Evaluating Visual Criteria for Identifying Carbon- and Iron-Based Pottery Paints from the Four Corners Region Using SEM-EDS

Abstract

Paint types on black-on-white pottery in the prehistoric American Southwest have had significance for both chronological and sociocultural interpretations. Visual attributes have formed the basis for distinguishing carbon- and mineral-based paints on ancient black-on-white pottery in the American Southwest for over 60 years. In this study, an SEM-EDS (scanning electron microscope-energy dispersive X-ray spectrometer) system was first used to make an independent objective determination of the mineral or non-mineral paint present on 15 Mesa Verde White Ware sherds. Then, a group of 19 people (including experienced archaeologists and newly trained individuals) examined and classified the paint on these sherds, achieving an overall accuracy of 84.2 percent. This group also ranked in priority order the visual attributes they felt were most useful in determining pottery paint type: nature of edges (fuzzy, sharp), absorption (soaks in, sits on top), luster (shiny, dull), color range (black-gray-blue; black-brown-reddish), flakiness (doesn't flake off, flakes off), thickness (thin, thick), and surface polish (polish striations visible through paint, striations not visible through paint). In each case, the attribute applicable to carbon-based paint is listed first. The most difficult sherds for the group to identify displayed attributes of both carbon and mineral paints. A category for "mixed" paint type, already in use by archaeologists, is a reasonable third category for labeling sherd paint, as long as it does not become a "catch-all" category. For problematic sherds, the SEM-EDS can be used to characterize paint type, then the visual attributes adjusted to improve investigator accuracy in paint type determination.

Raman microscopy in archaeological science

“…Raman spectra similar to that shown above the photograph were recorded from the black areas of this sherd. Previous SEM-EDS analysis had revealed the presence of iron in the paint, but the specific phase of ferruginous mineral could not be identified [132]. The Raman band at 672 cm ÿ1 is the principal feature of magnetite [80], thus confirming the presence of that iron oxide in the paint sample.…”

“…The precise identification of these pigments is important for understanding the technology and use of raw materials in the production of these pots as well as in determining typologies, chronologies, and geographical extent of these ceramics [132]. Of the many techniques applied to the analysis of these paints [131][132][133]142], only Raman microscopy was able to detect the carbonaceous component and simultaneously identify the specific mineral phases in pigment mixtures. Moreover, the analyses were performed rapidly and non-destructively.…”

Identification of black pigments on prehistoric Southwest American potsherds by infrared and Raman microscopy

“…Analyses of the black pigments on prehistoric Southwest American potsherds of the Ancestral Puebloan period have been discussed in a number of recent archaeometry studies [17,23,24,25]. Although these reddishbrown to black pigments are thought to be based on carbonaceous chars and/or iron oxide minerals, only elemental iron has been detected by the analytical methods used so far.…”

Evaluation of Painted Pottery from the Mesa Verde Region Using Laser Ablation-Inductively Coupled Plasma-Mass Spectrometry (LA-ICP-MS)

“…First, sample preparation for SEM can be destructive to the artifact and somewhat cumbersome since a typical quantitative analysis requires the sample be impregnated in epoxy, cut, polished, and carbon-coated. Second, analysis of carbon-based paint surfaces is difficult because paints are generally analyzed in cross-section and the carbon layer may not be visible (Stewart and Adams 1999: Figure 5). Third, trace elements, which include several transition metals likely to be present in ores used for pigment, are impossible to measure using SEM-EDS.…”