New protocol for compound‐specific radiocarbon analysis of archaeological bones

• Published in: Rapid communications in mass spectrometry 2018-03, Vol.32 (5), p.373-379
• Main Authors: Deviese, Thibaut, Comeskey, Daniel, McCullagh, James, Bronk Ramsey, Christopher, Higham, Thomas
• Format: Article
• Language: English
• Subjects: Amino acids; Archaeology; Bones; Carbon; Chromatography; Collagen; Contamination; Dating techniques; Eluents; Environmental Sciences; High performance liquid chromatography; Ions; Mass spectrometry; Pretreatment; Radiometric dating; Separation
• ISSN: 0951-4198; 1097-0231
• DOI: 10.1002/rcm.8047

Rationale For radiocarbon results to be accurate, samples must be free of contaminating carbon. Sample pre‐treatment using a high‐performance liquid chromatography (HPLC) approach has been developed at the Oxford Radiocarbon Accelerator Unit (ORAU) as an alternative to conventional methods for dating heavily contaminated bones. This approach isolates hydroxyproline from bone collagen, enabling a purified bone‐specific fraction to then be radiocarbon dated by accelerator mass spectrometry (AMS). Methods Using semi‐preparative chromatography and non‐carbon‐based eluents, this technique enables the separation of underivatised amino acids liberated by hydrolysis of extracted bone collagen. A particular focus has been the isolation of hydroxyproline for single‐compound AMS dating since this amino acid is one of the main contributors to the total amount of carbon in mammalian collagen. Our previous approach, involving a carbon‐free aqueous mobile phase, required a two‐step separation using two different chromatographic columns. Results This paper reports significant improvements that have been recently made to the method to enable faster semi‐preparative separation of hydroxyproline from bone collagen, making the method more suitable for routine radiocarbon dating of contaminated and/or poorly preserved bone samples by AMS. All steps of the procedure, from the collagen extraction to the correction of the AMS data, are described. Conclusions The modifications to the hardware and to the method itself have reduced significantly the time required for the preparation of each sample. This makes it easier for other radiocarbon facilities to implement and use this approach as a routine method for preparing contaminated bone samples.