Dating Human Bone: Is Racemization Dating Species-Specific? Mehdi Moini,* Christopher M. Rollman, and Christine A. M. France Museum Conservation Institute, Smithsonian Institution, Suitland, Maryland 20746, United States * S Supporting Information ABSTRACT: Our recently developed dating technique based on the racemization rate of aspartic acid was applied to dating human bone, as well as that of other mammals, utilizing capillary electrophoresis mass spectrometry. First, several well- dated (mostly 14 C-dated and with strong archeological evidence) human bones ranging in age from 150 to ∼10 000 years were used to develop a calibration curve for human bone. The D/L ratio of aspartic acid for these specimens ranged from 2.4% to ∼10%, with a correlation coefficient of better than 0.99, indicating a strong linear relationship between the D/L ratio of aspartic acid and the age of the specimens. This calibration curve can now be used to date human archeological specimens of unknown age, up to ∼10 000 years. However, when the technique was applied to well-dated mixed species of larger mammal bones such as bison, whale, llama, etc., the calibration curve showed a slower rate of racemization with a lower correlation (0.88). As additional large mammal bones with less certain age (i.e., using archeological evidence alone with no 14 C- dating) were dated the correlation coefficient decreased to 0.70. The correlation coefficient decreased further to 0.58 when the racemization data from all mammals (including human) were added to the calibration curve, indicating the importance of using well-dated, species-specific specimens for forming a calibration curve. This conclusion is consistent with our previously published calibration curve for a single species of silk (Bombyx mori), which followed the expected reversible first-order kinetics. These results support species specificity of amino acid racemization dating. D etermining the age of human bones is important in archeology, forensic science, and other disciplines. Currently, radiocarbon ( 14 C) dating (using both conventional and accelerator mass spectrometry) is the most common technique for dating bone. However, 14 C-dating has several disadvantages, including (i) a large sample (often >1 g of bone) is necessary, which leaves undesirable visible and morphologic damage to valuable specimens; (ii) the process is expensive and usually costs several hundreds of dollars per analysis; (iii) sample preparation is time-consuming and labor intensive; and (iv) 14 C- dating does not provide accurate dating for objects younger than ∼500 years or older than 100 000 years. 1−6 To address these shortcomings, decades ago, an amino acid racemization (AAR) dating technique was developed and applied to a wide range of archeological and historic samples, including shells, teeth, and bone ranging in age from modern to millions of years, using a variety of analytical techniques. 7−11 The main advantages of AAR are (i) a minimal amount of sample is required, and (ii) widely available technology, such as gas and liquid chromatography (GC and LC, respectively), can be used with a variety of detectors, such as flame ionization detectors (for GC), ultraviolet (UV) spectrometers (for LC), and mass spectrometers (with both GC and LC). 12,13 The use of GC and LC with chiral columns allowed individual laboratories to analyze their own samples, rather than outsourcing them to a commercial laboratory for dating. The use of AAR for dating has been controversial, because of reports of unreliable dates for bones. 14,15 To re-examine the use of AAR for dating bone, we have investigated several factors that could have contributed to the development of unreliable data in past experiments. The first factor is ambiguity in the age of the bones studied. To date an unknown bone, one must first develop a calibration curve for the bone using well-dated specimens. 14,16 However, in some previous studies, the bone calibration curves were obtained using bone dated solely from archeological evidence. 9 To eliminate this uncertainty, our study used mostly 14 C-dated bones with good archeological constraints to create the calibration curves. A second factor is the failure to address bone species as a possible variable in racemization. 17 In this study, the emphasis is on species-specific dating. Specifically, a calibration curve was developed exclusively for Homo sapiens bones. Additional mammal species were subsequently added to the curve in order to examine the effect of mixed species on the slope and variance of the calibration curve. A third factor is the use of older technologies such as HPLC-UV or GC-FID to identify and quantify amino acids and their D/L ratio. 18−20 These non-mass-spectrometric techniques are unable to identify overlapping peaks and other impurities hidden within the peak, which could lead to inaccurate quantification of the D/L ratio. Tremendous advances in chiral chromatography have provided Received: September 4, 2013 Accepted: October 24, 2013 Published: October 24, 2013 Technical Note pubs.acs.org/ac © 2013 American Chemical Society 11211 dx.doi.org/10.1021/ac402917z | Anal. Chem. 2013, 85, 11211−11215