NDT.net • May 2005 • Vol. 10 No.5

Non-contact damage monitoring by laser AE technique

M. Enoki, S. Nishinoiria
Department of Materials Engineering, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8656, Japan

*Corresponding Author Contact:
Email: enoki@rme.mm.t.u-tokyo.ac.jp; phone 81 3 5841-7126; fax 81 3 5841-7181

© SPIE - The International Society of Optical Engineers.
This paper was originally published in the SPIE Proceedings vol. 5852
The 3rd Int' Conf' on Experimental Mechanics at 29.11.- 1.12.2004 in Singapore
Back To Session: Smart Structures and Non-Destructive Testing


ABSTRACT

AE method is a well-known technique for in-situ monitoring of fracture behavior by attaching piezoelectric transducer. However, conventional AE transducer cannot be used at elevated temperature or severe environment. Laser based ultrasonic (LBU) technique has been developed to characterize materials properties and detect flaws in materials. We developed the AE measurement system with laser interferometer to apply this technique to microfracture evaluation in various materials. AE during sintering of alumina ceramics and thermal spying of alumina powder on steel substrates were successfully measured by laser interferometers. The effect of processing parameters on AE behavior was clearly observed by analyzing AE waveforms. One of the most advantages of this laser AE technique is to estimate the temperature where microcracks are generated. These results could give a feed back to control processing conditions in order to avoid damage in materials. It was concluded that the laser AE method was very useful to detect microcracks in ceramics during fabrication.

Keywords: Acoustic emission, laser interferometer, microfracture, ceramics, coatings

1. INTRODUCTION

Recently, ceramics is extensively applied to structural materials because of its strength, abrasion resistance, chemical stability and so on, and ceramics components become larger and more complex. Ceramic components are produced by heating and densifing green compact at elevated temperature. These components are sometimes easily fractured during sintering due to large size and complex shape of ceramic components, so it becomes a problem that this fracture interrupts manufacturing process. It is important to control the crack initiation and propagation in green compact during sintering to resolve this problem. Furthermore, it is necessary to understand the evolution of mechanical properties during sintering to control the crack behavior [1, 2].

As the operation temperature of gas turbines has been rapidly increased to achieve high efficiency in recent years, materials are required improve their stability and mechanical properties at elevated temperature above 1773 K. Thermal barrier coating (TBC) with ceramic coating layer has been developed to shield heat from the outside and increase thermal stability of the surface [3]. As these coatings are subjected to both external stress and internal thermal stress constantly or periodically, the evaluation of mechanical properties and failure process of TBC at elevated temperature is desired to ensure the integrity. A health monitoring of structural components in service is important to assess the integrity of TBC, and a process control is also very important to provide uniform coating thickness and maximize the coating properties, as well as an estimation of the mechanical properties of coatings such as thermal shock resistance,thermal fatigue resistance and creep resistance .

Acoustic emission (AE) technique is a promising tool for reliability assessment of materials because it can monitor the generation and growth of microcrack in real time. However, conventional contact AE technique has a limit in application at elevated temperature, because a conventional piezoelectric transducer cannot be used above about 800 K. We have investigated the non-contact AE measurement technique using laser interferometer as a AE sensor [4-11]. This laser AE technique has several advantages such as non-contact measurement, absolute velocity measurement of AE signals, and applicability for severe environment. The purpose of this study is to investigate the influence of processing parameters on the generation and growth process of defects during fabrication process of ceramics by means of an in-situ monitoring system based on laser AE technique.

2. EXPERIMENTAL

2.1. In-situ monitoring during plasma spray coatings
Experimental setup of in-process monitoring system for plasma spraying is shown in Fig. 1. A sample was fixed to the stage equipped on a turntable in the direction perpendicular to the spraying direction. This coating process consists of three stages: First is a preheating of the substrate (stage-1), second is a spraying of topcoat (stage-2), and the last stage is an air cooling of the coated specimen (stage-3). After preheating of the surface by gun stroking without powder feeding, a feeding-rate control valve was opened to start the spraying of the topcoat. When the thickness of topcoat reached a desired value, the spraying apparatus was switched off and then the specimens were air-cooled until room temperature. After spraying, the sample stage was turned and the rear surface was faced to the laser beam of interferometer. AE signals during cooling period were detected using a heterodyne type interferometer (AT-0022, Graphtec Corp.). Acquisition of AE signals was started after the power source of the spraying robot, exhaust duct and compressors were switched-off to avoid the influence of mechanical or electromagnetic noise on AE signals.

Fig. 1. Schematic figure of in-process monitoring system using laser interferometer for plasma spraying.

Some dead time for AE measurement was required to adjust the focus of laser beam on the measuring surface. Total dead time before the start of an acquisition was about 1 min. Low noise type demodulator (AT-3600S, Graphtec Corp.) was used to measure an out-of-plane surface velocity on a sample with range of 1 mm/s/V. In order to reduce noise level, output signals were filtered with high pass filter (HPF) of 50 Hz and low pass filter (LPF) of 200 kHz. Detected AE waveforms were recorded by AE analyzer (DCM-140, JT-Toshi Corp.).

2.2. In-situ monitoring during sintering of alumina
Experimental setup for sintering of ceramics is shown in Fig. 2. Pre-treated alumina green compact, 35 by 35 by 20 mm, was heated by electrical furnace, and AE signals during sintering were measured by using laser interferometers. Deep notch was introduced into some samples to investigate the effect of stress distribution. Two types of sintering pattern with high and low shrinkage rate were also used to change the stress during sintering. Refractory was holed to introduce a laser beam and AE signals through SiC plate was detected by the above measuring system. Generally long waveguide is used to measure AE signals at elevated temperature in order to avoid the influence of heat on AE transducers. However, it is very difficult to characterize AE sources quantitatively because AE signals through waveguide are strongly affected by the shape of waveguide. AE signals detected by this setup have an advantage over the ones using waveguide.

Fig. 2. Schematic figure of in-process monitoring system using laser interferometer for sintering of ceramics.

3. RESULTS AND DISCUSSION

3.1. AE behavior during plasma spray coatings
Figure 3 shows the temperature history and AE behavior for the different thickness of bond coat, where the other processing parameters were common, that is, pre-heating temperature Tp was 773 K, thickness of top coat dTC was 1 mm, and gun velocity v was 0.1 m/s. In the sample without bond coat, only two AE with large amplitude were detected and immediately ceramic layer was delaminated. On the other hand, a large number of AE events with various amplitudes in the sample of thickness of bond coat dBC of 70 µm were detected in the temperature range of 200 K before the delamination of top coat.

Fig. 3. Relationship between temperature history and AE behavior for different thickness of bond coat, (a) dBC= 0 mm and (b) dBC= 70 µm.

Fig. 4. Relationship between the (a) amplitude and (b) cumulative AE event and generation temperature of AE for different traverse speed of spraying gun.

Figure 4 shows the relationship between the amplitude and generation temperature of AE for different traverse speed of gun, where dBC was 70 µm, Tp was 773 K and dTC was 1 mm. AE generation temperature TAE in the sample of 0.1 m/s was higher compared with that in the sample of 0.2 m/s. The number of AE events in the case of 0.2 m/s was lager than that of 0.1 m/s, and the temperature range during AE generation of 0.2 m/s sample was wider than that of 0.1 m/s. Many interlamellar microcracks were observed in top coat of coating samples, and delamination was found near the interface between top coat and bond coat. It is difficult to quantitatively characterize the damage induced sample during coating. However, non-contact measurement of AE behavior enables to estimate the damage during coating process.

AE signals detected during cooling process could be classified into two types by the peak frequency, that is, Type-A and Type-B, respectively. Type-A signals have a peak around 75 kHz, on the other hand Type-B signals have several characteristic peaks in 100-200 kHz. Type-A signals were especially detected before the final delamination of specimens, and Type-B signals were broadly observed during cooling process.

3.2. AE behavior during sintering of alumina
Figure 5 shows a typical AE behavior during firing. AE events were detected only in brittle region (below 900°C) during cooling. Within tested firing conditions, no AE signal was detected during heating or sintering period of firing. Figure 6 shows the change in cumulative AE events with generation temperature tested for various holding time at the maximum temperature. Generation temperature of first AE event increased with the increase in holding time regardless of heating rate. Cumulative AE events also increased with the increase in holding time. Although AE signals were detected in samples fired with heating rate of 800 K/h and holding time over 60 min, no surface crack was observed. Some of these samples cracked few days later.

Fig. 5. Typical AE behavior during sintering of alumina detected by laser AE technique.

Fig. 6. Cumulative AE events as a function of generation temperature, (a) results of mixed-grain samples with heating rate of 1600 K/h , (b) results of fine-grain samples with heating rate of 800 K/h.

Single-channel inverse analysis was applied to estimate the microcrack size using deconvolution method. In this study, the breaking of a pencil lead was used as the simulated AE signal source. With assumption of mode-I type penny-shaped cracking, the Green's function of the media was calculated from the simulated AE signal by the response waveform detected at the epicenter of the media and converted numerically to a dipole from the monopole source. The crack radius, a, was estimated as follows [12]:

     (1)

where D0 is the first peak of source function, s0 is the strength for microfracture and n is the Poisson's ratio, respectively. Crack radius was estimated by this equation. Properties used in calculation were that of fully-sintered alumina ( n= 0.23, s0 = 340 MPa).

Fig. 7. Crack radius as a function of generation temperature. Samples in (a),(c) and (b) were mixed grain and fine grain one, and were fired with heating rate of 1600 K/h for (a) and (b), 800 K/h for (c) and holding time of 30 min for (a) and 120 min for (b) and (c), respectively.

Figure 7 shows the relation between generation temperatures of AE signals and crack radii estimated by the inverse analysis of AE. As shown in Fig. 6(a), generation temperature of the first AE of notched sample was higher than that of smooth one. In the case of the notched sample, no AE signal was detected after generation of the microcrack radius of over 1 mm, while some AE events were detected in the case of the smooth sample. Smooth samples with holding time of 120 min as shown in Fig. 7(b) and (c) showed same behavior. However, larger number of AE events than in case of 30 min, especially relatively smaller microcracks were detected. Most frequent crack size determined from Fig. 7(b) and (c) was around 250 and 300 µm, respectively. In case of the sample in Fig. 7(c), no surface crack was observed after firing while other samples in Fig. 7 cracked. Figure 8 shows the relation between crack generation time and crack radii. It could be seen that microcracks with radius of 150-1500 µm and generation time of 1.5-5.5 µs were generated during firing and larger crack had relatively longer generation time.

Fig. 8. Crack radius as a function of generation time. Samples in (a),(c) and (b) were mixed grain and fine grain one, and were fired with heating rate of 1600 K/h for (a) and (b), 800 K/h for (c) and holding time of 30 min for (a) and 120 min for (b) and (c), respectively.

4. CONCLUSIONS

We developed a non-contact in-process monitoring system for ceramics and coatings with laser AE technique, and applied this system to the detection of microfracture during fabrication process. Conclusions of this study are as follows:
  1. Using the developed in-process monitoring system with laser AE technique, AE generated in cooling process of plasma spraying and alumina sintering was successfully detected.
  2. Processing conditions in plasma spraying affected AE behaviors such as generation temperature, number of AE and frequency characteristics.
  3. It was demonstrated that cracks with radius of 150-1500 µm and generation time of 1.5-5.5 µs occurred during sintering of alumina.

REFERENCES

  1. S. Harato, T. Mitsudome, J. Hasegawa, Y. Hara, H. Matsumoto, T. Nose and M. Sugawara, Nippon Steel Technical Report, 59 (1993) pp21-29
  2. N. Miyata, T. Shiogai, C. Yamagishi and Y. Matsuo, J. Ceram. Soc. Jpn., 106 (1998) pp494-499
  3. R. A. Miller: J. Thermal Spray Technol. 6 (1997) pp35-42
  4. M. Enoki, M. Watanabe, P. Chivavibul and T. Kishi: Sci. Technol. Adv. Mater. 1 (2000) pp157-165
  5. M. Watanabe, T. Okabe, M. Enoki and T. Kishi: Sci. Technol. Adv. Mater. 4 (2003) pp205-212
  6. M. Watanabe, M. Enoki and T. Kishi: Mater. Sci. Eng. A359 (2003) pp368-374
  7. S. Nishinoiri, M. Enoki and K. Tomita: Mater. Trans. 45 (2004) pp92-101
  8. S. Nishinoiri, M. Enoki, T. Mochizuki, H. Asanuma: Mater. Trans. 45 (2004) pp257-263
  9. S. Nishinoiri, K. Nozawa, M. Enoki: Mater. Sci. Forum, 449-4 (2004) pp717-720
  10. M. Enoki, S. Nishinoiri, T. Seki, A. Kishimoto: Key Eng. Mater. 270-273 (2004) pp485-490
  11. S. Nishinoiri, M. Enoki, T. Mochizuki, H. Asanuma: Key Eng. Mater. 270-273 (2004) pp1821-1826
  12. M. Enoki and T. Kishi: Int. J. Fracture, 38 (1988) pp295-310
© NDT.net |Top|