43rd Annual Meeting – OSG Tips Session, May 16, "Plaster Cleaning Tests" by Kathryn Brugioni

In this tip presented during the OSG Tips session luncheon, Kathryn Brugioni discussed the use of Reflectance Transformation Imaging (RTI) for evaluating whether certain dry cleaning methods for plaster abraded or damaged the surface of the object.
Dry methods are preferred over wet methods for cleaning plaster because of the risk of solubilizing the substrate during treatment. When Kathryn was presented with a heavily soiled plaster bust that required cleaning, she turned to the use of vinyl erasers as a cleaning method. Using previously published information that evaluated various types of erasers (Williams and Lauffenburger 1995; Pearlstein, et al. 1982) she decided to test two different PVC-based erasers made by Staedtler: the 526 50 Mars plastic eraser and the 527 05 Mars eraser strip refills, to evaluate not only how well they cleaned soiled plaster, but whether they abraded the surface.
Once she chose what dry cleaning method/materials to test, Kathryn was left with the question of how to evaluate surfaces after cleaning to determine the level of abrasion or scratches resulting from the treatment.  Examination using SEM imaging has been used (Wharton, et al. 1990) , but it is a technique that may not be available to all conservators.  So she looked to a method that could be more accessible: RTI.
RTI, or polynomial texture mapping, is an imaging technique that allows for an interactive display of an image under different lighting conditions.  Multiple images are taken of an object where the object is kept in a fixed position, but the light source moves.  The images are processed using using freely available software which combines all the images taken into a single image presented in an interface that allows for the direction of light to be moved across the image at different angles highlighting surface features. (The non-profit organization Cultural Heritage Imaging (CHI) is one of the leaders in this type of imaging for cultural heritage and has lots of information on its website about this technique, steps on how to do it and the software needed to process the images).  The software  also allows for different types of light or shadow effects to be rendered which may improve or further highlight surface examination.  All you need for RTI is a camera, moveable light source and some metal spheres (ball bearings) as markers that help the software determine the direction/angle of the light.  These are all things that conservators have on hand or can readily purchase (like the ball bearings) making this type of surface examination/imaging more accessible and much cheaper than an SEM.
Kathryn cleaned the surfaces of plaster test coupons using the erasers and imaged them with RTI before and after cleaning.  She soon saw that it was possible to see scratches on the surface using this technique.  However, she wanted a way to quantify the scratches and determine what the limit was in terms of scratch size observable using RTI.  She abraded plaster coupons with a range of grades of micromesh, from 400-1800, and then examined the surfaces using RTI.  She noted that you could detect scratches made with up to 800 grit micromesh, but higher grits, like 1800, created more subtle scratches that were not as easily discernable.
Comparing the scratches made by the two erasers on the plaster coupons to those of different micromesh grades, the scratches made by the Mars plastic eraser were similar to those made by 1200 grit micromesh (measured to be about 34μm size scratches) and the eraser strips made scratches similar to 466 grit micromesh (measured to be about 60μm sized scratches).  So the eraser strips are much more abrasive to plaster surfaces than the plastic eraser.
Based on Kathryn’s findings, it looks like RTI can be used to evaluate any surface scratches or changes caused through the abrasive action of erasers used for dry cleaning plaster. Though there are limitations to the use of this technique, and fine scratches may not be readily visible, RTI is a useful, and accessible, examination tool and can provide important information on surface changes caused by certain cleaning methods.

Pearlstein, E., D. Cabelli, A. King, and N. Indictor. 1982.The Effect of Eraser Treatment on Paper. JAIC 22(1): 1–12.
Wharton, G., S. Lansing Maish, W.S. Ginell. 1990. A Comparative Study of Silver Cleaning Abrasives. JAIC. 29(1): 13-31.
Williams, J. and J. Lauffenburger. 1995. Testing Erasers used to Clean Marble Surfaces. Objects Specialty Group Postprints, Vol. 3: 118-124.

43rd Annual Meeting – Objects Session, 16 May, "Beloved Treasures: Assessing the Effects of Long-Term Display on Models Containing Wax" presented by Marissa Stevenson

Marissa Stevenson presented on the effects of long-term display on the “Libbey Dolls” in the Toledo Museum of Art. The dolls were created in 1915 and are named for the collector who purchased them. They depict characters popular in the later nineteenth and early twentieth century using multiple types of media that include wax, plaster, feathers, fur, human hair, lace, plaster, cotton, plant material, wire armatures, and more. After full-time display for over fifty years, along with two interventions that date to pre-collection in 1917 and to 1951, the dolls were showing damage. During that time, the dolls were exposed to incandescent lighting with the associated light and heat conditions, fluctuating environment, and lack of support on display. The result? Disfigurement, cracking, light damage, and other structural damages, topped off with a healthy dose of dust and grime.

Before treatment photograph of 1917.650, a Libbey doll in the Toledo Museum of Art. Photo provided by Marissa Stevenson.
Before treatment photograph of 1917.650, a Libbey doll in the Toledo Museum of Art. Photo provided by Marissa Stevenson.

The conservation study focused on the wax and included identification and treatment. Through IR analysis, the wax was identified as bee’s wax, which was observed to become brittle and darken. It was applied in four layers, with two layers of plaster supporting it. Cleaning tests of 1% ammonium citrate, saliva, methylcellulose with deionized water rinses, and gels, showed that wheat starch paste and saliva were the most effective. Adhesive tests with Jade 403, Aquazol 500, Rhoplex AC33&N58, Avalure UR450, Avalure AC120 showed that Avalure AC120 and Aquazol 500 preformed best. The Avlaure also supported good pigment dispersion, so this was used in areas that needed pigmented fill material. Since Avalure has not been tested as an adhesive, Stevenson chose to use it to adhere a broken leg in her case study. The leg’s hidden location will allow them to observe how the resin holds up over time without compromising aesthetics should it prove to be problematic in any way. Since the head is in a visually prominent location, and Aquazol has been tested as adhesive with good results, it was used there.
With the wax issues under control, the next step will be to investigate what the clothing’s needs and to address them. Finally, storage mounts need to be made. Solutions will need to consider how to fully support the dolls, and will in some cases necessitate horizontal storage, and in others vertical storage. Fosshape is a likely candidate for helping to create these supports, forming it on Ethafoam model. An audience member brought up the idea of using Varaform fabric for the mounting system since he said it is less bulky than Fosshape, comes in 3 weights, and though it is heated to use, it is not thermally conductive. This was a new material for Stevenson, who said she would investigate. This also led to a brief discussion regarding to long-term storage. Since Fosshape has pass Oddy testing, it seems like a good idea, but we do not know about its aging properties, so question was raised about it being appropriate for long-term storage.

39th Annual Meeting, ASG Morning Session, June 2nd, Student Papers, “Rediscovering an American Master: The analysis and proposed treatment of the decorative plaster ceiling of Robert Winthrop Chanler’s Gertrude Vanderbilt Whitney Studio, New York” Lauren Vollono Drapala, University of Pennsylvania

Lauren’s presentation introduced me to a true treasure hiding out in my downtown New York City neighborhood at 8 West 8th Street on Washington Square. In what was once the original location of the Whitney Museum lays the intimate studio retreat of patroness to the arts, Gertrude Vanderbilt.  Lauren’s focus on this room was to conduct historic research, assessment, analysis and documentation of the decorative ceiling created by artist Robert Winthrop Chanler.  What first appears as textured white plaster ceiling is actually a highly detailed bas relief plaster composition that still has some indications of paint. [PHOTO].   Like Angela’s Curmi’s talk before, Lauren knew there was an original decorative finish below the surface based on similar works by the artist, documentation, and cross-sectional analysis, but was having problems with accessibility to the paint after the ceiling had been overpainted white, most likely due to unaddressed failings of the plaster and paint over time.  Lauren’s goal was to eventually discover a means to safely remove the overpaint  and conserve the original paint below. I was rooting for her!


After creating high res photodocumentation of the ceiling, including an overall raking light image to accent the relief, Lauren mapped the condition and motif of the ceiling to aid with her investigation.  Lauren also visited other sites with similar work by the artist at Peebles Island and Vizcaya Museum to note pigments and decorative effects he employed at a similar time.  Lauren’s paint analysis [PHOTO]revealed that the newer layers of paint were bound more strongly to the original paint than the original paint was bound to its plaster substrate; the removal of the overpaint was not looking favorable.  Lauren was even given a detached portion of the ceiling to test  overpaint removal gels on, unfortunately with little success.


Following what looked to be a disappointing end to all of her hard work, Lauren took it a step further.  She made mock up panels of the ceiling based on her research and analysis to get a better idea of the surface then combined that information from her analysis to propose what it “May have once looked like”.  Here is the initial result of her efforts [PHOTO].  I’m still rooting for this project.  In the meantime if you are interested in learning more, read Lauren’s thesis: http://repository.upenn.edu/hp_theses/151/