2:30pm – 3:00pm
Evaluation of Cleaning Agents for Artists’ Acrylic Paints with the Aid of High Throughput (HTP) Testing / Alan Phenix and Thomas J.S. Lerner, Getty Conservation Institute, Los Angeles, CA; Malinda H. Keefe, The Dow Chemical Company, Midland, MI; Bronwyn Ormsby, Tate, London, UK –
The presentation was made by Melinda Keffe from The Dow Chemical Company. The material she presented was building on the development of the HTP test methods for measuring cleaning efficiency which was presented at the AIC’s 2009 Annual Meeting in Los Angeles. The study reported on the effectiveness of Dow specialty ethoxylate surfactants (biodegradable, nonionic), Ethylene oxide/butylenes oxide diblock copolymers (biodegradable, nonionic), and non-polar (i.e. aliphatic hydrocarbon) solvents on soiled acrylic paints. The aim of the research is to provide results for aqueous systems based on parameters such as: pH; conductivity; surfactant type and concentration; chelate type and concentration; and combinations of these variables using the HTP test methods for measuring cleaning efficiency. The test results indicated that the Dow cleaners were at the top of the list in most of the cleaning samples with mineral spirits solutions. However most of the highly effective solutions also had a high conductivity which could contribute to swelling and cleaning issues with acrylics. It was also noted that the acrylic painted surfaces were dirty but not sooty – and Ms. Keefe admitted it would be interesting to test the affects with soot. Analytical tests to post-cleaned surfaces will be posted in publications being made in scientific journals. They are trying to modify the solutions and test the systems further using volunteers. Ms. Keefe extended an informal invitation for those willing to participate as a volunteer for these tests. If you are interested, please contact authors.
3:00pm – 3:30pm
A Question of Technique: Condition Issues Associated with Layering Structure in Richard Diebenkorn’s Ocean park Series / Ana Alba, Postgraduate Fellow and Susan Lake, Director of Collections Management and Chief Conservator, Hirshhorn Museum and Sculpture Garden, Smithsonian Institution; Mel Wachowiak, Senior Conservator, Museum Conservation Institute, Smithsonian Institution – Ana Alba presented the research. Richard Diebenkorn’s Ocean Park paintings began in 1967 and continued over 25 years. An artist applied, preparatory synthetic layer has been discovered in cross-sections from some of the paintings from this series. It was initially found as a dripped material along one of the tacking edges. Ms. Alba traveled to look at over a dozen other Ocean Park paintings from this series and discovered that those with this heavy, clear resinous layer beneath the paint have suffered condition issues (cracks and flaking paint), while those without it did not. The Hirshhorn’s Diebenkorn, No. 111, 1978 has a complex pattern of lifting cracks that extend over most of the painted surface. Analysis of materials was done at Smithsonian Museum Conservation Institute (MCI). FTIR identified the ground as an acrylic gesso and the paint medium as oil. It also identified the clear layer as an ethyl- acrylate-methyl methacrylate polymer (possibly Rhoplex AC-33), noting that an aged Rhoplex is similar in spectra as aged acrylic. Many of the cracks correlate with the charcoal under-drawings and white preparatory layer. The artist’s manipulation contributed to the cracking, as the paintings with thick layers have cracked more. The Ocean Park paintings on commercially prepared canvas have fared better (these have no clear synthetic layers but the paint application is the same). It was surmised that the alkyd’s polyester backbone has a higher molecular weight; that it dried faster; became more brittle with age; and is now causing cracks. The alkyd layers are separating from both the oil layers and the acrylic gesso. Summary: brittle layers are failing over a flexible support – Rhoplex AC-33 may cross-link but remains flexible. Ms. Alba concluded with an appropriate quote from Diebenkorn, stating “if you get an image, try to destroy it.”
4:00pm – 4:30pm
Do Weave Matches Imply Canvas Roll Matches? Don H. Johnson, Department of Electrical & Computer Engineering, Rice Univeristy; Ella Hendriks, Conservation Department, van Gogh Museum; Muriel Geldof, Netherlands Instiute for Cultural Heritage; C. Richard Johnson, Jr., School of Electrical & Computer Engineering, Cornell University – Mr. Johnson’s presentation augments the 2009 presentation given by C. Richard Johnson at the AIC meeting in Los Angeles. Their findings are based on computational algorithms for measuring thread counts and angles in both warp and weft determined from scanned x-ray images that produce “maps” of these variables across an entire painting. Mr. Johnson’s background as an electrical engineer prompted him to look at the horizontal and vertical threads (warp/weft) as a ‘signaling’ or frequency; with some interference (he jokingly referred to the painting is the interference). The peaks of the weave are the signals – axes; where an angle measure can be taken. (I hope I got that right because I was trying to type it out as quickly as he was saying it) Anyhow, an estimated periodic structure in 2-D spectrum with wedged areas around the peaks can be determined. This produces the weave maps (similar to heat maps), where the weft signals are wavy and the warp are crisper. These maps are like fingerprints for the canvas (not the painting) and may tie it to a roll of woven canvas. It is Easier to identify warp direction alignment in these maps. Over 42 van Gogh paintings were lined up by the computer program and placed in a possible roll sequence. The warp matching indicated cusping at edges, and the warp angle identified the type of loom used to crate it. But there were inconsistencies and so off they went to Belgium to learn more about the preparation of the pre-primed canvas, from the loom to the hand-primed artist canvas (at a factory that still hand-prepares canvas today as it was in van Gogh’s time). They discovered that the canvas is shipped to priming factory in large bolts of woven fabric, then cut to 10 meter lengths and put on priming frames. The canvas is hand sewn to the priming frames using hook and lace tensioning, which is what causes the cusping that was found (as it is looped around nails). The cusping is weak along the top edge and more pronounced along the bottom. After observing this manufacturing technique, their findings were most consistent in the warp direction, which related the canvas more to the bolt production, not the roll. Therefore, weave match means a bolt match. (Phew)
Unfortunately the remaining presentations were missed by this blogger as she had to catch her shuttle bus to the airport.