·Home ·Table of Contents ·Methods and Instrumentation

In depth Analysis and Characterisation by Photoacoustic Imagery Berquez L., Marty-Dessus D., Mousseigne M., Franceschi J.L. Laboratoire de Genie Electrique - Universite Paul Sabatier 118 route de Narbonne, 31062 Toulouse Cedex Contact

Abstract

Photoacoustic imagery is a non-destructive testing technique, which allows in- depth images in thick samples to be realised. With the aim of obtaining more information about the position of various defects into the sample, one must associate layer decomposition inside the object with the Zhang and Chen model. Then, it becomes possible to extract the photoacoustic signal at a given depth and to produce dierent images, each characteristic of a given layer. The work proposed shows the efficiency of our experimental method when in-depth analysis is required. Images got with various samples are shown and illustrate the techniques used for selective acoustic detection.

1 Introduction

An intensity modulated laser beam (figure 1) is focused on the surface of the sample and generates ultrasounds by thermoelastic effect beyond the thermal zone [1].

Fig 1: Experimental setup

The presence of defects into the sample added to a local change of the physical characteristics within the thermal zone give place to some variations of energy transported by the acoustic waves. These ones, detected by a piezoelectric sensor, produce a useful subsurface signal after processing by a lock-in amplifer. This signal is representative of a selected depth inside the studied sample. The depth selection can be obtained by two ways[2]: either varying the detection phase relative to that of laser beam [3], or modifying the modulation frequency of it. This assumption is based on the study of the Zhang and Chen model [4] which shows that, to change the phase detection, implies the possibility to modulate the contribution of each point of the thermal zone to the total photoacoustic signal. Moreover, the modification of the modulation frequency generates a widening or a reduction of the thermal zone size that's to say varies the depth of investigation [5].

2 Principle of depth profiling

Fig 2: Experimental setup

In order to qualitatively understand the mechanisms of depth profiling we used a simplified one-dimensional theory (figure 2) presented by Zhang and Chen [4]. So we can consider that each elementary thermal source situated at a given depth x^' within the sample produces an elementary voltage V (x ^' ) at the transducer output.

(1)

Where m_s is the thermal diffusion length.Thermal energy is converted into acoustic waves all along the path of propagation of these waves.Each signal V(x^') will have a magnitude and a phase depending on the beam modulation frequency, and in particular on the depth at which this elementary source is located. Because thermal generation vanishes with penetration in the sample, the 'in surface' generated signal will naturally be more important than those from the subsurface. To analyse an objet in depth, we must emphasise or amplify the signal from a depth x^'>0 which indicates the area of interest and suppress all other signal from interference. This is achieved by using a lock-in amplier where f₀ is the phase shift of the reference signal. The figure 3 shows the signal amplitude distribution versus x^' for different phase shift.

(2)

The total output voltage is represented by the integral of all these voltages. Thus, experimentally we can realise depth profiling by adjusting the phase shift.

(3)

Fig 3: Distribution of the outut signal V(x') versus x' for different phase shift settings

3 Photoacoustic extraction

In order to extract the photoacoustic signal generated at a given depth, we supposed that the sample is constituted by multiple layers and that each point of a same layer pro duces a elementary voltage . The magnitude of this elementary voltage is constant in a same layer but different between layers. So the photoacoustic signal can be expressed by:

(4)

To perform photoacoustic extraction,the modulation frequency is modified,so the contribution of each layer to the total photoacoustic signal is changed.
The different modulation frequencies F_reqi give the different thermal diffusion length m_i.Each thermal diffusion fixes the limit of a layer and the figure 4 depicts the geometry problem.

Fig 4: Geometry of the problem

For the different frequency the equation 4 become:

(5)

We can express that in a matrix form:

(6)

The magnitude of the photoacoustic signal generated on each layer can be obtained by inversion of the equations system 6.

4 Experimental result

In order to illustrate the efficiency of the method, we have observed a crack on an integrated circuit for different frequencies.
The images on figure 5 are a size of 256 m by 256 m with a resolution of 1 pixel by micron. The image 5a is a surface image; the others are subsurface images "in phase". The crack appears on subsurface images to frequencies of 1MHz, 250kHz and 65kHz. For others it has nearly disappeared. From these images, it is diffcult to quantify until where the crack is present. In order that we have applied our in depth extraction technique and we obtain images of the figure 6. One can then see that the crack is present only in a zone understood between 0 and 4 m. behind, the crack is quasi invisible.

Fig 5: (left)Photoacoustic images of a integrated circuit for various frequency Fig 6: (right) Photoacoustic images of a integrated circuit issued from various depth

Conclusion

Photoacoustic microscopy is very attractive for the non-destructive examination of materials, subsurface structure detection and depth profiling. We can see inside objets. With our technique, we can locate the defect into the objet and have a subsurface image from a know depth. We are hopeful that in the used model there are substantial simplications. More detailed theorical and experimental studies are in progress.

References

1. R.J. Von Gutfeld R.L. Melcher. Appl. Phys. Lett., 30:257, 1977.
   
2. G.F. Kirkbright M.J. Adams. Analyst, 102:281, 77.
   
3. J.L. Franceschi D. Marty-Dessus. Depth profiling by phase shift detection in scanning electron-acoustic microscopy. Electronics Letters, 29(10):843{844, 1993.
   
4. S.Y. Zhang L. Cheng. Photoacoustic microscopy (PAM) and detection of subsurface features of semiconductors devices. Ed Mandelis, elsevier edition, 1987.
   
5. A. Rosencwaig. Solid Stat Tech., (91), March 1982.
   

© AIPnD , created by NDT.net

|Home|    |Top|