Pretreatment and gaseous radiocarbon dating of 40–100 mg archaeological bone

Helen Fewlass (1) , Thibaut Tuna (2) , Yoann Fagault (2) , Jean-Jacques Hublin (3, 4) , Bernd Kromer (1, 5) , Edouard Bard (2) , Sahra Talamo (1)

1 Department of Human Evolution [Leipzig]
2 Collège de France - Chaire Evolution du climat et de l'océan
3 Collège de France - Chaire internationale Paléoanthropologie
4 Max Planck Institute for Evolutionary Anthropology [Leipzig]
5 Universität Heidelberg [Heidelberg] = Heidelberg University

Radiocarbon dating archaeological bone typically requires 300-1000 mg material using standard protocols. We report the results of reducing sample size at both the pretreatment and 14 C measurement stages for eight archaeological bones spanning the radiocarbon timescale at different levels of preservation. We adapted our standard collagen extraction protocol specifically for <100 mg bone material. Collagen was extracted at least twice (from 37-100 mg material) from each bone. Collagen aliquots containing <100 μg carbon were measured in replicate using the gas ion source of the AixMICADAS. The effect of sample size reduction in the EA-GIS-AMS system was explored by measuring 14 C of collagen containing either ca. 30 μg carbon or ca. 90 μg carbon. the gas dates were compared to standard-sized graphite dates extracted from large amounts (500-700 mg) of bone material pretreated with our standard protocol. the results reported here demonstrate that we are able to reproduce accurate radiocarbon dates from <100 mg archaeological bone material back to 40,000 BP. Bone is one of the most frequently radiocarbon-dated materials recovered from archaeological sites. However, many precious archaeological bones, such as human remains or Palaeolithic bone tools, are too small or valuable for extensive destructive sampling. The reduction of sample size to enable direct dating of precious bone is therefore a key concern for the archaeological community. In the 1960s and 1970s, gas proportional counters required many grams of bone to produce a radiocarbon date 1,2. The development and utilisation of Accelerator Mass Spectrometers (AMS) in the 1980s represented a revolutionary step in the reduction of sample size and time required for dating 3. Routine measurements today typically require 500-1000 micrograms of carbon (μg C) to produce a high precision date. In recent years, several AMS labs have worked on modifications to the graphitisation and AMS measurement process for smaller samples containing <500 μg C 4-13. However, the graphitisation of small sample sizes is often time consuming and can be prone to large contamination effects 14,15. A recent study by Cersoy, et al. 16 demonstrated that graphite targets containing ca. 200 μg C from archaeological bones can be successfully produced and measured using the IonPlus Automated Graphitisation Equipment III (AGE 3) 17 and MIni CArbon DAting System (MICADAS) 18,19 developed at ETH Zurich. However, the hybrid nature of the MICADAS system offers an alternative solution to the complex process of graphitising small samples. Organic samples containing <100 μg C can be placed into an elemental analyser (EA) directly coupled to the gas ion source of the MICADAS via the gas interface system (GIS) 15,18,20-24. The automated system reduces both sample preparation time and the risk of contamination through handling, and has been successfully utilised in environmental and climatic applications 23,25-28. In our preliminary study 29 we demonstrated that the gas ion source of the AixMICADAS 30 is suitable for dating bone collagen CO 2 samples of <100 μg C back to 35,000 BP (uncalibrated radiocarbon years before AD 1950). However, as sample size is reduced the effect of contamination during pretreatment and measurement increases greatly. Sample pretreatment involves the extraction and purification of carbon endogenous to the original bone. Any contamination remaining in the sample at the time of dating can lead to erroneous results. The effects become increasingly catastrophic with the increasing age of the sample due to the minute concentrations of residual 14 C. For example, in a bone extract ca. 40,000 BP, 1% modern carbon contamination would skew the resulting 14 C age by over 7,000 years.

Domaines: Archéologie et Préhistoire

Helen Fewlass, Thibaut Tuna, Yoann Fagault, Jean-Jacques Hublin, Bernd Kromer, et al.. Pretreatment and gaseous radiocarbon dating of 40–100 mg archaeological bone. Scientific Reports, 2019, 9, 5342 (2019), ⟨10.1038/s41598-019-41557-8⟩. ⟨hal-02096824⟩