Sampling for Instrumental Neutron Activation Analysis (INAA)


During the 2006 field season, in addition to regular processing, a selection of ceramic sherds were sampled for chemical analyses in the United States. Each sample consists of c. 300-400 mg of clay powder that will be entirely consumed in analyses. In total 289 samples were taken from 26 different survey units and points of interest, with a special focus on Early Bronze Age, Second Millennium BCE, and Lydian wares.


In 2006 powdered samples were taken from ceramic fragments collected by CLAS. These were submitted to the Research Reactor of the University of Missouri, Columbia (MURR), for INAA, and chemical fingerprints were obtained for each sample. The results suggest strongly that ceramics of several compositional groups were in use during the EBA; a different and single compositional group of ceramics was in use during the M–LBA; and a compositional group associated with production at Sardis was in use by the Lydian period. Thus, shifts of clay sources or ceramic recipes appear to correlate to shifts in time periods, which also correspond to dramatic changes in settlement patterns. Specifically, the multiple compositional groups of the EBA correspond to a period of diverse settlement with no apparent central organization; a much more centralized pattern of production is witnessed by the single compositional group of the M–LBA, when settlement patterns are dominated by an impressive centralized power centered on Kaymakçı; by the Lydian period another single compositional group predominates, in a period marked by the prominence of Sardis as the regional capital.


While the initial INAA results were positive, a number of issues still need clarification. Thus, one-gram powdered samples were collected from 82 sherds to fill out the previous analyses. These were packed in vials and exported to the US for expert analysis with an export permit from the Manisa Museum. Of specific note are ten sherds previously sampled with a diamond-tipped drill bit that were “re-sampled” with tungsten drill bits in 2008. These samples will be analyzed by MURR to test for previous contamination deriving from the use of a diamond-tipped drill bit and resulting in particularly high levels of Nickel and Zinc. Nickel, at least, occurs naturally and with some frequency in the soils around central Lydia, and, thus, whether its presence owes to accidental contamination or to actual ceramic composition needs to be determined.

In addition a number of sites first located in 2005 were under represented in the initial INAA sample, largely owing to restrictions on the size of collections imposed on CLAS by our Ministry Representative in 2005, and sites discovered after 2006 were not represented at all. Thus, three EBA sites were revisited in 2008 and powdered samples were taken from collected sherds. In addition, samples were taken also from ceramics collected after 2006 from POI07.01 and Kızbacı Tepesi, to assess further the organization of ceramic production during the M–LBA. In addition, and for the same purpose, powdered samples were taken from the LBA painted wares that seem to imitate Mycenaean wares.

Finally, based on ceramic analogues from around the world, it is clear that lake and river clays are often used to produce ceramics. In many cases the ratio of clay to sand in such clays precludes the need for temper. In 2008, CLAS will submit to MURR clay samples from cores taken from the Gygaean Lake as well as from the Hermos River floodplain. A chemical fingerprint of the clays from these cores should advance our understanding of local clay procurement by providing the location of a known source of clay.