Picture Meeting – 14/01/10 – Cleaning of Painted Surfaces
The half-day Picture Meeting that took place on 14/01/10 was the third devoted to the subject of surface cleaning. The presentations developed some concepts and topics that had been discussed in the meetings of 17/01/2007 and 05/06/2008, and also introduced new research projects.
Ever since the initial findings were presented in 2007, the conservation community has eagerly anticipated the results of ICN’s Dry Cleaning project. Maude Daudin, a private conservator and ICN researcher, summarized them in the first presentation of the day: ‘Application of dry cleaning methods on unvarnished modern paintings.’ A variety of materials – erasers, mouldable materials, sponges and cloths – were tested on prepared paint surfaces, which included water-sensitive oil paints and slightly underbound gouache paints. The suitability of the dry cleaning methods were evaluated based on six criteria: preservation of topographical integrity (T), stability when artificially aged (Ag), clearance of particulate residues (Cr), preservation of surface gloss (G), preservation of paint layer aspect under UV light (UV), and cleaning efficiency (Cp). The results were presented in the form of star-shaped graphs with the criteria plotted on different axes.
The painted surfaces were examined before and cleaning using light microscopy, SEM-EDS, and in some cases with additional methods (Hirox microscope, confocal microscope and/or gloss meter). It was unfortunate that these additional methods were not used systematically, as they appeared to give a detailed impression of the topography and “polishing” of the paint surface. That being said, the before-and-after images produced by light microscopy and SEM were clear enough to make alarming observations: for example, some residues left by groom stick and Absorene on the gouache surface became very difficult to remove after ageing. It seemed that based on the aforementioned criteria, the most damaging dry cleaning products were Absorene, groom stick and PVC erasers. In practice, the selection of a dry cleaning material is case dependent, and is necessarily based on a compromise between all of the criteria. Combinations of several methods, and/or a gradual testing approach are sensible practical suggestions.
The day after the Picture Meeting, Maude Daudin gave a preliminary workshop on dry cleaning methods. The ICN will organize two-day workshops in early June, 2010. Please contact Angeniet.Boeve for more information.
A table of the dry cleaning products analysed and tested as part of the project was distributed as a two-page handout during the meeting, and is available on the internet at: http://www.scribd.com/doc/28005037/Painting-Dry-Cleaning-Table.
The second presentation ‘Scanning Electron Microscopy for surface studies’ was given by Ineke Joosten, a researcher at the ICN. SEM can be used to provide topographical and compositional information at very high magnifications. As more studies use SEM technology to illustrate the before- and after-effects of cleaning, is it important to be able to interpret the images that it generates. The possibilities were made clear by a side-by-side visual comparison of images produced in the various modes: Secondary Electron Image (SEI), Backscattered Electron Image (BEI) and Low Scatter Electron Image (LSEI). The voltage, the choice between low or high vacuum, and the scan rotation also have an effect on the image that is produced.
Usually, a sample that is to be examined with SEM would be coated with carbon or gold to prevent charging effects (bright white areas); however, in surface cleaning studies involving a “before and after” comparison, coating is not possible. Images of uncoated samples can be produced in low voltage/high vacuum mode, and/or by using the LSEI detector, which works at a slow scan speed.
Although the Dry Cleaning project officially ended in 2009, some of the research will continue within the ‘Semi-humid cleaning methods’ and ‘20th century oil paints’ projects. Beatriz Verissimo, an MA/Sc student from the New University of Lisbon, introduced the Semi-Humid Materials project. She plans to test cleaning products that can be slightly moistened, for example: makeup sponges, microfiber cloths, absorption sponges, blitz fix and cotton swabs. These materials will be tested on a modern painting and on prepared samples of oil paint surfaces. The methodology and evaluation parameters for the Semi-Humid Cleaning study are similar to those used in the Dry Cleaning project. One interesting difference is the method that the researchers used to soil the test panels. In order to best emulate a dirt layer with both particulate and greasy characteristics, Maude Daudin (Dry Cleaning project) collected particulate dirt, applied it to the paint surface with a microfiber cloth, then left it in a kitchen. Beatriz Verissimo sprayed the particulate-covered paint surface with a combination of greasy substances, then artificially aged it.
Another recent ICN project – ‘Analysis of metal soaps in 20th-century oil paints’ – was presented by Francesca Izzo, a PhD student from Ca’ Foscari University in Venice. In 20th-century paints, metal soaps can be reaction products (e.g. zinc soaps formed through the chemical interaction of zinc white and linseed oil), or additives (e.g. metal stearates added by the manufacturer as dispersion agents. Aluminium stearate additives were described in a presentation by Klaas Jan van den Berg at the Picture Meeting on 16/01/2009.) In her research project, Francesca Izzo plans to devise methods to detect, quantify and distinguish between added and formed stearates using techniques such as micro-FTIR and GC-MS.
We look forward to hearing the results of both projects at an upcoming Picture Meeting.
The following presentation, ‘Some cleaning issues in the Jan van Scorel triptych from the Grote Kerk in Breda’, was made by Annelies van Loon, who conducted the research and analysis in 2008 as part of her private practice. The large Finding of the Cross triptych, dated ca. 1540, is currently being treated by Menno Dooijes. One of the challenges encountered during treatment was whether or not to remove a disfiguring grayish layer that covered the sky. It remained after most of the varnish and overpaint had been removed, but remarkably it was not present where it had been protected by the frame and by old retouchings. Cleaning tests showed that it could be removed with (an unspecified concentration of) triammonium citrate. At that stage, it was unclear whether the grey layer was an added material or the result of paint deterioration. Cross-sections taken from cleaned and uncleaned areas were compared. No distinct pigment particles were visible in the grey layer, and SEM-EDS showed that it included dirt, as well as lead and some other elements. Annelies van Loon suggested that lead soaps from within the paint layer had formed a crust at the surface, and that these soaps had mineralized into other compounds. The grey layer was separated from the original paint by a thin film of organic material – possibly a varnish or a skin of oil medium. This implies that the fatty acids could have migrated through this separating layer.
Annelies van Loon gave some examples were given of other recently treated paintings where there was no clear separation between the paint and surface layer. In the treatment Finding of the Cross, the decision was ultimately made (in consultation with an advisory committee) to remove the grey layer. In response to the questions that were posed after the presentation, the ethical implications of this decision were discussed.
Although the next two presentations were not specifically related to Surface Cleaning, they nevertheless followed on from the previous talks by highlighting some of the practical decisions that conservators face. Both discussed paintings by Piet Mondriaan (1872-1944). In ‘Conservation and restoration of Mondriaan’s Compositie nr. III 1927’, Vera Blok presented the research and treatment that she undertook as a freelancer at the Stedelijk Museum in Amsterdam. She outlined some of the visual and technical characteristics of paintings from Mondriaan’s “neoplastische” period (1920-1931). She also described some problems with the condition of works from this period, primarily caused by Mondriaan’s choice of materials, his working methods, the layering of paint layers over one another, and cadmium degradation in the yellow and red areas. Three of the five neoplastische paintings owned by the Stedelijk Museum were lined in the 1950s. Compositie nr. III 1927 received a double lining and a loose lining, and the format of the painting was changed through the addition of wooden strips to the edges. In the recent treatment, Vera Blok took Mondriaan’s aesthetic and technical choices into account, especially with regard to the painting’s format, sharply creased edges and framing.
Ruth Hoppe followed this presentation with an introduction to the Mondriaan Project currently being conducted at the Gemeentemuseum Den Haag, in collaboration with Madeleine Bisschoff, Leonore Speldekamp and Lidwien Speleers. The Gemeentemuseum owns 162 paintings by Mondriaan, spanning his entire career. Recently, a team examined 159 of these paintings under magnification and UV light to assess the conservation and restoration needs of the collection. Their observations regarding all parts of the painting’s structure, and estimates for the time required for treatment, were entered into a computer using the AD-LIB system.
Ruth Hoppe described some of the condition problems that they encountered during their collection scan: broken stretchers; tacking margins weakened by corroded nails; surface dirt on the front and back; flaking, brittle and collapsing paint; and paint from lower layers coming up through cracks in upper layers. So far, sixteen painting have been treated, and another forty have been earmarked for urgent treatment.
Unfortunately, the last presentation mentioned in the programme had to be cancelled due to illness. We hope to hear Joris Dik’s presentation – ‘Deterioration of cadmium yellow in Ensor and Van Gogh’ – at a future meeting.
The next Picture Meeting will take place on Thursday, 15/04/10, and we welcome suggestions for contributions or topics (please email: firstname.lastname@example.org or email@example.com). We would like to thank Monique de Louwere for organizing the previous Picture Meetings, and welcome Angeniet Boeve, who has taken over the organizational role. We received a positive response to the webcast of the last meeting, which can be viewed at: http://webcast.icn.nl. Although the next few meetings will not be webcast, we hope to broadcast videos of the Picture Meetings live on the internet in the future.
Summary by Abbie Vandivere