Effects of heating on the carbon and oxygen-isotope compositions of structural carbonate in bioapatite from modern deer bone

Abstract

Carbon and oxygen-isotope compositions of mammalian bone carbonate from bioapatite can be used to reconstruct diet, track migration and determine climatic conditions during ancient times. Hence, the extent to which primary isotopic compositions are preserved during post-mortem alteration is of interest. Bone samples are commonly obtained from contexts where they have been heated (e.g., middens, hearths, cremations). In addition to naturally occurring alteration, burning and boiling may also contribute to modification of primary isotopic compositions. We have examined the changes in isotopic composition of structural carbonate carbon (δ¹³C_carb) and oxygen (δ¹⁸O_carb) during the heating and boiling of modern deer bone, and compared these results with values reported elsewhere for coexisting phosphate oxygen (δ¹⁸O_p). Recently deceased White-Tailed deer leg bones (Odocoileus virginianus) were sectioned and heated incrementally from 25° to 900 °C in 25° intervals. In a second experiment, these bones were boiled before heating in an oven. Changes in crystallinity were assessed for 45–63 µm bone powders using X-ray diffraction and Fourier transform infrared spectroscopy. Weight loss was measured using thermogravimetry.

Heated samples and samples boiled before heating exhibited similar behaviour. Original oxygen isotopic compositions were preserved for δ¹⁸O_carb (± 2‰) up to 225–250 °C. At higher temperatures, δ¹⁸O_carb (and δ¹⁸O_p) values became progressively lower relative to original values by as much as 8‰. Only small changes in crystallinity indices (CI) were recorded from 25–675 °C, despite much larger changes in oxygen-isotope composition above 225–250 °C. The δ¹⁸O_carb and δ¹⁸O_p values were positively correlated over the entire heating range, with carbonate being enriched in ¹⁸O relative to phosphate. Carbon isotope compositions remained constant (< ± 1‰) during heating to 650 °C. At higher temperatures, there was a rapid rise in δ¹³C_carb values (2–5‰), accompanied by an increase in CI, as carbonate was lost. No structural carbonate remained above 725 °C.

Introduction

Stable carbon and oxygen isotopic compositions of bioapatite from terrestrial mammals have been used extensively in palaeoclimatological and bioarchaeological investigations (Longinelli, 1973, Sullivan and Krueger, 1981, Longinelli, 1984, Luz et al., 1984, Koch et al., 1994, Bryant and Froelich, 1995, Kohn and Cerling, 2002, Ambrose and Krigbaum, 2003). The possibility of post-mortem isotopic alteration is therefore of interest, with the behaviour of structural carbonate being of most concern (Zazzo et al., 2004). Burned bones are frequently found in archaeological contexts (White, 1992). Potential for their isotopic compositions to be altered therefore exists not only during processes associated with burial (natural or otherwise; e.g., cremation) (Pate and Hutton, 1988, Hedges, 2002, Lee-Thorp, 2002, Trueman and Martill, 2002), but also from cooking associated with food preparation (Brain and Sillen, 1988, Nicholson, 1993, Taylor et al., 1995, Shahack-Gross et al., 1997, Balter, 2001, Roberts et al., 2002).

Here we assess the stable isotopic behaviour of structural carbonate oxygen (δ¹⁸O_carb) and carbon (δ¹³C_carb) during the heating (“burning”) and boiling of White-Tailed deer bone bioapatite. These experiments mimic cooking, food-waste disposal, and cremation practices that can be encountered in archaeological contexts. The data may also provide some insight into changes in isotopic composition during burial (i.e., diagenesis), which generally occurs at lower temperatures over a much longer time (Roberts et al., 2002). We also consider whether a systematic relationship is preserved between the oxygen-isotope composition of structural carbonate and coexisting phosphate (δ¹⁸O_p) during these post-mortem processes. In terrestrial mammals, the oxygen-isotope values of structurally bound carbonate and phosphate are closely related to body water composition (Bryant et al., 1996, Iacumin et al., 1996a, Iacumin et al., 1996b). There remains, however, considerable debate concerning the usefulness of the δ¹⁸O_carb value as a proxy for the δ¹⁸O_p value of archaeological bone. It is commonly thought that the δ¹⁸O_carb value is more susceptible to post-mortem changes (Cerling and Sharp, 1996, Iacumin et al., 1996a, Kohn and Cerling, 2002, Lee-Thorp, 2002, Vennemann et al., 2002, Zazzo et al., 2004).

Bone samples that have been exposed to heating or burning, for example in middens and hearths, are often used in archaeological studies (Shipman et al., 1984, Brain and Sillen, 1988, Nicholson, 1993, Stiner et al., 1995, Taylor et al., 1995, Shahack-Gross et al., 1997, Balter, 2001, Roberts et al., 2002). Relatively few studies describe the isotopic behaviour of burnt bone in any detail. Notable amongst these reports, Person et al. (1996) investigated the variation in carbon isotopic composition of burnt bone at eight temperature steps between 300° and 700 °C. They reported a general increase in δ¹³C values of total carbon (organic carbon and structural carbonate) with increasing temperature. Lindars et al. (2001) studied removal of non-phosphate oxygen from bioapatite as a prelude to laser-fluorination isotopic analysis. They showed that δ¹⁸O_p values of bone heated in 100 °C steps from 100–1000 °C varied with temperature. Building on this research, we describe δ¹³C_carb and δ¹⁸O_carb variations during the heating and boiling of bone over a temperature range of 25–900 °C in 25° increments.