Consolidants

Medieval plasterwork currently faces preservation issues due to the lack of cohesion between their polychrome sections, which, if not adequately treated in time, cause the irretrievable loss of their original appearance. It is relatively frequent to observe how the plasterwork preserve only scarce remains of their original color appearance in monuments like the Alhambra, the Alcazar in Seville or in pieces stored in Archeological Museums all over the world. The loss of vividness in the color of the plasterwork has been caused mostly by exposure to humidity, aggressive cleaning agents or by superposed paint layers applied at a later period. In view of these facts, the main aim of this study is to determine the best consolidant of choice to fix and preserve medieval plasterwork, from a colorimetric perspective. A set of 48 plaster probes have been built and the polychromy has been applied according to the procedures and materials used in the mediaeval period. Each probe was 6.5 x 13 cm in size, and three different natural pigments have been applied on 16 probes each (natural azurite, cinnabar and lead red) with two different binders (animal glue and arabic gum). Figure 1 shows an instance of one of the probe series for lead red. In the study, the methodology used to evaluate the color changes induced by six different consolidants is presented, and the results analyzed. The consolidants are: ethyl silicate (Estel 1200, BE), ethyl silicate (Nanoestel, NE), acrylic resin (Paraloid® B72, PAB72), hydroxypropyl cellulose (Klucel®G, KL), polyvinyl butyral (Mowital® B60H, MW) and thermoplastic polymer (Aquazol®, AQ). Two factors have been analyzed in the preliminary results: uniformity of the applied pigments across probes, and effect of applying the consolidant on the probes, in average across the different pigments. For the first time to our knowledge, a hyperspectral imaging system (Resonon PikaL camera) has been used to measure the spectral reflectance of the probes before and after applying the consolidants. Using the spectral reflectance data, CIEDE00 color difference and two additional spectral difference metrics (Root-Mean-Square-Error or RMSE and Goodness-of-Fit coefficient or GFC) have been computed between corresponding probes to analyze the two factors mentioned above. Besides, changes in perceptual attributes like Lightness, Chroma and Hue have been studied using the CIELAB color coordinates of the samples. Figure 1. Left: Lead Red probes prepared for spectral reflectance measurement with the PikaL system (center). Right: CIEDE00 results comparing the probes before and after the consolidants have been applied, in average for the three pigments used. The results show that the least homogeneous pigment across probes is azurite, due to the effect of the grind size on the pigment obtained and the difficulty of the binding process. Also, the least homogeneous binder is arabic gum. Regarding the effect of applying the consolidant, the Klucel® is the agent that best preserves sample color, and Mowital®B60H is the consolidant which causes the most noticeable change in color in average across pigments. The use of a hyperspectral capture device will enable us to analyze the effect of the consolidant agent in the local spatial texture of the sample, which poses an additional advantage over traditional area-based color measurement devices. This methodology is also potentially valid to perform colorimetric quality assessments of other preservation actions for plasterworks, like cleaning and color reintegration operations.