Biological mineral content in Iberian skeletal cremains for control of diagenetic factors employing multivariate statistics

Journal of Archaeological Science (Impact Factor: 2.14). 02/2013; DOI: 10.1016/j.jas.2013.01.022


The aim of this study was to define a strategy for a correct selection of bone samples by employing inductively coupled plasma optical emission spectroscopy (ICP-OES) for reconstructing the biological mineral content in bones through the determination of major elements, trace elements and Rare Earth Elements (REE, lanthanides) in skeletal cremains of ancient Iberians (III–II B.C), discovered in the Necropolis of Corral de Saus (Moixent, Valencia) between 1972 and 1979. The biological mineral content was determined taking into account diagenetic factors. A control method for a better reading of results was applied. To explore large geochemical datasets and to reduce the number of variables, Principal Component Analysis (PCA) was used, thus, providing a deeper insight into the structure of the variance of the dataset. PCA shows that the elemental profiles of bone and soil samples are clearly different. Bone samples obtained from the outer bone layer were shown to have a different elemental composition; more similar to soil samples than samples of the inner bone layer. PCA scores and loadings plots were preferred to dendrograms obtained using Cluster Analysis, due to the limits of the latter one to appreciate the spatial ordering of samples. Partial least squares discriminant analysis (PLS-DA), a frequently used supervised classification method, was applied to differentiate between degradation states of bone samples. PLS-DA results obtained in this study confirmed that changes derived from different burning conditions were associated with transformations in the mineral part of the bones. Accordingly, carbonized bones can be differentiated from cremated bones. Class assignment of bone samples with uncertain thermal conditions in dependence on their elemental composition has shown to be feasible. Consequently, for biochemical-archaeological studies the analysis and statistical classification of carbonized and cremated archaeological bones, as well as those exposed to unknown thermal conditions together with experiments in modern bones, are recommended.

  • [Show abstract] [Hide abstract]
    ABSTRACT: The thermal behaviour of the animal by-product meat and bone meal (MBM) has been investigated in order to assess how it is affected structurally and chemically by incineration. Initially composed of intergrown collagen and hydroxyapatite (HAP), combustion of the organic component is complete by 650°C, with most mass loss (50–55%) occurring by 500°C. No original proteins were detected in samples heated at 400°C or above. Combustion of collagen is accompanied by an increase in HAP mean crystallite size at temperatures greater than 400°C, from 10nm to a constant value of 120nm at 800°C or more. Newly formed crystalline phases appear beyond 400°C, and include β-tricalcium phosphate, NaCaPO4, halite (NaCl) and sylvite (KCl). Crystallite thickness as judged by small angle X-ray scattering (SAXS) increases from 2nm (25–400°C) to 8–9nm very rapidly at 550°C, and then gradually increases to approximately 10nm. The original texture of HAP within a collagen matrix is progressively lost, producing a porous HAP dominated solid at 700°C, and a very low porosity sintered HAP product at 900°C.
    Journal of Materials Science 08/2007; 42(23):9807-9816. · 2.31 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Wood is a ubiquitous material used in everyday life. Accurate identification of species can be of importance in a historical context enabling appropriate conservation treatment and adequate choice of material to be applied to historic wooden objects, and in a more modern context, in the identification of forgeries. Wood is also often treated to improve certain physical characteristics, often strength and durability. However, determination of whether or not a piece of wood has been treated can be very difficult. Infrared spectroscopy has previously been applied to differentiate between different wood species or between treated and untreated wood, often in conjunction with chemometric analysis techniques. Here, we report the use of mid-IR spectroscopy, coupled with partial least squares discriminant analysis for the discrimination between two walnut wood species and to differentiate between steam-treated and untreated samples of each of these wood species. We show that the discrimination between species and between steam-treated and non-steam-treated wood from Juglans nigra is very clear and, while analysis of the quality of the discrimination between steam-treated and non-steam-treated J. regia samples is not as good, it is, nevertheless, sufficient for discrimination between the two groups with a statistical significance of P < 0.0001.
    Analytical and Bioanalytical Chemistry 09/2010; 398(6):2713-22. · 3.66 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The Tri-State Crematory Incident in Nobel, GA (February 2001) revealed limitations in traditional human cremated remains (cremains) analytical methodology. The goal of this study was to develop a method for effectively classifying questionable sets of cremains as legitimate or contaminated. Eighty-eight samples of known human cremains, concrete, mixtures of the two, and questionable sets of cremains were acid digested and analyzed for 21 elements by Inductively Coupled Plasma-Optical Emission Spectroscopy (ICP-OES). Variable cluster and principle component analyses identified the seven elements (Sb, B, Li, Mn, Sr, Tl, and V) used to develop discriminant functions to classify questionable sets into two groups: cremains and concrete. The discriminant analysis shows that at the 0.90 probability level, mixtures of 50% or less human content were classified as concrete. Mixtures with 90% human content classified as cremains. Sixty percent and 75% human content mixtures remained in the questionable classification, but as the concentration of human increased in the mixture, the probability of assignment to the known cremains group increased. Most of the questionable human samples classified as cremains. This is a pilot study and cannot yet satisfy Daubert standards for courtroom admissibility, but it indicates that it is possible to determine the legitimacy of cremains using elemental analysis by ICP-OES coupled with multivariate statistical analysis.
    Journal of Forensic Sciences 10/2006; 51(5):967-73. · 1.31 Impact Factor


Available from
May 29, 2014