Focus
Comparison of transmission FTIR, ATR, and DRIFT spectra: implications for assessment of bone bioapatite diagenesis

https://doi.org/10.1016/j.jas.2014.03.008Get rights and content

Highlights

  • Fourier transform infrared (FTIR) spectroscopy is used to evaluate bone diagenesis.

  • The commonly used sample preparation techniques are transmission FTIR, ATR, and DRIFT.

  • Two indices calculated from FTIR spectra, IR-SF and C/P, are used to assess bone diagenesis.

  • The IR-SF and C/P values are not equivalent between the preparation techniques.

  • The transmission FTIR, ATR, and DRIFT spectra are not interchangeable.

Abstract

Evaluation of diagenesis in bioapatite samples is an important step for screening bone and tooth samples for stable isotope analysis to ensure in vivo signatures are obtained. Fourier transform infrared (FTIR) spectroscopy is one tool used to evaluate diagenesis by anthropological geochemists and commonly employs calculating the infrared splitting factor (IR-SF) and carbonate-to-phosphate ratio (C/P). There are three commonly used sample preparation techniques for vibrational spectroscopy: transmission FTIR, attenuated total reflection (ATR), and diffuse reflectance infrared Fourier transform (DRIFT). Each technique characterizes the internal vibrations of particular molecular groups, such as carbonate (CO3) and phosphate (PO4), using different optical properties to detect absorbance bands. Spectra are correlated between techniques using correction equations that account for differences in optical properties. Traditionally, anthropologists have used spectra produced by transmission FTIR to assess diagenesis, most commonly using two indices (IR-SF and C/P); however, recently the ATR and DRIFT techniques have been used as an alternative to transmission FTIR. The spectra produced by the three techniques are thought to be interchangeable in calculating the indices used to assess diagenesis. In this study, we evaluated the interchangeability of the three FTIR techniques by analyzing 452 prehistoric and modern bioapatite samples. Results indicate that IR-SF and C/P values are not equivalent between the three techniques. However, ATR produced more reliable results and was comparable to transmission FTIR. The DRIFT method showed much lower resolution, and did not distinguish between modern and prehistoric bioapatite samples as clearly.

Introduction

Diagenesis is a complex process that involves physical and chemical postmortem alterations to bones and teeth that is directly influenced by the burial environment, including local geological and groundwater conditions (Nielsen-Marsh and Hedges, 2000). As bone collagen and bioapatite show signs of diagenesis, changes in the molecular structure of bone can be detected by vibrational spectroscopy (i.e., infrared and Raman) (Carden and Morris, 2000, King et al., 2011). Anthropologists commonly use a potassium bromide (KBr) pelleting technique to prepare samples for transmission Fourier transform infrared (FTIR) spectroscopy to evaluate diagenesis of the carbonate and phosphate components of bone and enamel bioapatite (Garvie-Lok et al., 2004, Lee-Thorp and Sponheimer, 2003, Lee-Thorp and van der Merwe, 1991, Weiner and Bar-Yosef, 1990, Wright and Schwarcz, 1996). FTIR spectroscopy is considered a semi-quantitative tool that uses infrared radiation to determine what fraction of the incident light is absorbed at a particular wavelength. This produces a spectrum characterizing the vibrations of the bonds within a molecule for analyzing the structure of various materials. Each spectrum acts as a chemical fingerprint for mineral identification, and can be analyzed for unique information about mineral structure (Ferraro and Krishnan, 1990, Griffiths, 1983). Advances in instrumentation have generated alternate sample preparation techniques for vibrational spectroscopy that are less costly and labor-intensive; these techniques produce the same spectra but show less variation than the method involving the manual creation of pellets for transmission FTIR (Bruno, 1999, Cardell et al., 2009, Fuller and Griffiths, 1978, Haberhauer and Gerzabek, 1999, Yan et al., 1999). Two of these sample preparation techniques, attenuated total reflection (ATR) and diffuse reflectance infrared Fourier transform (DRIFT), have been used for spectral analysis as an alternative to transmission FTIR for decades in other fields, including chemistry, medicine, biology, and geology. These reflectance techniques operate with different optical properties, which do not require the traditional KBr pelleting preparation used for transmission FTIR spectroscopy. Correction equations can be used to account for the differences in how the infrared light beam is absorbed by the sample, so each spectra produced by the three techniques should be comparable (Ferraro and Krishnan, 1990; see Fig. 1). We compare the three techniques in this study to assess whether values of C/P and IR-SF (measures calculated from spectra) obtained from KBr pelleting for transmission FTIR correspond to the values obtained from spectra produced by ATR and DRIFT.

One reason to explore alternatives to the KBr pelleting method is that studies have shown the preparation technique for producing KBr pellets for transmission FTIR can introduce variation in diagenesis indicators due to differences in sintering pressures and times, KBr concentration, and individual preparation experience with sample preparation (Surovell and Stiner, 2001). Only a few studies have used ATR in an anthropological context, including studies on burned bone (Thompson et al., 2009) and diagenesis (Hollund et al., 2013, Stathopoulou et al., 2008); one study used DRIFT to characterize archaeological bone (Cardell et al., 2009). A pilot study comparing the diagenesis measures from transmission FTIR, ATR, and DRIFT spectra concluded that the three techniques produce similar spectra (i.e., identify the same positions of absorption bands for a sample), but show different relative peak intensities resulting in different values for calculations of diagenesis measurements (i.e., C/P and IR-SF) (Beasley and Carman, 2009). The purpose of this study is to analyze these measures from spectra produced by these three preparation techniques to evaluate whether the data are comparable for evaluating diagenesis in archaeological samples.

Section snippets

Bone diagenesis

Bone is a biphasic material composed of an organic component (predominately collagen) and an inorganic carbonated calcium phosphate mineral (bioapatite) fraction. Bone mineral crystallites and collagen fibers create a matrix that forms the structure of bone. Stable carbon and nitrogen isotope values from the collagen fraction primarily track dietary protein. Early studies using the carbonate component of bone mineral (bioapatite), however, were initially rejected because biogenic signatures can

FTIR spectroscopy

Infrared (IR) spectroscopy uses infrared radiation to measure what fraction of the incident radiation is absorbed at a particular wavelength, which can be used to establish semi-quantitative measures of bone composition (Carden and Morris, 2000, Ferraro and Krishnan, 1990, Griffiths, 1983, Thompson et al., 2009). Photons of IR radiation transmit through a sample and excite the molecules in bioapatite to higher rotational or vibrational states. This results in some wavelengths of light being

Bone samples

A total of nine sets of bone bioapatite samples (n = 452) from prehistoric, historic, and modern contexts were used in this research (Table A.1 Supplement Material). The prehistoric sample (n = 405) consisted of samples spanning 5000–1000 B.P. from various shell and earthen mound archaeological sites surrounding the San Francisco Bay and the Central California Delta region (Bartelink et al., 2010, Beasley, 2008). The historic sample (n = 22) consisted of nine human samples and one canid (Canis

Results

Table 1 presents the descriptive statistics and correlation coefficients between IR-SF and C/P for each preparation technique. Table 2 presents the statistical comparisons for each of the three preparation techniques by subsample grouping. The data produced by the three different FTIR preparation techniques were compared and found to be significantly different (p < 0.001).

Discussion and conclusion

This study demonstrates that while the three FTIR techniques identify the same chemical properties of a sample (based on the similar peak locations of carbonate and phosphate), the differences in resolution in the spectra result in different C/P and IR-SF values for each technique. Thus, the alternative ATR and DRIFT accessories do not result in C/P or IR-SF values that are comparable to the KBr pellet transmission FTIR technique (Fig. 3). Variation in values for each technique would be of no

Acknowledgments

We would like to express our gratitude to Dr. Cassady Yoder, Randy Wiberg, Ramona Garibay, Alan Leventhal, Rosemary Cambra and Dr. Frank Bayham for allowing access to bone samples. The majority of prehistoric bone bioapatite samples were from previous studies that sampled skeletons from the archaeological collections at the Phoebe A. Hearst Museum of Anthropology. A special thanks goes to Dr. Tim White, Natasha Johnson and the staff at the Phoebe A. Hearst Museum of Anthropology for allowing

References (34)

Cited by (0)

View full text