|By Liz Peirce Samuel H. Kress Conservation Fellow at Winterthur Museum email@example.com
The furniture conservators at Winterthur have been treating the Chinese export lacquer within the collection after receiving a lacquer specific grant from the IMLS 3 years ago. Three of the pieces are multi-panel room screens which were separated and stored on custom built storage shelves in the former Upholstery lab within the conservation department. One of the screens, a six-panel screen from the first quarter of the 1800s, was selected for treatment. The panels are approximately 83 x 21 x 1 and are comprised of a wooden substrate coated with lacquer (a resin from a tree in the poison sumac family) and decorated with gold. Between September 21 and November 13, 2015, the screen was disassembled and stacked on the shelves. There is good airflow in this space and the HVAC system maintains an acceptable temperature and relative humidity range. The panels were stacked two panels per shelf, with the following stratigraphy:
-Mylar to cover esterfoam
-Thin foam support
Tyvek was chosen as a barrier to prevent interaction between the foam supports and the sensitive lacquer surface and to protect the object from excess light while awaiting treatment. Three 2 thick esterfoam blocks measuring approximately 16 x 4 were used as spacers to allow for airflow between the panels as well as to provide a cushion for stacking. Mylar was placed in between the lacquer surface and the esterfoam to prevent scratching. The panels were moved one at a time from the shelves for treatment.
Panels 1 and 2, which were stored on the lowest shelf, were removed for consolidation on May 19th, 2016. In the five to six months that the panels were stored in this orientation, the esterfoam had degraded. The Tyvek had become stuck to the lacquer surface where it had been in contact with the foam and, once removed, showed a brown discoloration on the surface. While the esterfoam did not visually appear to be degraded, it felt stiffer and crunchier than new esterfoam. Additionally, there were corresponding rectangles of discoloration on the lacquered panels. Panel 1 (Figure 1) had dark rectangles that were slightly tacky when touched and appeared to be supersaturated when examined under magnification (Figure 2). Three of these rectangles were present on the reverse of the panel. Panel 2 had rectangular shadows of the same dark supersaturation consisting of a dark outline that faded away from the Mylar barrier. The three marks are directly related to the esterfoam supports and are located on the reverse of the panel.
Figure 1: Darkened areas found on the reverse of Panel 1 are located within the red rectangles. The bright areas are reflected light from the fluorescent lights in the lab. Courtesy of Jim Schneck
Figure 2: Photomicrographs of the darkened area. On the left image, the line between unsaturated (left of dotted line) and saturated (right of bottom line) can be seen. The image on the right depicts the saturated surface from a damaged area. Photos courtesy of Liz Peirce.
The Tyvek sheet which had been in contact with the lacquer as well as the esterfoam were sampled and sent to the Scientific Analysis and Research Lab at Winterthur. Dr. Chris Petersen, scientist at Winterthur, ran gas chromatography/mass spectrometry (GC/MS) of the stained and unstained Tyvek and the degraded esterfoam and pyrolysis gas chromatography/mass spectrometry (py-GC/MS) of the esterfoam alone. GC/MS of the Tyvek and foam showed no chemical overlap between the two materials to support the migration of volatiles or the solid phase extraction of materials. GC/MS was also unable to identify the discoloration that appeared on the stained Tyvek, although it is similar in color to the lacquer coating. The conclusion of this comparative analysis is simply that the instrument detection limit is not as low as the color change detected by the human eye. Py-GC/MS of the foam showed the presence of urethane groups, but in smaller quantities than would be expected for a urethane foam. Py-GC/MS also identified chlorinated propylene, which may be a component in the foaming agent. However, there are no matching materials for any propylene compounds that contain chlorine without a fluorine component, and there is no evidence of hydrochloric acid in the results to imply the presence of a tetrachloride. The identification of the saturation agent is still an unsolved puzzle. If you are interested in the chromatographs of the analysis, we would be happy to share them.
The current system recommending Tyvek as a sufficient barrier between objects and degrading esterfoams has clearly been ineffective in this case. While the esterfoam may be acceptable for short term storage or transport, a less permeable barrier should be used between the esterfoam and objects. Care should also be taken to ensure that the esterfoam is new to ensure that older, degrading esterfoam is not recycled into storage materials. The same storage system was used for the rest of the panels in the screen, none of which exhibited any discoloration due to the esterfoam. It is possible that an older piece of esterfoam was selected in conjunction with newer, non-degraded materials from the packing area in the museum. We are fortunate in one way: the damage occurred on the reverse of the panels. The display orientation for the screen keeps the back away from visitors. Because the discoloration appears to be stable, we are not pursuing further treatment to try to reverse the damage at this time. However, after the incident, we have removed all of the esterfoam from the furniture lab to prevent a similar issue occurring again.