Nonlinear optical imaging techniques (NLO) for painting investigation

Nonlinear optical imaging techniques (NLO imaging), i.e. multiphoton excitation fluorescence (MPEF), second and third harmonic generation (SHG, THG), are high-resolution imaging modalities which may provide non-destructive determination of thickness and composition within multi-layer objects as a function of depth. NLO techniques have been recently introduced into artworks analyses, giving promising results. As regards paintings, the guiding idea is to exploit the low absorption in the near-infrared (NIR) of most materials, using a laser light in this spectral region which may generate non-linear interactions within materials. The objective is to make feasible the in-depth profiling of thin films on the basis of refractive index changes, variation of optical activity and presence of fluorophores. Such information is definitely useful for the analysis of painted objects and can be also crucial for the monitoring of restoring operations. In this study, we present the preliminary results obtained through the application of NLO imaging techniques on a series of single- and multi-layers systems simulating real egg-tempera wood panel paintings. This research is part of the wider Iperion CH Project, whose WP6 entails the evaluation of NLO-imaging to obtain highly resolved en-face and 3D images of small areas on paintings. To that purpose, a set of multi-layer painted samples, simulating real paintings, was designed and realized. They are currently under analysis with different operating systems and modalities within the partnership. In order to assess the reliability of NLO measurements, data were complemented with other well-established techniques, such as Fourier Transform Infrared Spectroscopy (FTIR) and X-Ray Fluorescence (XRF) for the chemical characterization of materials, Optical Coherence Tomography (OCT) for the determination of layers thickness and laser scanning micro-profilometry for the study of surface morphology. In addition, colorimetric analyses, i.e. reflectance spectroscopy in the visible region, were performed in order to chromatically characterize pigments and lakes. Finally, the transparency of the different materials was investigated by means of Scanning Multispectral VIS-NIR Reflectography. The preliminary results suggest that the 740 nm wavelength, while enabling the visualization of the painting surfaces micromorphology, is not effective for the determination of thickness through the detection of the non-linear signals.