Received: 29 November 2016 Revised: 17 November 2017 Accepted: 20 November 2017 DOI: 10.1002/xrs.2826 RESEARCH ARTICLE Application of XANES spectroscopy to investigate Sb species in corroded bullets crust material oriented to evaluate the potential toxic effects in the environment Marcelo Rubio1 | María F. Mera2 | Carlos A. Pérez3 | Flavio C. Vicentin4 1 Unidad Estudios Físicos, CEPROCOR, Alvarez de Arenales 230 Barrio Juniors, Córdoba X05004AAP, Argentina 2 XRF Laboratory UEF, CEPROCOR, Alvarez de Arenales 230 Barrio Juniors, Córdoba X05004AAP, Argentina 3 Fluorescence and Absorption group, Brazilian Synchrotron Light Source, Caixa Postal 6192, Campinas, São Paulo CEP 13084‐971, Brazil 4 Centro Nacional de Pesquisa em Energia e Materiais Campinas, Brazil, Brazilian Synchrotron Light Source, Giuseppe Maximo Scolfaro, 10000, Campinas, São Paulo 13083970, Brazil Lead, antimony, and other toxic metals from pellets alloy are dispersed in the soil of the shooting fields. As long as the corroding bullets are present in soil, secondary Pb and Sb phases appear in the weathering crusts being an important source of bioavailable Pb and Sb. Knowledge about the corrosion mechanism of Sb from the bullet is limited, and reports on Sb speciation in crust and soils are still scarce. Considering that Sb species have different toxicological properties in the environment, this work has focused attention in X‐ray Absorption Near Edge Structure measurements at the Sb L‐edges in order to identify its chemical speciation in crust (Sb(0), Sb(III), and Sb(V)). X‐ray Absorption Near Edge Structure measurements were carried out at the D04A Soft X‐ray Spectroscopy beamline at the LNLS. Samples consisted of dust crust taken from physically deformed and strongly corroded metallic bullets Correspondence Marcelo Rubio, Unidad Estudios Físicos, CEPROCOR, Alvarez de Arenales 230 Barrio Juniors, X05004AAP Córdoba, Argentina. Email: mrubiocba@yahoo.com retained in soil samples sieving from shooting fields of the North and East Funding information LNLS ‐ National Synchrotron Light Laboratory, Campinas (Brazil), Grant/Award Number: Proposal SXS‐16953; CEPROCOR, Grant/Award Numbers: Government of Province of Córdoba CEPROCOR 2014 and CEPROCOR 2014; Brazilian National Synchrotron Light Laboratory (LNLS), Grant/Award Number: SXS‐16953 ronmental conditions, pentavalent Sb was the predominant species after region of Córdoba, Argentina. The results showed that the main species found in all samples were Sb(V) (Sb2O5) followed by Sb(0) (metallic). Sb(III) was not observed, and it is known that Sb(III) is more toxic than Sb(V). The results suggested that in these enviweathering of metallic Sb from corroding bullets. 1 | INTRODUCTION The Province of Córdoba is a region recognized worldwide due to dove hunting. Its northern territory has regions of native forests surrounded by grain production fields allowing a great population of doves. Every year, thousands of hunters release in the shooting fields ammunition made of approximately 90 wt% Pb and 2–6 wt% Sb X‐Ray Spectrometry. 2017;1–5. alloy with traces of other elements. Metals are not stable and progressively react to the soil, coating a crust of mineral phases over the surface of the spent bullets. During the weathering process, grains of crust removed from the coating become an important source of bioavailability of Sb and other toxic metals in the biosphere. For this reason, it is interesting to study the mineralogical composition and chemical speciation around corroded wileyonlinelibrary.com/journal/xrs Copyright © 2017 John Wiley & Sons, Ltd. 1 RUBIO 2 ammunitions to understand how the metallic Sb reacts in the soil influencing the bioavailability and the potential toxic effects on the environment. There are many scientific works dealing with soil contamination in walled shooting ranges,[1,2] but scarce information is available on contamination in hunting sites located in native forests. In a previous work, Rubio et al.[3] used X‐ray fluorescence analysis (XRF) to quantify Pb concentration in samples of soil, prepared by material collected in 315 sampling pits of 36 firing sites of native forest in the north of Córdoba, Argentina. In many countries, even in Argentina, regulation norms are oriented to lead but not antimony when it is necessary to screen this contaminant too. In Argentina, the intervention regulatory limits for agriculture soils are 20 ppm for Sb, whereas for Pb, the limit reach 375 ppm. Therefore, a contaminant source such as metallic ammunitions under weathering process may release to the environment Pb and Sb mineral compounds at a rate potentially equivalent, meaning that total Sb can reach the intervention limit for soils. This is not as straightforward to predict a priori and it depends on the composition of soils, weathering process, and meteorological conditions. Nevertheless, it should be a warning due to the potential carcinogenic properties and the progressive presence of Sb in the environment because of human activities. In a previous work, Mera et al.[4] characterized by synchrotron radiation micro X‐ray fluorescence analysis (SR μXRF), several samples of crust dust removed from pellets used for hunting, showing quantitative elemental analysis results. A typical composition for crust was: Pb ~50 wt%; Fe ~1.2 wt%; Si ~1.1 wt%; P ~1 wt%; Sb ~0.7 wt%; Ca ~0.5 wt%; K ~0.2 wt%; Ti ~0.1 wt%; and traces of other elements. Na and Al were not analyzed. In the same work, XRD patterns were obtained and the main crystalline mineralogical compounds of Pb and Sb quantified were HC: hydrocerussite (37.8%), Lit: litharge (1.6%), Lst: lead sulfate (1.8%), Lsd: lead sulfide (0.3%), Mar: margarosanite (0.4%), Hp: hydroxypyromorphite (4.4%), Gb: gebhardite (3.4%), Sh: shanonite (3.7%), AO: antimony oxide (1%), St: stibnite (2.5%), Tr: tripuhyite (0.8%), and Potassium Sodium Antimony Oxide (1.6%). In the referenced work, the authors concluded that a positive correlation was found between Sb and Fe in crust around cores of bullets. This is due to the high sorption affinity of Sb to Fe oxides present in rock parents as well as in organic iron compounds of the soil matrix. That study confirmed how Fe oxides act as important sinks for Sb. It was also shown that Sb is present in crust grains as elemental Sb, as antimony oxides, stibnite (Sb2S3), and tripuhyite (FeSbO4). Antimony and its compounds are considered by the United States Environmental Protection Agency and the ET AL. European Union as pollutants of priority interest due to their potential toxicity and carcinogenicity, and Sb is not cataloged as an essential element for plants or animals.[5] The mobility of antimony in soil and its interaction with plants and food crops depends on the speciation and mineralogy of the Sb–soil interaction.[6] The toxicity of Sb compounds is understood as arsenic, and it is the priority knowledge about the presence of Sb(V) or Sb(III) because of the carcinogenic risk of the Sb3+ species. It is known that inorganic species of Sb are more toxic than the organic ones, and Sb(III) compounds are more toxic than Sb(V) oxo‐anionic species. Indeed, the International Agency for Research on Cancer classifies Antimony trioxide (Sb2O3) as possibly carcinogenic to humans (Group 2B).[7] Moreover, Antimony in conjunction with other metals may produce several adverse effects, and the related health problems have still not been documented considerably. Therefore, research on the speciation, distribution, and remediation of sites contaminated with Sb is a need that has to be resolved in the short term. In this work, X‐ray Absorption Near Edge Structure (XANES) spectroscopy at the Sb L1‐edge was measured to identify its oxidation states in crust, and thus assess different toxicological properties of Sb species in the environment. 2 | MATERIAL A ND METHODS 2.1 | Soil material Soil material was collected in 36 hunting sites located in lands of the north of the Province of Córdoba. The sampling procedure consisted of collecting on each analyzed site at 50 mm depth in 10 GPS referenced points symmetrically distributed in a 100 × 100 m2 surface. Nine sampling points were equally distributed in the square perimeter and one in the center. The soil from the center point was taking at 5 and 15 cm depth. An aliquot of soil raft material from each point sampled was carefully prepared for multielemental quantitative XRF analysis as it is described by Rubio et al.[3] A selection of weathered bullets was prepared at CEPROCOR within the same process to provide the soil for pressed pellets samples for XRF analysis. Ammunition retained in soil material sieving (200 mesh) were separated, cleaned of external dust, and classified. The procedure to obtain crust powder consist of blending crust material from groups of approximately 15 weathered bullets collected in the sieve during the soil RUBIO ET AL. material sieving process. A time‐controlled mixer was used to shake each group of bullets. 2.2 | Antimony L1‐edge XANES analysis XANES spectroscopy is restricted to low energy range in relation to the binding energy of the element of interest. XANES spectra contain information related to the oxidation state and species of the atoms in the surrounding of the element of interest.[8] As oxidation state increases, the corresponding binding energy increases as well, giving different XANES spectra that allow identification of Sb(III), Sb(V), and Sb(0) for instance. Chemical speciation of Sb was carried out at the Soft X‐ray Spectroscopy (SXS) beamline of the Brazilian Synchrotron Light Laboratory (LNLS) in Campinas, Brazil, performing XANES measurements at the Sb L‐edge. The beamline is equipped with a toroidal Ni coated focusing mirror, a Si(111) double‐crystal monochromator and a high‐vacuum experimental chamber.[9] The corroded bullets crust material and reference standard Sb oxides (Sb2O3 and Sb2O5) were uniformly spread on double coated carbon conductive tape provided by Ted Pella, Inc. and mounted on a stainless steel sample holder, as well as on a standard Sb metallic foil (tin base white metal (BCS‐CRM 178/2) from British Chemical Standard). All the standards were measured at the SXS beamline using the same experimental setup of the samples. XANES measurements were performed in fluorescence mode for the samples and in Total Electron Yield mode for the standards. The energy scale was calibrated with a titanium metallic foil, setting the Ti K‐edge to 4966 eV. The intensity of the signal from L1‐edge is lower than the signal from L2,3‐edges, which results in low quality spectra with high noise. Even in this measuring conditions, the Sb L1 edge was used because it is rather sensitive to the oxidation state of antimony.[10] Therefore, each XANES spectrum consisted of an average of 8 to 15 individual scans. XANES spectra of the samples were collected in fluorescence mode using an Amptek X‐123 FIGURE 1 (a) Experimental L1‐edge X‐ ray Absorption Near Edge Structure spectra of Sb in crust from northwest zone corroded bullets (scatter dots) and reference standards materials for Sb(V), Sb(III), and Sb(0) oxidation states (short dash dot lines, short dot lines, and short dash). (b) Best linear combination fit to experimental data (solid line) 3 silicon drift detector, whereas XANES spectra of the standards were collected in total electron yield mode using an electrometer measuring the drain current. Synchrotron radiation was monochromatized by a double‐crystal monochromator equipped with Si(111) crystals, providing an energy resolution of 0.85 eV at the Sb L1‐edge (4,698 eV). All XANES spectra were scanned in the following energy regions: from 4,650 to 4,695 with a step size of 1 eV, from 4,695 to 4,720 with a step size of 0.2 eV, and from 4,720 to 4,820 eV with a step size of 1 eV. The integration time was 3 s per point and the detector dead time was kept below 10%. The final XANES spectra were obtained after background subtraction and normalization to the post‐edge intensity. Data analysis was performed using the Athena software in the computer package IFEFFIT.[11] The background was corrected by fitting a first‐order polynomial to the pre‐edge region from 4,650 to 4,690 eV, and the spectra were normalized over the reference energy of 4,706 eV. Linear combination fitting (LCF) analysis of Sb‐ XANES spectra for the crust samples, weighting with the standards spectra (Sb(III), Sb(V), and Sb(0)), was performed over the spectral region from 4,686 to 4,800 eV using Athena. The selected spectral region was chosen because it covers all of the spectral features for Sb species of interest in our study. 3 | R ESULTS A ND DISCUSSIONS Figure 1a shows normalized L1‐edge XANES spectra of Sb present in crust from bullets collected in the northwest zone of the hunting region (scatter dots), compared with standard materials of well‐known oxidation states (short dash, short dot, and short dash dot lines). Figure 1b shows the best LCF to experimental data (solid line). Under this curve, the two components (short dash and short dot lines) of the predominant antimony oxidation state are shown. From Figure 1b, it is possible to identify the two RUBIO 4 main Sb species composing the crust material (Sb(V) and metallic Sb), being antimony pentavalent Sb(V) the predominant oxidation state as shown from Athena calculations. Figure 2 shows four Sb L1‐edge XANES spectra. Experimental data (scatter dots) correspond to the measurement of crust material from bullets coming from the northeast zone of the hunting region. Table 1 shows the LCF results of normalized XANES reference spectra to the experimental spectra from northwest and northeast measured samples. Columns show the percentage fraction of each antimony oxidation state present in samples. From calculations, Sb(III) species were not detected as a percentage fraction lower than 4% of the total. Results from Athena fitting process show that antimony pentavalent oxidation state is predominant in all samples, with significant differences between northwest ET AL. or northeast hunting sites of the region studied. Metallic antimony represents approximately 38% of the species in the northwest samples, whereas in the northeast samples, Sb(0) fraction represents 52% of the total. Our results in crust are in good agreement with those from Scheinost et al.,[2] pointing that Sb(V) is the principal oxidation state after weathering of metallic Sb(0) in oxic soils, whereas Sb(III) fraction was not detected within the detection limit mentioned in this section. From the point of view of environmental toxicology, it is a remarkable result that the more toxic Sb(III) was not present in crust. Assuming that the pristine bullets used in the entire hunting area have on average the same chemical composition, the difference in percentages of Sb(V) and Sb(0) between both zones should be attributed to the nature of the soil and meteorological conditions. However, this is subject to another work based on the study of soil, agricultural land management, and meteorology (especially in the northeast region). This region is considerably more humid and cultivable than the semi‐arid hunting zone of the northwest. 4 | C ON C L U S I ON S FIGURE 2 L1‐edge X‐ray Absorption Near Edge Structure spectra of Sb in crust from the northeast zone corroded bullets and reference standards materials for Sb(0), Sb(V), and Sb(III) oxidation states TABLE 1 Linear combination fitting results of Sb XANES reference spectra onto the experimental spectra from northwest (N) and northeast (E) bullet crust samples. The values in columns are the percentage fraction (%) of each antimony oxidation state in the total Specie Samples Sb metal (%) Sb5O2 (%) Sb2O3 (%)a 1E 46 ± 5 54 ± 2 0±3 1N 34 ± 5 66 ± 2 0±3 2E 57 ± 5 43 ± 7 0±4 2N 43 ± 6 57 ± 3 0±4 3E 52 ± 4 48 ± 5 0±3 A C KN O WL ED G EME N T S 3N 36 ± 6 64 ± 2 0±3 This work was partially supported by the Brazilian National Synchrotron Light Laboratory (LNLS), under the proposal SXS‐16953 and by CEPROCOR, Córdoba Note. XANES = X‐ray Absorption Near Edge Structure. a Species of Sb were determined by synchrotron radiation XANES analysis on crust material removed from weathered bullets, collected in two different geographic zones within the dove hunting region of Córdoba. Three Sb species were measured, although only two were detected and quantified: Sb(0) and Sb(V). The fraction (in percent) of Sb(III) species was not detected in the analyzed samples, considering a detection limit of 4% for presence relative of this species in the total. It means that Sb(III) could be present in a fraction lower than 4% of Sb(V) + Sb(0). It can be concluded that geomorphological changes, different environmental conditions, different forms of soil management, and the proximity of hunting sites to the cultivated areas are factors that differentiate the presence of the detected phases Sb(0) and Sb(V) between the northwest and northeast hunting zones. The results suggested that in these conditions, pentavalent Sb was the species of antimony oxide of greater growth in the corroding process of metallic Sb from the weathering bullets. Values according to the best Linear Combination Fitting (LCF) results. RUBIO ET AL. 5 (Argentina), under the regular budget of the Government of the Province of Córdoba/CEPROCOR 2014. ORCID Marcelo Rubio http://orcid.org/0000-0001-8052-9657 R EF E RE N C E S [6] D. Amarasiriwardena, F. Wu, Microchem. J. 2011, 97, 1. [7] IARC International Agency for Research on Cancer (IARC)‐ Summaries & Evaluations, 47, 1989, 291. [8] G. Bunker, Introduction to XAFS: A Practical Guide to X‐ray Absorption Fine Structure Spectroscopy, February 2010 edition, Cambridge University Press, ISBN‐10 052176775X, ISBN‐13: 978‐0521767750 2010 270. [9] H. Tolentino, V. Compagnon‐Cailhol, F. C. Vicentin, M. Abbate, J. Synchrotron Radiat. 1998, 5, 539. [1] D. Dermatas, X. Cao, V. Tsaneva, G. Shen, D. G. Grubb, Water, Air, and Soil Pollution: Focus 2006, 6, 143. [10] J. Rockenberger, U. Zum Felde, M. Tischer, L. Tröger, M. Haase, H. Weller, J. Chem. Phys. 2000, 112(9), 4296. [2] A. C. Scheinost, A. Rossberg, D. Vantelon, I. Xifra, R. Kretzschmar, A. K. Leuz, H. Funke, C. A. Johnson, Geochim. Cosmochim. Acta 2006, 70, 3299. [11] B. Ravel, M. Newville, J. Synchrotron Radiat. 2005, 12, 537. [3] M. Rubio, A. Germanier, M. F. Mera, S. N. Faudone, R. D. Sbarato, J. M. Campos, V. Zampar, E. Bonzi, C. A. Pérez, X‐ Ray Spectrom. 2014, 43, 186. [4] M. F. Mera, M. Rubio, C. A. Pérez, V. Galván, A. Germanier, Microchem. J. 2015, 119, 114. [5] C. Bini, J. Bech, PHEs, Environment and Human Health: Potentially harmful elements in the environment and the impact on human health, May 14, 2014 edition, Springer, ISBN‐13:978‐ 9401789646, ISBN‐10:9401789649 2014 467. How to cite this article: Rubio M, Mera MF, Pérez CA, Vicentin FC. Application of XANES spectroscopy to investigate Sb species in corroded bullets crust material oriented to evaluate the potential toxic effects in the environment. X‐Ray Spectrometry. 2017. https://doi.org/10.1002/ xrs.2826