Smithsonian American Art Museum Lunder Conservation Center Analytical Test Results and Treatment Report Artist: Tom Wesselmann Title: Still Life #12 Accession #: 1986.23 Date: 1962 Materials: oil and acrylic paints, fabric, paper, metal, photomechanical images printed on paper, fiberboard Dimensions: 48 x 48 x 3/8 in. (122 x 122 x 1 cm) Requested By: Ann Creager, Paintings Conservator Examiner: Sharra Grow, Graduate Intern Supervisor: Ann Creager, Head of Conservation at SAAM; Amber Kerr-Allison, Kress Fellow Contributors: Sharra Grow, Graduate Intern (SAAM); Jia-Sun Tsang, Senior Paintings Conservator (MCI); Rebecca Gieseking, Paintings Conservation Intern (MCI); Nicole Little, Conservation Scientist (MCI); Figure 1: Still Life #12, normal light, before treatment Suzanne Lomax, Senior Scientist (NGA); Melvin Wachowiak, Senior Conservator (MCI); and Judy Watson, Conservation Scientist (MCI) Report Date: August 1, 2008 Analytical Test Results Polarized Light Microscopy The colorant in the red paint sample appears to be a lake pigment. A McCrone pigment sample set was used for comparison. No more conclusive results could be made using polarized light microscopy. Cross-sectional Microscopy This cross section was taken from Sample 2 (see fig.20). The paint layering on the red of the table cloth shows white layers under the top red paint layer and cellulosic fibers below the white layers (see figs. 11, 12). Three distinct white layers can be seen underneath the red paint layer when the sample is viewed under ultraviolet (UV) light (see fig. 13). The presence of the cellulose fibers underneath the lowest white layer confirms that this is the ground or priming layer on the fiberboard support. The ground contains large and coarse particles in comparison with the two white layers and red layer above, which contain very small and evenly ground pigments. 1
Scanning Electron Microscopy Energy Dispersive Spectroscopy (SEM-EDS) This analysis negates the hypothesis that the red paint contained cadmium-based pigments, as no cadmium was detected. In fact, there were no heavy elements found in the red paint layer except for chlorine which is likely an artifact of the embedding materials (see figs. 18, 19). It is believed that the red paint layer contains an organic dye or lake pigment, though the presence of chlorine is not yet understood. The presence of silicon in the ground layer may suggest glass or sand as a coarse filler (see figs. 14,15). Calcium is also found in the ground layer, which could be from calcium carbonate, often found in ground and priming layers (see fig. 17). A significant amount of titanium was detected in all three white layers suggesting that titanium white is a pigment in these paints (see fig. 16). Fourier Transfer Infrared- Attenuated Total Reflection Spectroscopy (FTIR-ATR) FTIR-ATR spectrum was obtained from efflorescence taken from Sample 3 and the surrounding area (see fig. 20). After comparison with known spectra, the spectrum from the efflorescence sample indicates the presence of free fatty acids, likely palmitic acid or stearic acid. The two sharp peaks at 2850 cm -1 and 2920 cm -1 indicate straight hydrocarbon chains like those in fatty acids. The peak at 1700 cm -1 indicates carbon-oxygen double bonds and the broad peak between 2500-3500 cm -1 is caused by oxygen-hydrogen bonds, indicating the presence of carboxylic acid groups which are found on fatty acids. FTIR spectra were also obtained for the ground and the white and red layers used to paint the tablecloth (see fig 21, 22). The ground appears to be acrylic, and both paint layers appear to be oil. The spectra of oil-based and acrylic-based paints differ significantly in the carbon-hydrogen region just below 3000 cm -1. While oil paints have two distinct peaks in this region (indicating the long hydrocarbon chains in the oils), acrylic paints have a single broad peak caused by the overlap of several smaller peaks. X-Ray Diffraction (XRD) XRD performed on efflorescence resulted in spectra indicating the presence of n-paraffin primarily, with trace amounts of two forms of palmitic acid (Perhydrotriphenylene palmitic acid and α-palmitic acid) (see fig. 23). Although initial examination of the XRD spectra indicated the presence of lead silicate, the lack of any other lead or silicate peaks does not support this possibility. Stearic acid was not found in the sample, indicating that palmitic acid is the only fatty acid present. Gass Chromatography/Mass Spectroscopy (GC/MS) GC/MS performed on a sample of the efflorescence resulted in a spectrum which included primary peaks from methyl palmitate and methyl stearate, with a P/S ration of 4.40 (see fig. 24). Small amounts of C-14, C-15, C-17 and C-20 fatty acids were also found to be presence. It is not possible from these results to determine whether the initial species were fatty acid salts or free fatty acids. Samples Taken Sample 1: Red paint from painted fabric at lower PR corner Sample 2: Red paint from checkered tablecloth taken from lower edge 2
Sample 3: Efflorescence crystals taken from the red paint of the checkered tablecloth at lower edge Sample 4: Red paint from the vertical collage strip on the PR edge Sample 5: Coated paper from the upper PR corner Sample 6: Red paint from the top edge of the red apple Sample 7: Red paint pigment from checkered tablecloth taken from PR lower edge (see fig. 11) Sample 8: Fiberboard from the lower PL corner Sample 9: Efflorescence crystals taken from red paint of the checkered table cloth near the lower edge (taken at Jia-Sun Tsang s request) Sample 10: White paint on the checkered table cloth near the lower PR edge (taken at Jia-Sun Tsang s request) Sample 11: Ground layer taken from the lower edge (taken at Jia-Sun Tsang s request) Conclusions and Further Research The red paint is most likely an organic lake pigment and the three white layers beneath it contain a significant amount of titanium white. The red paint layer and the white paint layer immediately beneath are oil based and the ground is acrylic based. Results from FTIR-ATR, XRD, and GC-MS confirm the presence of free palmitic acid as a major component and paraffin as a minor component. The observation that this fatty acid is able to form solid crystals on the painting further supports that it is likely palmitic acid, being a saturated fatty acid. Unsaturated fatty acids, such as oleic acid, tend to be liquid at room temperature and therefore would not form solid crystals on a painting in a museum environment. Additionally, unsaturated fatty acids tend to crosslink with each other, while saturated fatty acids are unable to crosslink because of their lack of carbon-carbon double bonds, making the saturated fatty acids more likely to migrate to the surface of the painting. One important question remaining is what instigated the migration of the fatty acids and surface crystal formation? It could have to do with polymorphic transformations these compounds are able to make. Polymorphism is the occurrence of several different crystal forms from the same chemical compound. For example, calcium carbonate is dimorphous (having two possible crystal forms), crystallizing as calcite or aragonite. Saturated triglycerides, which may be found in low quality or slow drying oil paints that Wesselmann may have used, can transform and assume three different crystals alpha, beta prime, and beta. It appears that the beta form of the crystal best matches the visual observation of the white crystals found on the Wesselmann painting; long, opaque, white needle crystals. This crystal form results from the slow cooling of the oils. This process of slow cooling oils, often called winterizing, has been used by the food industry and candle manufacturing to remove traces of wax and higher melting glycerides from vegetable oils. In this process waxes can generally be removed by chilling and filtering. Separation of high-melting glycerides, or stearine, usually requires very slow cooling in order to form crystals that are large enough to be removed by filtration or centrifuging. It is possible that non-drying oils present in the Wesselmann painting underwent a kind of winterization through very slow cooling that resulted in the efflorescence formation on the painting surface. This slow cooling may have occurred during the 3
environmental change of the painting when it was removed from sto 1 rage and placed in the newly renovated gallery space in the museum. This painting has been and will again be glazed with Plexiglas, so another important question to consider is whether or not the framing and glazing has an effect on the formation of efflorescence on the paint surface. If it does exacerbate the problem, what is an acceptable alternative framing and glazing? Tests further exploring the influence of environmental change on the formation of fatty acid efflorescence on paintings have also been suggested. Research to date on the formation of efflorescence has given us a clearer understanding of the composition of these crystals. However, the crucial issue of prevention still requires further research, which would benefit not only this painting, but the many paintings, objects and other artworks which suffer from similar surface formations. Treatment 1. Documented the condition of the artwork before treatment in written and photographic form (see figs. 2, 3, 6, 8). 2. Re-adhered loose edge of the ham paper collage element using Beva 371, first confirming the insolubility of the red paint layer in petroleum benzine (see figs. 8, 9). 3. Removed the surface efflorescence using the CO 2 snow gun after testing (see figs. 4, 5, 10). Because of the ease of removal no other techniques were required to clear the efflorescence from the paint surface. Visual and microscopic examination of the paint surface upon removal of these crystals showed no degradation or change to the original paint surface. (I will probably add a more detailed explanation of the CO 2 snow gun) 4. Documented the artwork after treatment in written and photographic form. 1 Joyce Hill Stoner told me of a discussion she had with Steve Kornhauser at the Wadsworth Atheneum; three temperas on panel by Andrew Wyeth in the Wadsworth collection are all housed differently (one has no glazing, one is glazed with glass, and one is in a climate-controlled box), but all three have developed efflorescence. 4
Figures Figure 2: Still Life #12, normal light, before treatment 5
Figure 3: Still Life #12, raking light, before treatment 6
Figure 4: Still Life #12, normal light, after treatment 7
Figure 5: Still Life #12, raking light, after treatment 8
Figure 6: Still Life #12, detail, raking light, BT Figure 7: Still Life #12, detail, raking light, AT Figure 8: Still Life #12, detail, BT Figure 9: Still Life #12, detail, AT 9
Figure 10: Still Life #12, CO 2 snow gun used for efflorescence removal, DT Figure 11: Sample 2 cross section, dark field and transmitted light 10
Figure 12: Sample 2 cross section, transmitted light Figure 13: Sample 2 cross section, UV light Figure 14: Sample 2, SEM image Figure 15: Sample 2, SEM-EDS, silicon 11
Figure 16: Sample 2, SEM-EDS, titanium Figure 17: Sample 2, SEM-EDS, calcium Figure 18: Sample 2, SEM image Figure 19: Sample 2, SEM-EDS, chlorine 12
Figure 20: FTIR Spectra of the efflorescence from Sample 3 and known palmitic acid and stearic acid spectra for comparison 13
Figure 21: FTIR spectra of the ground from Sample 11 Figure 22: FTIR spectra of the ground from Samples 6 and 10 14
Fig 23: XRD Spectra of efflorescence and known spectra for paraffin and two palmitic acids for comparison 15
Figure 24: GC/MS spectrum of the efflorescence 16
Figure 20: Still Life #12, sample locations for technical analysis 17
Bibliography Boon, J. J., P. Noble. 2007. Metal soap Degradation of Oil Paintings: Aggregates, Increased Transparency and Efflorescence. Paintings conservation catalog Vol. 19, American Institute for conservation Paintings Specialty Group. Washington, D.C.: AIC. Eccher, D. 2005. Tom Wesselmann. Rome, Italy: Museo d Arte Contemporanea Roma. Garver, T.H. 1971. Tom Wesselmann: Early Still Lifes: 1962-1964. Kansas City, Missouri: Nelson Gallery-Atkins Museum. Glenn, C. 1974. Tom Wesselmann: The Early Years: Collages 1959-1962. Long Beach, California: The Art Galleries, Californian State University. Hunter, S. 1994. Tom Wesselmann. New York, New York: Rizzoli International Publications, Inc. Loon, A. v. 2008. Color Changes and Chemical Reactivity in Seventeenth-Century Oil Paintings. Amsterdam, The Netherlands: FOM Institute for Atomic and Molecular Physics (AMOLF), Molecular Paintings Research Group. McCrone, W.C. and J.G. Delly. 1973. The Particle Atlas: Edition Two. Ann Arbor, Michigan: Ann Arbor Science Publishers, Inc. Ordonez, E., J. Twilley. 1998. Efflorescence on Works of Art. WAAC Newsletter 20 (1). Schilling M. R., S. Lake, E. Steele, and S. Q. Lomax. 2002. Modern Science and Contemporary Paintings: Preserving an Evolving Legacy. Conservation: The GCI Newsletter 17 (3): 4-10. Stoner, J. H. 2008. Personal communication. Lunder Conservation Center, Smithsonian American Art Museum, Washington D.C. *Select text, data, and images were taken from Smithsonian Museum Conservation Institute Technical Report MCI 6213 Still Life #12 by Tom Wesselmann, Smithsonian American Art Museum, compiled and written by Jia-Sun Tsang, in order to complete this report. 18