14C- AMS- dating of the ceramic ware

In the project ”Extraction methods of carbon in ceramic ware for 14C- AMS- dating. A pilot study on Bronze Age pottery from Scania” (Lindahl & Skog forthcoming), the aim is to develope methods to use ceramic pots and sherds for absolute dating using 14C-AMS-analysis. The working hypothesis in the project is that when a pot is exposed to a fire, soot from the fire will be trapped in the porous ware. This would be of great use and benefit for the archaeological society since the pottery material with no doubt is part of the human activities on the site. In this pilot project we have used 9 sherds from well dated find contexts i.e. the pits/ layers in which these sherds were found has been dated by standard carbon-14 dating mainly on grains of Hordeum Vulgare. The study also comprises recently produced “traditional” cooking pots from Zimbabwe as well as simulated manufactured vessels at KFL in order to study the variation in carbon content in newly made pots fired in an open fire and pots used for cooking over e period of time.

The study also comprises experiments on how deep the soot particles penetrate the ware during a specified time of use (Fig.1). To further verify and correlate 14C-AMS-datings on soot in the ceramic ware two simulated manufactured cups were made (iron rich clay tempered with 20% vol. crushed granitic rock and fired to 800oC in an oxidized atmosphere). The cups were placed over a fire with wood of a known dendrochronologically dated age.


Fig. 1. Penetration of sooth in the ware after ca 3 hours of use

In an initial analysis the carbon content and type of carbon (organic and in-organic) in the ceramic samples was tested by means of a carbon determinator. The pots used for cooking contained very high amounts of organic carbon – more than 7%. Also pots not considered to have been used for cooking but only fired in oxidized atmosphere in an open fire contained a reasonable amount of organic carbon – normally 1-2.5%. Considering the very low amounts of carbon that is required for a 14C-AMS-dating – ca 0.2 mg of carbon –  a sample size of only 2 gram of the ceramic ware is needed if the carbon content is 1%.

The sample preparation involved various steps of extracting carbon from the ceramic ware treating it with HF and HCl. Furthermore one sample was picked out to carefully investigate the homogeneity of the 14C concentration in the sample. This was made by stepped combustion where we picked out four different temperature fractions. The sample was mixed with CuO and stepwise heated to 200oC, 500oC, 700oC and 900oC. The collected CO2 gas was transported to a reaction vessel where CO2 was mixed with hydrogen and reduced to elemental carbon over a hot (600oC) iron surface. The graphite samples were pressed into an aluminium cathode. Radiocarbon dating was made using the SSAMS machine at the GeoBioSphere Science centre at Lund University (Skog, G. 2007, The single stage AMS machine at Lund University: Status report. Nuclear Instruments & Methods in Physics Research Section B-Beam Interactions with Materials and Atoms).

all dat

Fig. 2. All datings from the wares compared to the datings of seeds.

The dating by means of 14C-AMS-analysis of the ceramic samples correspond extremely well with the dating of the cereal grains from the same find contexts (Fig. 2). Two sherds show a clear deviant dating indicating that waste of much younger date was mixed in the pits.

The experiment of 14C-dating the carbon in the the cups sooted from wood of a known dendrochronologically dated age matched very well.

Anders Lindahl and Göran Skog

Page maintained by Anders.Lindahlgeol.lu.se
This page was last modified 2011-05-27 17:01
To the first page