Research and Technical Studies Afternoon Session, May 14, 2010

During the afternoon, Dr. Margaret G. MacDonald, presenting on behalf of her co-author Dr. Barbara H. Berrie, discussed “The effect of metal ions on early stages of curing in linseed oil models”. In an effort to gain a fundamental understanding of pigment-binder interactions, the

authors investigated early stages of paint film curing on the molecular level. The research focused on mechanisms driving these interactions and their effects on long-term behavior. The authors developed a series of experimental paint systems to model the interaction between ethyl linoleate (component of linseed oil) and various metal acetates. These experimental compositions remained soluble following polymerization allowing isolation for analysis. Samples were made under argon gas with varying concentrations of metal acetate. Sample thin films were cast on quartz plates (eliminating issue of oxygen penetration) and monitored using Attenuated total reflection Fourier transform infrared (ATR-FTIR) spectroscopy, ultraviolet-visible spectroscopy (UV-VIS) and fluorescence spectroscopy.

Based on results, the authors show that ethyl linoleate oligomerized films to form new carboxylates, as evidenced by loss of the C=O stretch peak at 1730 cm-1 and the formation of new bands between 1600 – 1500 cm-1. The presence of birefringent spheres with slight fluorescence may reflect a hybrid inorganic-organic framework structure in Pb-ethyl linoleate films. This suggests that there is some crystallinity at these sites and a possibly more thermodynamically stable end product. Further research is needed to delineate these structures and characterize their effects on overall paint film stability.

Dr. Joseph Swider, presenting on behalf of his co-authors Elaine F. Schumacher and Joseph G. Barabe, discussed “TEM as a complementary technique for pigment identification”. Transmission electron microscopy (TEM) can be used in specific instances for pigment identification, where morphological differences on the nano- and atomic scales are critical for pigment differentiation. The technique’s use is limited by the need for samples to withstand substantial heat during analysis. As well, samples must include features of interest that exhibit contrast on the nano-/atomic scale due to differences in thickness, chemistry or crystallinity.

Swider presented a series of case studies where TEM was extremely effective in identifying pigments including analysis of carbonaceous materials, differences between synthetic ultramarine and lapis lazuli, and iron tannate inks. Carbonaceous pigments appear similar on the macro-scale, but can be differentiated using morphological characteristics, which are best imaged using TEM on the nano-/atomic scale. Diffraction and elemental data also provides information that helps to identify specific carbonaceous pigments. Similar nano-/atomic

scale morphological data, obtained using TEM, help to differentiate between synthetic ultramarine and lapis lazuli, as well as TiO2- based pigments. Swider concluded that TEM’s high spatial resolution is helpful for characterizing unknown samples within these pigment classes, as well as pigments characterized by small particles. Finally, a web version of the McCrone Atlas of Microscopic Particles is available online and free for users.

Dr. Christina Cole presented her dissertation research in a talk titled “The identification of Early Eastern Woodlands quillwork dyes via LC-MS with a different approach to sample collection”. Quillwork objects, executed prior to 1856 and made by Native American groups

living east of the Mississippi River, were analyzed to identify natural dyes used during pre-industrial production. Literature associated with natural dyes (used during this period and geographic location) is based predominantly on primary source documents with little or no published scientific data.

Cole developed a method using a sample paper (1 mm x 1mm) swab to collect dye samples. This technique absorbs dyes in situ and dye samples measure approximately five nanograms. Cole tested the technique’s efficacy for liquid chromatography mass spectrometry (LC-MS) using prepared experimental dye samples. This technique is low-cost, low-tech and able to achieve fine sensitivity during analysis. Both unmordanted and mordanted dyes can be analyzed using this method.

Artifacts from five collections (Mashantucket Pequot, McCord Museum, NMAI, Peabody Museum, Harvard, UPenn museum) were screened by curators and Native Americans to ensure that no NAGPRA eligible or cultural sensitive objects were included in the study, while conservators identified artifacts with stable condition. Based on all these criteria, Cole sampled 53 objects to collect 126 samples from eight colors. Samples were selected to obtain the range of color distribution. All samples were extracted in situ from the quill cuticle.

Interpretation of analytical results identified a number of trends. Vulpinic acid, associated with moss found only in California, was identified on objects made dating to 1720 – 1750. These results suggest the presence of an early and extensive trade network between the Great Lakes region and California. Green quillwork was produced using a mixture of dyes, as no green dye materials were identified. Generally, green was produced using a mixture of blue and yellow, or, were the result of over-dyeing. Overall, there was limited use of European dyes. When used, European dyes were always associated with indigenous ones. There appears to be no trends in overall dye use, as a function of time. A number of previously unknown dye materials were identified, including a red dye associated with brazilwood, a wood limited to Brazilian rain forests. Further research is needed to delineate the trading network that resulted in access to the brazilwood material used during dyeing. Possible sources include trade of raw/unmodified brazilwood, recycling of brazilwood dyes preserved in traded blankets or reuse of exported brazilwood objects.