By Francesca Piqué
The Getty Conservation Institute's Organic Materials in Wall Paintings project (OMWP), which began in 2003, was inspired by research undertaken at leading Italian conservation institutions, such as the Florence-based Opificio delle Pietre Dure (OPD), and by the work of well-known Italian wall paintings conservator Leonetto Tintori (1908–2000). The objective of the OMWP project has been to improve wall paintings conservation practice by enhancing the support that scientific investigations can provide by identifying organic materials in wall paintings.
Organic materials are particularly vulnerable during conservation intervention, and identifying them prior to treatment is critical. These materials deteriorate faster than inorganic components, and often only traces remain in centuries-old murals, making these organic components challenging to identify and conserve. Toaddress this problem, the GCI partnered with scientific research groups with differing expertise on a two-phase project that first assessed a group of investigation techniques on wall paintings replica samples and then applied the results of the assessments to ongoing conservation projects (see Conservation, vol. 20, no. 2).
DEVELOPING A METHODOLOGY
For its assessment of investigation techniques, the OMWP project used a set of wall paintings replicas made by Leonetto Tintori in his Laboratorio per Affresco di Vainella (see Conservation, vol. 18, no. 3). The replicas had been prepared with different types of binders and pigments applied on lime-based plaster. Each OMWP partner laboratory evaluated the potential and limitations of one or more analytical techniques on the Tintori replicas. The partners grouped the techniques to develop an investigation methodology that started with approaches not requiring sampling (noninvasive), followed by investigations that required removal of material (invasive).
Technologies are constantly improving and—significant for the study of wall paintings—laboratory instruments are becoming portable. Portable noninvasive methods are important in the study of wall paintings (which are typically vast and heterogeneous in nature and in condition), because they allow scientists to collect a great deal of data from a large surface without taking away any original material. In addition, when scientists are on site, direct dialogue is facilitated with the conservators regarding both selection of study area and preliminary interpretation of results. Sampling, then, can be limited to specific and characteristic situations representing many analogous areas.
THREE CASE STUDIES
In the OMWP project's second phase, three wall paintings cycles undergoing conservation were used as case studies to assess the project's methodology. This phase was done in collaboration with various sections of the Italian Ministry for Cultural Heritage.
The first case study, begun in 2005, was on the wall paintings cycle by Filippo Lippi (1406–1469) in the Cathedral of Prato (see Conservation, vol. 20, no. 2). The paintings, originally executed in a mixed fresco secco technique (very rich in organic materials applied over a fresco base), suffered from harsh and unguided cleaning interventions and heavy wax and soot deposits. As conservation intervention was already under way, only noninvasive investigative techniques were applied here. These techniques provided surface data that were difficult to interpret because of the extensive surface treatment performed on the painting in past interventions.
During work on the Lippi cycle, it became evident in discussion with head conservator Mark Gittins that the noninvasive tools had significant potential for assessing surface condition, for tracking its variation during intervention, and for post-treatment long-term monitoring. This potential was the focus of the second case study—the late-fourteenth-century wall paintings cycle Legend of the True Cross by Agnolo Gaddi (1350–1396) in the Alberti Chapel of the Santa Croce Church in Florence, which was the subject of a conservation program conducted by the OPD, the Opera di Santa Croce, and Kanazawa University, Japan. This large cycle, covering approximately 800 square meters (8,610 square feet) has a complex conservation history, having been previously treated several times. The OWMP project's noninvasive investigations confirmed the presence of an organic (protein-based) surface fixative (beverone), which was suspected by conservators. The challenge was to identify and protect the original organic material while the cleaning—aimed at removing the surface beverone following OPD's cleaning procedures—was undertaken.
Noninvasive measurements used on the Gaddi cycle included accurate technical photographic imaging, portable fiber optic mid-FTIR reflectance spectroscopy (mid-FITR), and portable fiber optic UV-Vis fluorescence spectroscopy (FOFS). Measurements were carried out before cleaning, after waterbased cleaning, and after ammonium bicarbonate cleaning of the paintings. Using mid-FITR, it was possible to follow the removal of alteration compounds such as sulfates and oxalates during cleaning and also to confirm the continued presence of proteins in specific paint layers applied a secco. The results were validated and compared with gas chromatography–mass spectrometry (GC-MS) characterization of the organic materials removed during cleaning and collected on cotton swabs.
This case study, carried out in collaboration with head conservator Mariarosa Lanfranchi, confirmed the capacity of noninvasive techniques in detecting changes in the composition and distribution of material on the surface of the paintings. Noninvasive methods potentially could be used not only to assess treatment but also to evaluate conditions over longer periods, thereby becoming a supporting tool for a long-term monitoring program.
The final case study was a seventeenth-century paintings cycle by Andrea Pozzo (1642–1709) in the Church of San Francis Xavier in Mondovi. Pozzo is well known not only as an artist but also for his treatise on architecture, with its appendix on how to create wall paintings. Pozzo's paintings in Mondovi are exceptional, not only for quality and beauty but also because they showed no obvious sign of previous intervention. Given the apparent authenticity of the material, these paintings provided a rare opportunity to understand the original technique and to use this knowledge to guide the intervention and to support a minimalistic approach to conservation that was completely respectful of the original.
Cleaning the paintings proved a challenge, even though they had only a limited amount of accumulated surface dust and grime. The investigations, carried out in collaboration with head conservator Mariano Cristellotti and the supervisor for the work from the Italian Superintendence, Walter Canavesio, confirmed that although Andrea Pozzo is well known as a fresco artist, he used organic material extensively in painting this cycle—evident by the UV-induced fluorescence of some areas.
In this case, the OMWP protocols made it possible to characterize the nature of the top layer, which fluoresced strongly, suggestive of the presence of organic material. Determining if this material were original or if it were a fixative applied in the past was an important objective, especially in those areas where it had discolored and a decision needed to be made regarding its removal. Based on the results of noninvasive testing, a specific set of samples was collected that was aimed at illuminating the painting technology. In cross section it was possible to identify the presence of a thin surface layer not containing inorganic components and presumably corresponding to the surface finish visible in UV-induced fluorescence.
The type of organic material present was studied by gas chromatography–mass spectrometry and by enzyme linked immunosorbant assay (ELISA).¹ The tests identified egg white used to apply a malachite layer, while whole egg was identified in other pigment applications. Carbohydrates have been identified in some of the transparent glazes, indicating the use of natural gums and/or organic pigments. The presence of carmic acid² suggests the use of cochineal, an organic pigment extracted from insects.
In this final case, the results of the OMWP investigations provided the professionals with scientific support in the decision on how to clean the surface. It was decided not to use reagent cleaning but, rather (and only where necessary), to use deionized water, while the majority of the surface was only dusted.
CREATING A BENEFICIAL DIALOGUE
The three OMWP project case studies are examples of in situ scientific investigation carried out according to a specific protocol. A beneficial dialogue was created among conservation professionals, an occurrence that is becoming more common thanks to the portable instruments. It is essential that scientific research be designed with the conservator and that it focus on actual conservation problems, that it is conducted with the most efficient techniques, and that its results are interpreted and evaluated by scientists along with conservators and art historians. This approach is now being applied at the Peruzzi and Bardi Chapels by Giotto in the Church of Santa Croce in Florence, in a project supported by the Getty Foundation. The project's research program is focused on understanding the conservation problems and the materials present in preparation for possible treatment, with investigations following the protocol advocated by the OMWP project.
Despite the fact that the case studies were different in terms of the original painting technology, physical history (such as previous cleanings), conservation interventions, and deterioration caused by natural and other environmental factors, the OMWP protocol allowed the conservators to develop better treatment plans and to be more fully informed about the effects of the interventions. The experience of working with an exceptional group of collaborators and with access to so many instruments demonstrates the enormous potential of integrated research.
The ability to make sound conservation decisions requires adequate resources in terms of both time and expertise—a situation that is, unfortunately, rare. Typically, access to wall paintings is simultaneous with the beginning of the conservation interventions, and there is little time available for investigations to be conducted and results to be interpreted so as to influence decisions about interventions.
The case studies confirm, through the before- and after-treatment investigations, how sensitive wall paintings can be to any direct intervention and how difficult it is to assess what is happening. In order to improve the preservation of wall paintings, it is important to have efficient and competent teamwork, good communication, and sufficient time for investigation before making decisions on treatment—including maintaining the option not to intervene directly on the paintings.
The most important contribution of the OMWP project is related to the noninvasive investigations and the assessment of the capacity and limitation of each investigation technique tested, particularly for new portable methods such as mid-FTIR. Another valuable contribution was bringing scientists to the site to work side by side with conservators to address conservation issues. The project made abundantly clear that there are no pure frescoes as such; they were executed in mixed techniques, and organic materials were always used. None of these great painters carried out the whole scheme exclusively in buon fresco— they always used, in different amounts, some organic materials. Clearly, thorough study of an artwork's surface before interventions are planned is crucial.
Francesca Piqué, formerly a project specialist with the GCI, is a researcher at the University of Applied Sciences and Arts in Lugano, Switzerland, and has been the coordinator of the OMWP project.
1. Tests were conducted by Gwenaelle Gautier and Perla Colombini at the
University of Pisa, and Joy Mazurek at the Getty Conservation Institute.
2. Identified by Ana Claro a visiting postdoctoral fellow at the GCI, using high-pressure liquid chromatography.