Mathieu Boudin

Improved radiocarbon dating for contaminated archaeological bone collagen, silk, wool and hair samples via cross-flow nanofiltrated amino acids

RATIONALE Radiocarbon dating and stable isotope analyses of bone collagen, wool, hair and silk contaminated with extraneous carbon (e.g. humic substances) does not yield reliable results if these materials are pre-treated using conventional methods. METHODS A cross-flow nanofiltration method was developed that can be applied to various protein materials like collagen, hair, silk, wool and leather, and should be able to remove low-molecular and high-molecular weight contaminants. To avoid extraneous carbon contamination via the filter a ceramic filter (molecular weight cut-off of 200 Da) was used. The amino acids, released by hot acid hydrolysis of the protein material, were collected in the permeate and contaminants in the retentate (>200 Da). RESULTS (14)C-dating results for various contaminated archaeological samples were compared for bulk material (pre-treated with the conventional methods) and for cross-flow nanofiltrated amino acids (permeate) originating from the same samples. Contamination and quality control of (14)C dates of bulk and permeate samples were obtained by measuring C:N ratios, fluorescence spectra, and δ(13)C and δ(15)N values of the samples. Cross-flow nanofiltration decreases the C:N ratio which means that contaminants have been removed. CONCLUSIONS Cross-flow nanofiltration clearly improved sample quality and (14)C results. It is a quick and non-labor-intensive technique and can easily be implemented in any (14)C and stable isotope laboratory for routine sample pre-treatment analyses.

Improved radiocarbon analyses of modern human hair to determine the year-of-death by cross-flow nanofiltered amino acids: common contaminants, implications for isotopic analysis, and recommendations.

RATIONALE In forensic investigation, radiocarbon ((14)C) measurements of human tissues (i.e., nails and hair) can help determine the year-of-death. However, the frequent use of cosmetics can bias hair (14)C results as well as stable isotope values. Evidence shows that hair exogenous impurities percolate beyond the cuticle layer, and therefore conventional pretreatments are ineffective in removing them. METHODS We conducted isotopic analysis ((14)C, δ(13)C, δ(15)N and C/N) of conventionally treated and cross-flow nanofiltered amino acid (CFNAA)-treated samples (scalp- and body-hair) from a single female subject using fingernails as a reference. The subject studied frequently applies a permanent dark-brown dye kit to her scalp-hair and uses other care products for daily cleansing. We also performed pyrolysis-gas chromatography/mass spectrometry (Py-GC/MS) analyses of CFNAA-treated scalp-hair to identify contaminant remnants that could possibly interfere with isotopic analyses. RESULTS The conventionally treated scalp- and body-hair showed (14)C offsets of ~21‰ and ~9‰, respectively. These offsets confirm the contamination by petrochemicals in modern human hair. A single CFNAA extraction reduced those offsets by ~34%. No significant improvement was observed when sequential extractions were performed, as it appears that the procedure introduced some foreign contaminants. A chromatogram of the CFNAA scalp-hair pyrolysis products showed the presence of petroleum and plant/animal compound residues, which can bias isotopic analyses. CONCLUSIONS We have demonstrated that CFNAA extractions can partially remove cosmetic contaminants embedded in human hair. We conclude that fingernails are still the best source of keratin protein for year-of-death determinations and isotopic analysis, with body-hair and/or scalp-hair coupled with CFNAA extraction a close second.