School of Engineering and Physical Sciences    Heriot-Watt University

Applied Optics and Photonics Research

Home    People    Links    Sensors    HPLA Research    Optical Diagnostics

 

NDT Ceramics

Mateusz Matysiak, Duncan Hand, Jon Shephard

Overview

The objectives of this project are to develop optical techniques that can identify flaws in a bulk ceramic sample which are sufficient to have a significant impact on its mechanical properties. One application would be in the fabrication of dental crowns and bridges. An ultimate aim is to develop a fully automated non‑contact system that could rapidly determine, non-destructively, the extent of flaws introduced during a fabrication process. This should be a non-contact, non-destructive technique that can be applied with the part in-situ without the need to “remove-inspect-relocate” during processing, hence reducing as much as possible the impact on the overall fabrication time.

Results using visible light

  • Sample inspected containing mm sized laser drilled holes
  • Significant changes of the light structure can be easily detected.
  • Applying visible light for the detection of bulk flaws and features in Zirconia works to some degree in that a change in the transmitted light indicates the presence of flaws.
  • It is not possible to determine the position and size of even mm-scale features.
  • Unlikely that it could be used to detect µm-scale defects due to the high degree of scattering.

Figure 1 a) Specimen tested with visible light, b) Intensity of light transmitted through sample without buried features, and c) False colour gray scale image of light transmitted through sample – LED illuminates the central part of the specimen

Mid infrared transmission imaging (MIR-TI) technique

FT-IR measurements carried out on Zirconia samples showed that the Mid-Infrared wavelength range has potential for optical, non-destructive inspection of Y-TZP components due to a “transmission window” existing between 3 and 7 μm. 

Figure 2 Set-up for mid-infrared transmission imaging technique

  • The smallest feature which could be observed was 50 μm.
  • Results for the Mid-Infrared system are significantly better compared to VIS technique.
  • All the drilled holes are visible.
  • Possible to precisely define all the features, for example depth and distance from the observed surface of each of the holes.

Figure 3 MIR-TI result of sample shown in figure 1 – picture consists of two combined images

  • Investigated the ability to detect mm scale cracks and flaws
  • Flaws in the surface facing away from the camera were detected.
  • ESEM images confirm that these features are indeed cracks, but with air gaps in the region of only 1 μm.
  • In the Mid Infrared image, these cracks appeared much larger due to shadowing.
  • Cracks actually propagate deeper in the material and the MIR TI system with a depth of field of 115 μm provides truly 3D information about an object rather than just imaging its surface.
  • MIR TI technique not only provides precise position of the crack, but also the crack shape can be established.

 Figure 4 Laser drilled sample and Figure 5 Flaws detected with MIR-TI technique; a) MIR-TI image; b, c) ESEM images

Conclusions

  • With ongoing development we are exploring the possibilities for the MIR-TI technique to become a reliable and robust technique suitable for a manufacturing environment.
  • There is also a drive to make a fully automated system without the need for a human operator.
  • This is potentially useful for a wide range of ceramic based applications, particularly biomedical.
  • Already we have demonstrated that the Mid-Infrared transmission imaging technique can be used for inspection of thick sections of Y-TZP material.
  • An optical transparency window in the Mid-Infrared wavelength region is exploited, resulting in much clearer images of flaws in comparison with visible light imaging.
  • The presented approach can be applied to test specimens up to 6 mm thick, considerably thicker than other techniques currently used to non-destructively examine Y-TZP.

 

 

 

 Contact us AOP@hw.ac.uk