Final Report for Phase 1:
Analytical procedures for DEAE in DEAE contaminated paintings
- Traces of DEAE were detected by GC analysis of water swabs from 4” x 5” areas of the acrylic glazing on “Salon Rose Roix”. The amount detected was about 1.7 ng of DEAE per square millimeter of acrylic surface. This is about 18 times less than the amount detected on surfaces at Johnson Museum, Cornell University, when analyzed in 1983 (30 ng per square meter).
- DEAE was possibly detected by GC analysis at CCI of large cleaning swabs from “Swing Landscape” by Stuart Davis, and “Madame Chinnery”. The swabs analyzed were old ones taken more than a year prior to CCI involvement in the project and are ten to 100 times larger than swabs taken by Williams in 1996. They were not stored in air-tight containers and volatile components may have been lost. Barrett-Wilt (Chemistry, IU) detected DEAE in the swabs from “Swing Landscape” by GC analysis in 1994 and 1995.
- DEAE was possibly detected by GC analysis at CCI of 1.1 mg of dust from the back of “Peinture” by Soulages Barrett-Wilt detected DEAE by GC analysis.
- No DEAE was detected by GC analysis of water swabs from small, 3 mm x 3 mm, areas of painting surfaces from “Ste Catherine” or “Magdalen Reading”.
- No DEAE was detected by GC analysis of relatively large samples from “Blue Sky” (water swab from 16 mm x 13 mm area, or 0.5 mg scraping of varnish surface from 3 mm x 4 mm area) or “Green Trees” (water swab from 12 mm x 13 mm area, or 0.7 mg of scraping from paint surface from 4 mm x 5 mm area). These painting fragments had been exposed to steam humidified air containing DEAE in air conditioning vents at Lilly Library.
- No DEAE was detected by GC analysis of any dry swab from any object sampled.
- No free, unreacted DEAE was detected by FTIR analysis of samples from any painting However, “Blue Sky” and “Green Trees”, exposed at Lilly Library, and “Beach Scene” and “Portrait of Leila in Red” by Engel, showed some spectral characteristics that are similar to reaction products produced in the laboratory by direct addition of liquid DEAE to samples of varnishes and paints from various paintings. These reaction products have IR spectra that are different from material not treated with DEAE. The reaction products may be esters formed by reaction of carboxylic acids in the varnish and paint media with the alcohol group of DEAE (DEAE esters), or substituted ainmonium carboxylate salts formed by reaction of carboxylic acids with the nitrogen in the amine group of DEAE (DEAE carboxylates, analogous to reaction of ammonia with acids).
- IR spectroscopy indicates that “Ste Catherine” and “Magdalen Reading” contain water sensitive or water soluble materials like starch and protein DEAE or its reaction products were not detected on these paintings. In these paintings, hazing and other problems ascribed to contamination by DEAE might be due to the presence of these water soluble materials in or on the paints and varnishes.
- DEAE is a very good solvent for varnish resins and oil paints. When fragments of paintings were suspended in the vapor above a few drops of DEAE in a closed vial, the varnish absorbed so much DEAE vapor that the varnish dissolved and dripped off the fragment, and the paint became very soft.
- Liquid DEAE dissolves fresh dammar and aged danimar varnish film dating from 1948, and when this varnish/DEAE solution is cast on glass and allowed to sit in air, a hard film like a typical varnish film is formed.
- FTIR microspectroscopy is the best method to survey the paintings in the IUAM collection for the presence of DEAE and its reaction products. This method requires the smallest sample (particles much less than 1 mm in diameter) and provides the most information in a single analysis (medium and pigment composition, DEAE and reaction products, information from different layers, etc) The GC method provides only a yes or no answer for the presence of DEAE.
- Small amounts of DEAE esters and DEAE carboxylates, the reaction products of DEAE with varnish resins and oil paint media, but not free DEAE itself, were detected by FTIR microspectroscopy of particles measuring 50 μm (0.05 mm) in diameter. This is much smaller than the size of sample that is normally considered acceptable for sampling for chemical analysis.
- DEAE cannot be detected in dry swabs or water moistened swabs from areas less than about 30 mm x 30 mm by the GC technique used in this analysis The minimum area required to be swabbed in order to obtain a detectable amount of DEAE has not been determined conclusively, but areas measuring 4” x 5” (100 mm x 125 mm) on the acrylic glazing of “Salon Rose Roix” yielded a just barely detectable amount of DEAE so this would appear to be close to the area limit.
- DEAE and its reaction products are present in small concentrations on some surfaces, in some accretion layers, in some varnish layers, or in some paint layers, of some paintings in the IUAM collections The amounts present are very small and do not appear to be uniformly distributed since they are not detected in all paintings.
- No sample analyzed appears to have been significantly changed by the presence of DEAE reaction products, although these products may be responsible for some of the effects attributed to DEAE contamination.
- In some paintings, water soluble components in or on the varnishes and paints, rather than DEAE reaction products from DEAE contamination, may be the prime cause of abnormal behaviours observed when some paintings are swabbed with water moistened swabs.
Recommendations for further work
The results of these analyses indicate that DEAE and its reaction products are present on some paintings. Some of the effects attributed to DEAE contamination may be due to the presence of water soluble components like starch or protein (glue, egg, etc.). Since the number of paintings examined for this report is so small, a true picture of the extent of the DEAE problem cannot be drawn. More paintings should be analyzed.
FTIR microspectroscopy is the quickest, most informative, and least intrusive technique to use for this analysis.
To clarify the extent of the effects of DEAE and the mechanism of its interaction with paintings the following further work is recommended:
- Minuscule samples should be taken of varnishes and paints from many paintings that are apparently subject to DEAE contamination problems, for analysis by FTIR microspectroscopy, specifically for the presence DEAE esters and DEAE carboxylates, and for the presence of water soluble or water sensitive materials like starch and protein. This should quantify the number of paintings that are affected by DEAE and the extent of the DEAE problem.
- Test samples or model paintings should be exposed to DEAE vapors then analyzed by FTIR microspectroscopy to clarify the chemical reactions between painting media, DEAE, and DEAE reaction products. This has already been done to some extent in developing the FTIR microspectroscopic analysis procedure. The additional work needs to use more realistic exposure conditions such as exposure to a DEAE/air mixture of about 1% DEAE, rather than pure DEAE vapors or immersion in liquid DEAE This would determine whether the interactions under less severe exposure conditions are the same as at the higher concentrations used in the FTIR analysis.
- Test samples or model paintings should be analyzed by microscopical methods, before and after exposure to DEAE, and before and after cleaning treatments to determine the physical or structural effects of DEAE exposure and subsequent cleaning of DEAE exposed paintings.
- The analysis reported here has shown that for some paintings, wiping with water moistened swabs removes water soluble compounds from the painting, including starch and protein (“Ste Catherine”) and DEAE reaction products (“Green Trees”). Additional tests should be made to determine if swabbing with water moistened swabs is an effective treatment for removal of DEAE reaction products DEAE reaction products are polar compounds and therefore soluble in polar solvents like water and alcohols. They are also surface active agents and therefore may be soluble in nonpolar solvents like aliphatic and low aromatic hydrocarbons such as mineral spirits or naphtha. Alkanolamine soaps such as those produced by reaction of monoethanolarnine (MEA) with coconut or tall oil fatty acids show some solubility in these solvents. Analogous DEAE soaps with the increased hydrocarbon content due to substitution of the two hydrogens on the nitrogen of MEA by two ethyl groups will be even more soluble in these solvents. The effects of using these solvents to remove DEAE reaction products should be investigated. Using nonpolar solvents that may be less damaging to paintings than water, may be a valuable alternative procedure.