© K. Douka,  T. Higham, E. Mastora


The Oxford Radiocarbon Accelerator Unit (ORAU), where all radiocarbon measurements of the PalaeoChron project will be produced, is the world's leading radiocarbon facility specializing in archaeological dating. 


Radiocarbon dating is the most widely used scientific dating method. It was developed by Willard F. Libby (University of Chicago) who was awarded the Nobel Prize in Chemistry in 1960 "for his method to use Carbon-14 for age determinations in archaeology, geology, geophysics, and other branches of science".

Radiocarbon is a rare, naturally occurring C variant (isotope). It forms in the upper atmosphere when secondary cosmic rays collide with ¹⁴N atoms, emit a neutron and produce ¹⁴C (see scematic reaction above). Soon after formation ¹⁴C oxidizes into ¹⁴CO₂, mixes with the atmosphere and enters circulation.

The Radiocarbon Cycle. Image: http://www.imprs-gbgc.de

All living organisms will absorb ¹⁴C via photosynthesis, feeding and respiration. When the organism dies ¹⁴C is no longer fixated therefore its concentration reflects the level of ¹⁴C in the atmosphere at the time of death.


Radiocarbon is unstable and will decay back to ¹⁴N through beta decay at a constant rate. The known rate of ¹⁴C decay forms the basis of the radiocarbon dating method.


 After ~5730 years (1 half-life)  half of the original ¹⁴C atoms will be present in a sample, after ~11500 years (2 half-lives) a quarter of original ¹⁴C, and so on.


We can radiocarbon date all organisms that once lived and exchanged C with their environment. The limit of the method is about 60,000 years ago.



Sample types and pretreatment chemistry

The efficient removal of C-bearing contaminants from the samples prior to AMS dating is one of the most important parameters in the reliable application of the radiocarbon method. Chemical protocols have been developed for all types of organic samples and these are constantly assessed and revised to address difficult and persistent cases of contamination.


Dating old samples (>20,000 years old) as we do in PalaeoChron is particularly challenging. The reason is that Pleistocene material contains much greater quantities and types of contaminants and the organic matter we use for dating is often badly degraded. PalaeoChron aims to research and further develop some of the current pretreatment protocols.



Since 2001 members of our team have been working on the application of a method called ultrafiltration for dating bone.





An ultrafilter

Ultrafiltration is useful in the purification of collagen. An ultrafilter is a molecular sieve that separates high from low molecular weight (MW) fractions. High MW components will include undegraded alpha chains of amino acids, whilst low MW components can include degraded amino acids and peptides, and soil-derived contaminants, all of which are discarded after separation.


Ultrafiltration was first used in the late 1980s (Brown et al., 1988) but it was never widely adopted. It was not until 2000 that testing of the method was undertaken in Oxford and it was applied first to material from the British Palaeolithic (~2000–2005) and later to the western European Palaeolithic (2006–2009).

Cleaned collagen after filtration, gelatinization and lyophilization.


The use of ultrafilters lead to significant improvements to the quality of extracted collagen. When applied to the dating of Palaeolithic bones, the new results showed that many previously obtained radiocarbon dates (between 70-100%) were inaccurate and almost always too young when the same sample was redated using ultrafiltration. The new dates are more reliable, based on the chemistry of the ultrafiltration process and the fact that contamination with modern carbon is the overwhelming influence on reliability in dating Palaeolithic material.


We have also extended our methodological approach for bone dating in the last 5 years, and tested a new technique called single amino acid dating which also provides significantly more reliable radiocarbon dates for extremely contaminated bone. See our specialised page for updated information.


Palaeolithic charcoal is also difficult to clean. The incorporation of external contaminants and the degradation of the plant structure make dating old charcoal very challenging.

We apply rigorous new methodologies (ABOx-SC) to remove organic contamination of Pleistocene-age charcoal. Like ultrafiltration, this protocol has led to significant revisions in previously established chronologies.


Our group has spearheaded the application of this technique to the European Palaeolithic, in sites in Russia and Italy, and demonstrated its applicability.

In situ charcoal.

Upper Palaeolithic shell beads from Lebanon with direct ¹⁴C determinations


Marine shells were used by Palaeolithic hominids to make jewellery or tools or were discarded after eating the flesh.

We use marine shell to date Palaeolithic activity when alternative terrestrial material (bone, charcoal) is not available or is badly preserved. We have developped high-resolution pre-screening methods and a new pretreatment protocol (CarDS) to assess the preservation of shell prior to dating.



We also use ostrich eggshell to date Palaeolithic contexts. 

See here for research updates on this aspect of the project.

Accelerator mass spectrometry

The pre-cleaned samples are purified and converted into CO₂ which is then reduced to graphite and is transferred into an aluminum target holder (shown below) prior to accelerator mass spectrometry (AMS).

All radiocarbon measurements in PalaeoChron will be undertaken at the ORAU using a tandem accelerator (HVEE Tandetron AMS system) (shown below).

Target holders with graphite samples

(photo © James King-Holmes)

Tandem Accelerator at the ORAU

(photo © James King-Holmes)

Calibration curve

Radiocarbon determinations are not calendar dates therefore they require calibration and convertion to the calendar timescale.

Calibration in the Palaeolithic has become available only recently. Prior to 2009, no internationally agreed curve existed. The INTCAL04 calibration curve, the first to include a large number of datasets, extended back to 26,000 cal BP. In 2010, the INTCAL09 internationally-agreed calibration curve was published, and extended back to 50,000 cal BP (Reimer et al., 2009).

An update of this curve was published in late 2013 (Reimer et al., 2009). INTCAL13 is the most recent curve and the one the PalaeoChron team will be using.

IntCal13 Calibration Curve


30-50,000 years cal BP


IntCal13 for the period between 30-50,000 years ago. Produced with OxCal.

Datasets used to produced IntCal13 (after Reimer et al. 2013)