Caas
Spectroscopy and Synchrotron Based Micro-Laminography
A. Schmidt-Ott (Technische Natuurwetenschapen, TU Delft) and J. Dik (Materials Science and Engineering, TU Delft)
Non-Destructive Characterization and Depth-Profiling of Drawings by Photoacoustic Spectroscopy and Synchrotron Based Micro-Laminography
There are a number of techniques capable of detecting structures and images below the surface of a painting or a drawing. However, none of these techniques are capable of seeing the colors buried under the surface. If buried colors are to be seen directly, this is not possible without the use of visible light. While visible light can penetrate quite deeply into most pieces of art in a manner that is completely non-destructive, the difficulty lies in obtaining a spectral response that also contains information about the depth beneath the surface. In principle, photoacoustic spectroscopy fulfills this requirement, and the first pilot experiments on double layers of paint on paper show that hidden layers can be detected.
- Fig. 1 Schematic setup of the photoacoustic spectrometer
The principle of photoacoustic spectroscopy is as follows: A light pulse of variable wavelength penetrates through the specimen. If the light pulse meets a color that absorbs it along its path through the specimen, there is a slight increase in temperature at this point. A temperature wave is created that travels to the surface, where it induces an acoustic wave. The time the temperature wave requires to reach the surface determines the depth of the absorbing color. Thus the delay with which a microphone receives the pulse determines this depth and the intensity of this acoustic signal determines the degree of absorption. By measuring the degree of absorption as function of the wavelength for a specific depth (i.e. a specific delay), the color at that depth can be determined. Rather than using single light pulses, a practically feasible set-up uses light that is chopped in a variable frequency. The acoustic intensity is measured as function of this frequency and as a function of the wavelength of the light. If the wavelength of a light beam is varied in addition to the chopping frequency and if the picture is scanned, this should reveal the colors of hidden layers at any depth of the sample.
- Fig. 2 Magnitude of acoustic signal as function of chopper frequency for double paint layers on paper: The shaded areas indicate the sections where most light is absorbed. For the green-green case (green curve) this happens far from the surface, and for the red-green case this happens close to the surface. The different shapes of the curves clearly indicate this
In our first experiment we used a monochromatic beam from a laser diode. This beam was passed through a double layer of acrylic paint on paper as schematically shown in the magnified cross section of fig. 2. Below this cross section a plot of the acoustic intensity vs. the chopper frequency is shown. The hidden layer of paint is green in one case and red in the other case. This layer is sandwiched between the paper and a surface paint layer, which is green in both cases. Expressed in a strongly simplified way, the frequency is a measure of depth (low frequency indicating large depth). The plot shows that the frequency profile of the acoustic signal is significantly different for the cases of a green and a red buried layer. The green curve, corresponding to the green buried layer, shows a much higher absorption at the depth of this layer compared to the depth of the surface layer. This is because the laser beam is absorbed more strongly by the green paint than by the red one. The experiment clearly shows the feasibility of the method. If the wavelength of the light could be varied over the whole visible spectrum, the absorption spectrum for each arbitrary depth could be determined, and the absorption spectrum precisely defines the color. This enables color tomography.