Printable PDF Version

December 6, 2006

Quantitative Determination of Absolute Organohalogen Concentrations in Environmental Samples by X-ray Absorption Spectroscopy

A.C. Leri¹, M.B. Hay², A. Lanzirotti³, W. Rao⁴, and S.C.B. Myneni^5,6
1Department of Chemistry, Princeton University, Princeton, NJ; ²Department of Civil and Environmental Engineering, Princeton University, Princeton, NJ; ³Consortium for Advanced Radiation Sources, The University of Chicago, Chicago, IL; ⁴Savannah River Ecology Laboratory, University of Georgia, Aiken, SC; ⁵Department of Geosciences, Princeton University, Princeton, NJ; ⁶Earth Sciences Division, Lawrence Berkeley National Laboratory, Berkeley, CA

A method for quantification of organic and inorganic halogen concentrations in environmental samples using x-ray absorption spectroscopy has been developed. Organochlorine (Cl_org) and inorganic chloride (Cl_inorg) concentrations are determined from Cl 1s XANES spectra. The absolute fluorescence intensity of these spectra is linearly dependent on Cl concentration. Calibration curves are obtained from NaCl standards in a matrix of uniform bulk density, and Cl concentration in natural samples is estimated from these curves with high precision. The fractions of Cl_inorg and aliphatic/aromatic Cl_org contributing to the total Cl 1s XANES signal are computed through least-squares spectral fitting with representative model compounds. Concentrations of organic and inorganic Br (Br_org and Br_inorg) in sediment samples are also measured using a combination of XRF and Br 1s XANES spectroscopy.

Graduate student author Alessandra Leri (left) and Michael Hay (right) collect natural organic matter samples in the field.

Halogens undergo complex biogeochemical transformations in terrestrial systems, cycling between organic and inorganic forms. Naturally produced organohalogens include a diverse array of relatively low molecular weight molecules, as well as macromolecules of indeterminate structure. Despite the ubiquity of natural organohalogens, the processes associated with their formation and degradation in the environment remain poorly understood, largely due to the inadequacy of available techniques to account for the myriad halogen species in heterogeneous soil, sediment, plant, and aqueous samples. Established quantitative methods require chemical isolation of organohalogen fractions from natural samples and are prone to partial recoveries and/or chemical alterations of halogens. We have applied synchrotron-based x-ray fluorescence (XRF) and x-ray absorption near-edge structure (XANES) spectroscopy to the in situ determination of Cl_org and Br_org concentrations in natural samples.

Quantification of Cl relies on the absolute fluorescence intensity of Cl 1s XANES spectra (acquired at beamline X19A) determined at 2850.8 eV, an energy ~20 eV above the Cl K-absorption edge (Fig. 1A). By this energy, the near-edge oscillations are attenuated and absorption intensity is independent of Cl speciation. Polyacrylic acid-based NaCl standards yield calibration curves with a strong linear relationship between Cl concentration and absolute Cl fluorescence intensity (Fig. 1B). X-ray beam absorption and scattering by polyacrylic acid-based standards and natural organic matter (NOM) are comparable, allowing for direct measurement of Cl concentrations in homogenized NOM sample pellets, which minimizes matrix effects.

Figure 1. Quantification of Cl concentration from Cl 1s XANES spectra. A) Unnormalized Cl 1s XANES spectrum of NaCl standard in PAA matrix (200 ppm Cl). The difference at 2850.8 eV between splines through the pre-edge and post-edge regions is used as a measure of absolute Cl fluorescence intensity. B) Relationship between Cl concentration in PAA-based NaCl standards and absolute Cl fluorescence intensity. Dashed line, equation, and R2 value represent linear fits to the data.

Unnormalized Cl 1s XANES spectra of oak leaf NOM at progressive stages of decay are shown in Fig. 2A (a = least degraded; c = most degraded). Comparison of the absolute Cl fluorescence intensity values of these spectra with standard curves (as in Fig. 1B) yields the following total Cl concentrations: a, 364 ppm; b, 141 ppm; c, 70 ppm. The decreasing trend in Cl concentration is consistent with leaching of soluble forms of Cl from the mulch material as part of the degradation process.

In addition to providing Cl concentration measurements, Cl 1s XANES spectra illuminate the chemical forms of Cl present in a sample. Cl_inorg exhibits a broad absorption maximum (Fig. 2B, x) higher in energy than the more intense maxima characteristic of Cl_org compounds. These sharp peaks denote 1s -> * or * transitions and differ in energy depending on C-Cl bond length, as becomes evident through comparison of aliphatic and aromatic Cl_org spectra (Fig. 2B, y-z). The substantial variations in spectral features depending on the coordination environment of Cl allow percentage estimates of Cl_inorg and aliphatic and aromatic Cl_org in natural samples to be ascertained via least-squares fitting of normalized sample spectra with spectra of representative model compounds (Fig. 2B). Combination of these proportions with total Cl concentrations yields the concentrations of Cl_inorg, aromatic Cl_org, and aliphatic Cl_org in NOM samples (Fig. 2, caption).

Figure 2. Determination of Cl speciation and concentration in NOM samples from Cl 1s XANES spectra. A) Unnormalized Cl 1s XANES spectra of pulverized oak leaves at different stages of decay (a = least degraded; b = more degraded; c = most degraded). Arrows illustrate the differences in absolute Cl fluorescence intensity among the spectra. B) Least-squares fit of background-subtracted, normalized spectrum b with representative inorganic (x, glycine-HCl), aliphatic (y, chlorodecane), and aromatic (z, chlorophenol red) Cl model compounds. Spectrum b data are represented by circles and fit by the solid line; r is the residual after the fit. Estimated Cl speciation by least-squares fitting: a, 78% inorganic (284 ppm), 16% aromatic (58 ppm), 6% aliphatic (22 ppm); b, 40% inorganic (56 ppm), 20% aromatic (29 ppm), 40% aliphatic (56 ppm); c, 16% inorganic (12 ppm), 42% aromatic (29 ppm), 42% aliphatic (29 ppm).

Total Br concentrations in environmental samples have been determined via XRF analysis for decades. The novelty of our application lies in the combination of quantitative information from XRF (acquired at beamline X26A) with Br speciation information from Br 1s XANES spectra (acquired at beamline X23A2) to determine absolute concentrations of Br_org and Br_inorg in chemically heterogeneous sediments.

This method provides a rigorous new approach to the determination of organohalogen concentrations in complex natural samples in situ. Robust, quantitative information about organohalogen fluxes in soil and sediment systems will help complete the description of natural halogen cycles, potentially illuminating the sources and sinks of naturally and industrially produced organohalogens in the environment.

BEAMLINES
X19A, X26A, X23A2

FUNDING
Office of Basic Energy Sciences within the U.S. Department of Energy
National Science Foundation (NSF) Chemical Sciences Program
NSF Graduate Research Fellowships

PUBLICATION
A. Leri, M. Hay, A. Lanzirotti, W. Rao, and S. Myneni. “Quantitative Determination of Absolute Organohalogen Concentrations in Environmental Samples by X-ray Absorption Spectroscopy” Analytical Chemistry, 78, 5711-5718 (2006).

FOR MORE INFORMATION
Alessandra Leri
Department of Geosciences
Princeton University
Princeton, NJ
Email: aleri@princeton.edu