Investigation of preparation techniques for δ2H analysis of keratin materials and a proposed analytical protocol.

Accurate hydrogen isotopic measurements of keratin materials have been a challenge due to exchangeable hydrogen in the sample matrix and the paucity of appropriate isotopic reference materials for calibration. We found that the most reproducible δ(2)H(VSMOW-SLAP) and mole fraction of exchangeable hydrogen, x(H)(ex), of keratin materials were measured with equilibration at ambient temperature using two desiccators and two different equilibration waters with two sets of the keratin materials for 6 days. Following equilibration, drying the keratin materials in a vacuum oven for 4 days at 60 °C was most critical. The δ(2)H analysis protocol also includes interspersing isotopic reference waters in silver tubes among samples in the carousel of a thermal conversion elemental analyzer (TC/EA) reduction unit. Using this analytical protocol, δ(2)H(VSMOW-SLAP) values of the non-exchangeable fractions of USGS42 and USGS43 human-hair isotopic reference materials were determined to be -78.5 ± 2.3 ‰ and -50.3 ± 2.8 ‰, respectively. The measured x(H)(ex) values of keratin materials analyzed with steam equilibration and N(2) drying were substantially higher than those previously published, and dry N(2) purging was unable to remove absorbed moisture completely, even with overnight purging. The δ(2)H values of keratin materials measured with steam equilibration were about 10 ‰ lower than values determined with equilibration in desiccators at ambient temperatures when on-line evacuation was used to dry samples. With steam equilibrations the x(H)(ex) of commercial keratin powder was as high as 28%. Using human-hair isotopic reference materials to calibrate other keratin materials, such as hoof or horn, can introduce bias in δ(2)H measurements because the amount of absorbed water and the x(H)(ex) values may differ from those of unknown samples. Correct δ(2)H(VSMOW-SLAP) values of the non-exchangeable fractions of unknown human-hair samples can be determined with atmospheric moisture equilibration by normalizing with USGS42 and USGS43 human-hair reference materials when all materials have the same powder size.