distance of 0.5mm from the heating point. The temperature measurement allows to record 25 temperature readings per second. A continuous wave He-Ne-laser with a wavelength of 633nm was used for sample irradiation and activation of speckle patterns at the measurement point. A fast CCD camera detected the speckle movement during head propagation at the measurement point with the frame rate of 200 to 500 images per second.

3 2.2 Analysis algorithm Figure 1. Experimental setup of LSP and the process chart of the LSP measurement Caused by short laser pulse energy, E [J/m 2 ] results as the time-dependent temperature distribution 2 E x T(r,t) = T0 + exp (1) π α t ρ c 4 α t p at the sample surface. Thereby, ρ is the density, c p - specific heat capacity, α - the thermal diffusivity, x - the distance from the excitation point to the measurement point, T o - the initial temperature of the solid. The temperature curve shows every point on the sample surface at temperature maximum at the time: t T max 2 x =. (2) 6 α The thermal diffusivity a can be determined by equation (2) [14]. The time t T also can be determined from distortions of the speckle images. In this case the max speckles shift is determined by the help of the correlation function: n 1 max 2 C(i, j,t) = S(i, j,n+ τ) S(i, j,t) dt of C (τ) = S(n + τ,i, j) S(n,i, j) (3) n max n= 1 providing decisive parameters. RMS is the root mean square of time series, n max is the number of video frames and gives the time dependence, S (i, j, n) is the gray value of the speckle-intensity signal in the n th image, where i and j are the coordinates of the speckle patterns pixel.τ is the time shift. The highest shift in the speckle-images leads to the maxima of this correlation function

4 [15]. The correlation function maximum time t C max equals t T. Using the temperature max measurement at the measurement point, the thermal diffusivity α t was calculated by equation (2); using the correlation - the speckle diffusivity α Corr. For the determination of the damage state of the material, the initial slope of the modified equation (3) was used. n max i max j 1 max q C (τ) = S(n + τ,i, j) S(n,i, j) (4) nmax n= 1 i= 1 j= 1 The autocorrelation function (4) was qualified by the practical application using the determination of scale behavior of the speckles images. <...> is the averaging over all measurement times t for a fixed value of time difference τ. The exponent q gives the permission to separate the random perturbation of measurement technique. In the case of fractal behavior from equation (4) follows: C(τ) q H(q ) τ. (5) The parameter fractal dimension is calculated as follows: D F (q) = 2 - H(q) (6) The parameter D F is a quantitative measure for the damaged state of material [16, 17]. 2.3 Samples In this work, the investigations were carried out for calibtating determined photo-thermal parameters from speckle dynamics to hardness, porosity and damage state of materials. For the hardness calibration, steel sample St 37 was used, see Figure 2. The four hardness grades along the sample surface were achieved by surface treatment with varying degrees of hammer blows. Figure 2. Hardness specimen of steel St 37; decrease of the hardness from area 1 to 4 The evaluation of the material porosity was done on cellular metal. The material was a set of cellular metal foam samples with the composition of Fe + 20 % Cr + 5 % Al (material number ) manufactured by arc spraying, followed by rolling to multiple structures with the goal of a large range of porosity (38% - 82%). The non-rolled material had a porosity of 90%. The specimen width was 20mm with a thickness from 16mm to 3.5mm. Figure 3 shows the two examples: sample on the left has highest porosity; on the right the lower porosity.

state of the material was evaluated at the so-called Compact Tension (CT) sample of 22NiMoCr37 steel alloy, which was cycle loaded, and as a consequence, an incipient crack of 18mm length was

5 Figure 3. Basic non-roller sample cellular metal sample (Fe + 20 % Cr + 5 % Al) manufactured by arc spraying (left) and the samples manufactured by rolling to multiple structures (right) The damage fatigue state of the material was evaluated at the so-called Compact Tension (CT) sample of 22NiMoCr37 steel alloy, which was cycle loaded, and as a consequence, an incipient crack of 18mm length was generated. In previous works [16, 17], the electro-magnetic measurement of Barkhausen Noise (BN) for the damage state evaluation was done at this sample. The BN measurement was performed at every raster point. The sample shows the decrease of fatigue condition with the distance to the notch. Figure 4 shows the used CT-sample with the marked raster. Figure 4. Compact tension sample of 22NiMoCr37 steel marked by a raster with about 10mm step width. The arrow shows the incipient crack caused by cyclic load 3. Results Results of the experiments with hardness determination are shown in Figure 5. It is clear that the thermal diffusivity α t and speckle diffusivity α Corr decrease with increasing hardness. The tendency of the diffusivity parameter for both is the same, but the absolute value is different. This effect is probably caused by the lesser frame rate of the radiation temperature measurement. The exact measurement of the t was problematic. Tmax

6 Figure 5. The thermal diffusivity α T determined from the temperature measurement and speckle diffusivity α Corr as a function of the hardness of the St 37 sample LSP measurement results to determine porosity at cellular metals is presented in Figure 6. Except for results of the material with the less porosity (38%), the thermal diffusivity α t and speckle diffusivity α Corr decreases with increasing porosity. Probably the manufacturing rolling process caused the loss of material continuity at the higher rolling plasticisation, which then caused the decrease of the thermal diffusivity of the material. Figure 6. The thermal diffusivity α T determined from the temperature measurement and the speckle diffusivity α Corr as a function of the porosity of metallic cellular material (Fe + 20 % Cr + 5 % Al) This paper explored the option of using LSP signals by applying the D F parameter to detect the fatigue with regard to the damage state of the material. The experimental results achieved at the fatigued CT sample are shown in Figure 7. It was found that the D F parameter increases with the fatigued state of the material. The resulting dependency of the fractal dimension as a function of measurement area looks quite similar to the behavior shown by the evaluation of the fatigue state by the Barkhausen Noise technique presented in [16, 17]. In addition, the result of the LSP

7 measurement by applying induction heating as the excited heating source is presented here. The tendency of all measurements is the same. Figure7. Results of the damage evaluation: the fractal dimension D F represented as a function of the measurement area at the CT sample, see Figure 4. For comparison, the results of the damage state evaluation using Barkhausen Noise technique [16, 17], and the result by the induction heating as the heating source are presented. 4. Conclusion The presented Laser Speckle Photometry (LSP) technique is a newly developed non-contact, nondestructive testing technique based on the detection and analysis of thermal or mechanical activated characteristic speckle dynamics. LSP uses the parameter of speckle diffusivity α for the investigation of structural material properties. Typically, this parameter is determined using the correlation function, which describes the degree of change in speckle images during heating and cooling processes and the thermal diffusion equation. In comparison to the previous works cited in this report, the thermal sample activation was substituted by the pulsed laser heating source. Thereby, the influence of thermal convection and sample displacement in the magnetic field was minimized, compared to the use of induction or flame heating. The additionally presented simplification of the evaluation method leads to the reduction of measurement time. The speckle diffusivity α Corr was determined directly from the correlation function by this modified LSP method, and was successfully used for the evaluation of hardness and porosity of the material. Moreover, the damage state of steel was determined from the slope of the same correlation function. This optical method was applied for investigations of compact and cellular materials. Acknowledgment The author thanks the colleagues of LOD Minsk, especially Mr. N. Khilo, for the ideas given and many thanks the colleagues of IZFP Dresden, Mr. S. Naumann and R. Mueller for their help with the realization and evaluation of the experiments.

8 References 1. H G Walther, D Fournier, J C Krapez, M Luukkala, B Schmitz, C Sibilia, H Stamm, J Thoen, Photothermal Steel hardness Measurements- Results and Perspectives, Vol. 17 Special Issue, pp , D Fournier, J P roger, A Bellouati, C Boue, H Stamm, F Lakestani, Correlation between hardness and thermal diffusivity, Analytical Sciences,, Vol. 17 Special Issue, pp , J Jaarrinen, A Lehto, M Luukkala, Photothermal measurement of the thickness of diffusion hardened surface layers in steel, Ultrasonics Symposium, pp , M Bertollotti, R Li Voti, G L Liakhou, S Paoloni, C Sibilia, Thermal characterization of nonhomogeneous materials, Analytical Sciences, Vol. 17 Special Issue, pp 85-88, A Sanchez-Lavega, A Salazar, A Ocariz, L Pottier, E Gomez, L M Villar, E Macho, Thermal diffusivity measurement in porous ceramics, Appl. Phys., Vol. A 65, pp 15-22, W J Parker, R J Jenkins, C P Butler, G L Abott, Flash Method of Determining Thermal diffusivity, Heat Capacity, and thermal Conductuvity, Appl. Phys., 32, pp , N Mukhurov, A Maschenko, N Khilo, and P Ropot, A speckle-photometry method of measurement of thermal diffusion coefficient of porous anodic alumina structures intended for optical sensing, Proceedings SPIE 8073, 80731B Prague, Czech Republic, T Mitsu, Small displacement measuring method and instruments, Patent EP A1, T Satoruo, Deformation measurement method and device using electronic speckle interferometry, Patent EP A1, Japan: , Korea: , US 6,943,870 B2, US A1, Y Sanichirou, Method and Apparatus for extracting phase from speckle image of objekt to be inspected, Patent Japan: A, A, D Know, ESPI (Electronic Speckle Pattern Inferometry) system using high-magnification lens and zoom lens and method for measuring deformation of microelement using ESPI system, Patent Korea: A, U Cikalova, B Bendjus, J Schreiber, Laser-Speckle-Photometry A Method for Non- ContactEvaluation of Material Damage,Hardness and Porosity, will be published in MP Materials Testing, 02/ U Cikalova, B Bendjus, J Schreiber, N Khilo, Application of Contactless and fast Laser- Speckle-Photometry Technique for Materials Characterization and Quality Control of Cellular Materials, Proceedings DGM Conference Cellular Materials CELLMAT 2010, Dresden, pp Supplementary sheet, October B Buchmeyr, Meatrial and Produstion Technics with Mathcad: Modellion und Simulation in Application field, Springer Berlin Heidelberg; Edition: 1, 11. April V Belyi, N Kazak, N Khilo, M Kroening, J Schreiber, International scientific laboratory for laser Diagnostics in service for non-destructive material testing, MP Materials testing, Volume 51/5, pp , J Schreiber, U Cikalova, and Ye Vertyagina, Use Of The Fractal Nature of Spatial and Temporal Response Behavior for Materials Damage Characterization, Proceedings of the Sixth International Conference on Low Cycle Fatigue Conference, P.D. Portella, T. Beck, and M. Okazaki eds., DVM, Berlin, Special Issue, pp 655, 2008.

9 17 U Cikalova, B Bendjus, J Schreiber, Bewertung des Spannungszustandes und der Materialschädigung von Komponenten industrieller Anlagen, Materials Testing, Volume 51/10,pp , 2009.

Crack Detection of Ceramics based on Laser Speckle Photometry

J. Ceram. Sci. Tech., 08 [01] 73-80 (2017) DOI: 10.4416/JCST2016-00090 available online at: http://www.ceramic-science.com 2017 Göller Verlag Crack Detection of Ceramics based on Laser Speckle Photometry