Earth and Planetary Science Letters 246 (2006) 458 – 465 www.elsevier.com/locate/epsl
Arsenic incorporation in natural calcite lattice: Evidence from electron spin echo spectroscopy Francesco Di Benedetto a,b,⁎, Pilario Costagliola a , Marco Benvenuti a , Pierfranco Lattanzi c , Maurizio Romanelli d , Giuseppe Tanelli a a
Dipartimento di Scienze della Terra, Università di Firenze, Italy b Museo di Storia Naturale, Università di Firenze, Italy c Dipartimento di Scienze della Terra, Università di Cagliari, Italy d Dipartimento di Chimica and CSGI, Università di Firenze, Italy Received 12 January 2006; received in revised form 28 March 2006; accepted 31 March 2006 Available online 22 May 2006 Editor: G.D. Price
Abstract Quaternary travertines of the Middle Pecora Valley (Tuscany, Italy) contain up to 257 ppm arsenic. Such a content is environmentally relevant, but low enough to make the exact chemical speciation of arsenic difficult by applying conventional investigation techniques. The task was addressed by use of the Electron Spin Echo (ESE) spectroscopy, taking advantage from the modulation by the arsenic nucleus of the decay spectrum of the paramagnetic ion Mn(II), occurring as replacement of Ca in the calcite lattice. Interpretation of the spectra suggests that arsenic occurs in the calcite lattice in the position of C, through the 3− substitution CO2− 3 ⇔ AsO3 . This mechanism of arsenic incorporation by calcite may be an effective limit of arsenic mobility under conditions where immobilization through sorption by iron and/or manganese oxyhydroxides is not operating. © 2006 Elsevier B.V. All rights reserved. Keywords: arsenic; travertine; electron spin echo spectroscopy; calcite
1. Introduction Arsenic contamination represents a serious environmental problem in many parts of the world [1–3]. Specifically, arsenic contamination of ground water has been recognised by far as one of the major threats for public health. The growing concern about the effects on human health of this metalloid has encouraged a wide range of studies on the chemical and physical state of ⁎ Corresponding author. Museo di Storia Naturale, Università di Firenze, Via G. La Pira, 4 - I 50121, Firenze, Italy. Tel.: +39 055 275 6349; fax: +39 055 284571. E-mail address:
[email protected] (F. Di Benedetto). 0012-821X/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.epsl.2006.03.047
arsenic in natural environment, because this feature controls the mobility of the element and, ultimately, its intake in the human food-chain [4–6]. Much attention has been paid to sorption–desorption reactions between arsenic and mineral surfaces, especially with Fe- and Mn-oxyhydroxides [7,8]. This is considered the most important and diffused mechanism controlling arsenic mobility and determining its release to groundwaters [9]. Comparatively, the influence of calcite on arsenic mobility has received less attention, in spite of the wide diffusion of this mineral at the Earth's surface [10,11]. With few exceptions (e.g. [12]), natural carbonates do not contain appreciable amounts of arsenic, thus limiting or hindering its determination by the most common
F. Di Benedetto et al. / Earth and Planetary Science Letters 246 (2006) 458–465
techniques used for investigating the micro-environment of this metalloid in minerals. In addition, natural samples quite often display a complex matrix, that inevitably makes analysis more difficult. As a consequence, the knowledge available in literature on arsenic speciation in calcite comes by far from studies on samples prepared under laboratory conditions. Hence, implications for natural systems result from extrapolation of these experiments, and not from direct evidence. In order to investigate the possibility of arsenic incorporation in natural calcite, we focussed our attention toward natural systems where calcite precipitation occurs in an As-bearing environment. A special interest has been dedicated to surface systems (at low temperature and pressure), where in some instances (cf. e.g. [3]) arsenic is presumably present not only in the
459
pentavalent, but also in the trivalent state that is considered to be more dangerous for environment and human health [5]. Quaternary travertines cropping out in the Middle Pecora Valley (MPV) in Southern Tuscany (Italy) meet all the above mentioned requirements, thus offering a unique opportunity to investigate the occurrence of arsenic in natural, albeit fossil, systems. These travertines, in fact, are known to contain appreciable amounts of arsenic (up to 257 ppm; [13,14]. They were deposited in a lacustrine environment, where redox conditions may stabilize both trivalent and pentavalent arsenic [15,16]. The arsenic content of these travertines is environmentally relevant, but low enough to make difficult the determination of the exact chemical speciation of arsenic with most routine mineralogical
Fig. 1. Schematic geological map of the study area.
460
F. Di Benedetto et al. / Earth and Planetary Science Letters 246 (2006) 458–465
techniques. This challenging task was addressed by an unconventional use of the Electron Spin Echo (ESE) spectroscopy. This is an Electron Paramagnetic Resonance (EPR) spectroscopy performed with pulsed microwave (∼ 9.5 GHz) radiation. This technique had so far only limited applications in Earth and environmental sciences, in spite of its remarkable potential [17]. As detailed below, arsenic was revealed through its interaction with a paramagnetic impurity in calcite, i.e. Mn(II). 2. Geological outline Southern Tuscany is a geologic district belonging to the Northern Apennines tectonic chain. This results from the superimposition of several compressive stages spanning from Late Cretaceous to Middle Miocene [18]. In Late Miocene, in a complex structural environment where both extensional and compressive structures are documented, the emplacement of magmatic rocks took place [19,20]. Moreover, an intense hydrothermal circulation led to the formation of several mineral deposits [21,22], and a number of post-nappe basins were filled by clastic sediments. They occupy much of the low elevation areas in Southern Tuscany, including the Pecora Valley (Fig. 1). Here, a Neo-autochtonous (Quaternary) alluvial–colluvial sediment pile, deposited by the palaeo- and present-day Pecora watercourse, covers a transitional clastic Late Miocene sequence, overlying the Apenninic orogenic rock pile. The alluvial–colluvial sediments are locally covered by lacustrine and phytoclastic travertines (meteogenic travertine; [23]). They extensively crop out in the lowermost areas of the Middle Pecora Valley (MPV).
Fig. 2. Outcrop view of MPV travertines: plant structures.
Hence travertines represent the topmost formation of the Quaternary succession of the area, but occupy the lowest topographic positions. At the depocenter, the travertine sequence reaches the maximum depth of about 30 m. Typically, MPV travertines are constituted by layers of irregular thickness, composed mainly by calcite micro-crystals, cementing variable amounts of other minerals, including iron oxyhydroxides, that confer to the bulk rock a brownishyellow color. Non-carbonate minerals may be found as thin centimetric layers, testifying the siliciclastic feeding of the travertine lake. Phytoclastic textures are common. They are represented by remnants of travertine encrustations around lacustrine plants, and by cavities of variable size (from few millimetres up to some centimetres), that preserve the shape of plant stems (Fig. 2). 3. Experimental procedures Seven MPV travertine samples (La Forra locality) were studied by optical microscopy and X-ray diffraction. Quantitative chemical analyses of major elements (Na, Mg, Al, Si, P, K, Ca, Ti, Mn, Fe) were performed by X-ray fluorescence using a Philips PW1480/10 apparatus. The concentrations of minor cations were determined by flame Atomic Absorption Spectrometry (AAS) using a Perkin Elmer AAnalist100 Spectrophotometer. Analytical solutions were obtained after aqua regia leaching of powdered samples. Cu, Zn and Pb required leaching at 70–80°C, whereas arsenic was leached with a less drastic treatment at 40–50°C in a sand bath for 4 h. The arsenic solutions were analyzed by Hydride Generation AAS using the same spectrophotometer, equipped with a Perkin Elmer FIAS 100 Hydride Generator. The analytical quality was controlled by using international standards (RTS-1, RTS-2, RTS-3 and RTS-4; [24]). The relative difference between our results and the certified arsenic contents in the standards was b 10%. Arsenic never occurs in a paramagnetic state in natural samples, thus preventing its direct investigation by conventional EPR technique. However, the 75As isotope (natural abundance = 100%) possesses an active magnetic nucleus (nuclear spin quantum number: I = 3/2), thus enabling its detection in ESE experiments, through the nuclear modulation of the ESE signal, the so-called ESEEM, provided that in the investigated phase a paramagnetic ion, acting as a “probe” or “tool”, is present. The interpretation of the nuclear modulation allows to characterize the crystal chemical surrounding of paramagnetic centers in terms of types of nuclei, of their distances and of the hyperfine and quadrupole interactions. Manganese is a common minor element in calcite
F. Di Benedetto et al. / Earth and Planetary Science Letters 246 (2006) 458–465
(see e.g. [25–27]), where it occurs as paramagnetic Mn (II) substituting for Ca. Indeed, the EPR spectrum of Mn (II) ions in calcite is ubiquitously detected, and may represent a fingerprint of trace calcite presence in sediments [28,29]. In our case, therefore, the presence of arsenic in the calcite structure was investigated by probing the chemical environment surrounding the Mn (II) ion. This approach presents several advantages; among them, the selective study of the arsenic–calcite relationships and the removal of any possible interference from arsenic in other minerals (e.g., adsorbed onto iron oxyhydroxides). As a consequence, no preliminary mineral separation and/or chemical leaching were necessary to obtain the spectral information. ESE spectra were preceded by conventional continuous-wave EPR measurements. Both experiments were directly run on raw powdered samples, without any chemical or physical pre-treatment. Continuous-wave EPR spectra were registered at room and liquid He temperature at X-band (∼ 9.5 GHz), by using a Bruker ELEXSYS E500 spectrometer equipped with a continuous flow liquid He cryostat for variable temperature measurements. ESE spectra were registered through a Bruker ESP380 spectrometer, equipped with a Bruker ER 4118 dielectric resonator. Pulsed experiments were run both in the field-domain (Echo-EPR) and in the time-domain (two-pulse Hahn echo, 2p), at T = 4.2° K with operating frequency ν = 9.717 GHz. Based on analysis of the field-domain spectra, time-domain 2p experiments were carried out at the magnetic field value of 0.3225 T, to improve the signal-to-noise ratio. The quantitative interpretation of the ESE spectra and the refinements of the relative parameters were performed by means of full-profile spectral simulations. ESE nuclear modulation (ESEEM) and decay were simulated by means of self assembled FORTRAN software. 4. Results In thin section, travertines are constituted by finegrained calcite with frequent cavities of different sizes, fossils, and detritic quartz grains. From a descriptive point of view they share the same general characteristics of the meteogene travertine of lake facies (lacustrine travertine) described by Pentecost [23]. The complete mineral assemblage, obtained by complementing optical microscopy with X-ray diffraction, includes, beside calcite and variable amounts of quartz, traces of phyllosilicates. Furthermore, SEMEDS allowed the recognition of small crystals of Casulphates, iron-hydroxides, and Ti-oxides. In one case, SEM-EDS semiquantitative analyses pointed out the
461
Table 1 Chemical compositon of MPV travertine samples F1a LOI Na2O MgO Al2O3 SiO2 P2O5 K2O CaO TiO2 MnO Fe2O3 FeO Y Rb Nb Sr Ce La Ba Cd Ag Pb Bi Zn Cu Sb As
F1b
Tr 1
Tr 4
Tr 6
36.05 34.37 33.33 34.65 38.90 0.18 0.23 0.22 0.16 0.10 1.57 1.65 1.87 1.75 1.43 6.92 7.49 10.01 9.11 4.37 14.68 17.28 19.74 17.04 9.25 0.03 0.03 0.04 0.02 0.01 1.02 1.21 1.59 1.29 0.64 37.33 35.16 30.17 33.53 43.92 0.19 0.23 0.25 0.19 0.11 0.07 0.07 0.06 0.05 0.05 1.97 1.97 2.73 2.21 1.22 0.32 tr tr 14 5 tr 55 67 77 58 28 tr tr tr tr tr 1371 1246 1059 1160 1562 32 42 36 44 34 28 37 5 16 tr 128 140 193 149 83 b5 b5 b5 b5 b5 b5 6 b5 b5 b5 23 46 67 b5 b5 b5 b5 b5 b5 b5 214 273 246 163 184 36 48 89 b5 b5 9 9 b5 b5 b5 206 257 191 142 198
Tr 7
Tr 8
Average
40.81 39.35 36.78 0.08 0.18 0.16 1.22 1.43 1.56 2.47 4.34 6.39 5.67 9.07 13.25 0.01 0.01 0.02 0.37 0.61 0.96 48.52 43.66 38.90 0.07 0.11 0.16 0.03 0.04 0.05 0.76 1.20 1.72 tr 16 tr 1485 19 217 60 b5 b5 b5 b5 239 b5 b5 127
tr 25 tr 1334 21 213 85 b5 b5 b5 b5 122 b5 b5 148
tr 47 tr 1317 33 74 120 tr tr tr tr 206 tr tr 195
The chemical contents in the upper rows (from LOI to FeO) are expressed in wt.%; the others in μg/g.
presence of small Fe-oxide fragments with high arsenic contents (up to about 6 % by weight). The chemistry of the travertine is reported in Table 1. As expected, calcium is by far the most abundant element, but the appreciable amounts of SiO2 and Al2O3 are in agreement with the presence of quartz and phyllosilicates. Variable As-contents are found to range from 127 to 257 ppm. A positive correlation between arsenic with both SiO2 and Fetot is also observed (Fig. 3a, b). In Fig. 4, the Echo-EPR spectrum of the Mn(II) ion is shown, as the echo intensity versus the applied magnetic field. The six hyperfine lines, characteristic of Mn(II) (nuclear spin quantum number: I = 5/2) are clearly evident, with isotropic hyperfine coupling of ∼94 G, typical for calcite [28]. Moreover, ten forbidden lines, characteristic of Mn(II) substituting for Ca in the calcite lattice [30], are also evident. The broad band underlying the sextet belongs to outer transitions, between the ± 5/ 2 ⇔ ± 3/2 and ± 3/2 ⇔ ±1/2 spin states (Mn(II) electron spin quantum number: S = 5/2). The 2p spectrum (Fig. 5) consists of a fast decay component, which is almost completely suppressed at 3.5 μs, and of superimposed nuclear modulations, ascribable to
462
F. Di Benedetto et al. / Earth and Planetary Science Letters 246 (2006) 458–465
Fig. 3. a, b — Binary plots of As vs. SiO2 (a) and As vs. total Fe (b) contents in MPV samples. (Fetot = FeO + Fe2O3).
the presence of magnetic nuclei surrounding the Mn(II) ions. Two very short periods are present, superimposed to a longer one, as shown in the inset of Fig. 5. Due to the fast oscillating components, the Fourier Transform (FT) appears dominated by the nuclear quadrupole contribution, thus requiring the necessity of full spectral simulation. Moreover, an important feature of the FT of the ESE decay spectrum is the absence of the nuclear H
modulation (see following discussion). Among elements listed in Table 1, those who possess magnetic nuclei, possibly giving rise to the nuclear modulation, are Na, Al, P, K, Rb, Nb, La, Ag, Bi, Cu, Sb, Y and As. Aluminium and the alkaline ions are most likely linked to phyllosilicates, whereas Y, Nb, Ag, Sb and Bi occur in such low contents that they are not considered to be relevant. Finally, Cu has a distinctive isotopic fingerprint ratio [31],
Fig. 4. Echo-EPR spectrum showing the signal of Mn(II) in calcite; the x arrow indicates the field position of the ESE decay experiments.
F. Di Benedetto et al. / Earth and Planetary Science Letters 246 (2006) 458–465
463
Fig. 5. ESE decay spectrum, evidencing the 75As nuclear modulation. In the inset detail of the spectrum, showing the fast-periodic nuclear modulation components.
which is not observed in our spectra. Only P (I = 1/2), La (I = 7/2) and As (I = 3/2) may be expected, therefore, to occur in calcite. The first must be excluded because it has no nuclear quadrupole moment. The two others, therefore, have to be taken into account. Hence, numerical simulations of the experimental nuclear modulation were devoted to refine the geometrical properties, the coordination and the nature of interaction between Mn(II) and the magnetic nuclei interacting with it. Field-domain spectra show Ca(II) substituted by Mn(II) in pseudooctahedral sites. The ESEEM patterns were simulated by numerical diagonalisation of the nuclear hamiltonian, with trial parameters concerning distance, isotopic hyperfine and nuclear quadrupole constants. Preliminary simulations excluded the presence of magnetic nuclei at distances N 4 Å, thus leaving 75As nuclei, located in the trigonal C sites, as the possible source of ESEEM. A Ca– La substitution [32], in fact, should occur in to adjacent octahedra giving rise to an hypothetical distance Mn–La ∼4 Å. Moreover, the slight distortion of the octahedral sites in calcite does not match with the high quadrupole coupling constant needed to fit the modulations. Refinements excluded also misalignment between dipolar hyperfine and quadrupole axis systems. Thus, the experimental patterns are best simulated assuming a remarkable quadrupole, due to the asymmetry of the chemical coordination, and a distance of 3.2 Å. The obtained best-fit parameters indicate that Mn(II) ions interact with a single nucleus of 75As located at the distance correspon-
ding to the Ca–C distance in the calcite structure. Moreover, the existence of a hyperfine interaction implies a chemical bond between Mn(II) and As. The magnetic nucleus of the 75As isotope was thus ascertained to modulate the two-pulse ESE spectrum of Fig. 5. It is noteworthy that the experimental Mn–As distance is fully comparable with the Ca–C distance in the calcite structural model, thus indicating the C site as host of the arsenic atoms. The existence, on the other hand, of a large axial quadrupole interaction was assessed, confirming arsenic to be in an axially symmetric chemical environment. 5. Discussion and conclusions The scanty literature data on the ability of carbonate to adsorb arsenic species consistently indicate that Asoxyanions can be adsorbed on calcite, due to its positively charged surface at low pH. Sadiq [33], suggests that, at pH between 7.5 and 9, carbonate may play an important role for arsenic adsorption in soils. Goldberg and Glaubig [34] determined a strong arsenate adsorption on calcite, peaking at relatively high pH level (about 10.5), possibly in response of the weakly acidic behaviour of arsenic acid. On the other hand, recent studies, conducted at comparatively lower pH (6.0 to 9.5), point out that, in laboratory conditions simulating natural water–rock interaction, calcite behaves as a minor adsorbent for arsenates [11]. As far as arsenite is concerned, laboratory
464
F. Di Benedetto et al. / Earth and Planetary Science Letters 246 (2006) 458–465
experiments reproducing the interaction between arsenite and freshly cleaved calcite surface using a H3AsO3 solution having an initial pH of about 9.8 indicate that arsenic may be incorporated by calcite through a substitution of CO32− by AsO33− groups [10]. This happens in spite of an unfavourable geometry, because the arsenite ion has a pyramidal trigonal shape, due to the nonbonding valence shell electron pair, instead of the planar trigonal shape of the carbonate group. Recent research emphasised the As–C substitution in the calcite lattice as a very promising tool for the mitigation of arsenic poisoning, since calcite may incorporate large amounts of arsenic (up to 8% by weight; [35]). Removal of arsenic by calcite and/or gypsum in a lacustrine environment was advocated by Roman-Ross et al. [15]. All these results provide evidences that arsenic may be incorporated in the calcite lattice, and support the idea that this process may take place also in nature. This behaviour has some relevant consequences on arsenic mobility under natural surficial systems, since arsenic immobilization into solid phases occurs mostly under oxidizing conditions and neutral to slightly acidic pH, by interaction with Fe (and Mn) oxides, through coprecipitation/adsorption processes [3,36–38]. On the other hand, arsenic can remain relatively mobile under a broad range of redox potentials, particularly at the neutral to slightly alkaline pH values typically found in surface and groundwaters [3]. Since calcite is generally stable under alkaline pH, Ca-carbonates may account for arsenic immobilization where Fe (and Mn) oxides lose their adsorbing effectiveness. The Echo-EPR spectrum confirms that Mn(II) occurs in a structural Ca position within the calcite crystal lattice, because of the absence of spectral features ascribable to near-surface lattice defects and/or distortion. Likewise, the absence of the nuclear H modulation in the ESE decay spectrum rules out any possibility that this nucleus, an ubiquitous external interference in ESE surface experiments [39], occurs within 10 Å from Mn (II). As a consequence, all experimental results point to interactions between As and paramagnetic Mn(II) ions within the mineral lattice, but far from its surface. Concerning the arsenic nuclear modulation revealed by the Mn(II)–As interaction, the numerical simulations allowed the assignment of arsenic to a structural C site, thus 3− suggesting a replacement mechanism CO2− 3 ⇔ AsO 3 . This implies that As occurs in the trivalent state, although ESE is not able to directly determine the arsenic valence in calcite. This feature is only inferred based on the distance between arsenic and Ca(Mn), which is fully compatible with the Ca–C distance in calcite, and the large quadrupole value, expected for a trigonal coordination of arsenic.
Combining the proximity of Mn(II) and arsenic, and the distance of the former from the surface, the 3− CO2− 3 ⇔ AsO 3 replacement cannot be considered as a surface capping, but a limited substitution mechanism occurring in the bulk mineral. The origin of arsenic in the MPV travertine is still under study. As a preliminary result, the positive linear correlation between As and SiO2 (Fig. 3) seems to indicate that arsenic was supplied to the travertine lake in association with the siliciclastic sediment [14]. The positive correlation between As and Fe (tot) and the SEMEDS microanalysis provides a direct evidence of the association of the metalloid with iron-oxyhydroxides, that were a component of the sediments discharged in the travertine lake. Mineralogical and chemical data, therefore, point to a clastic (alluvial) feeding of the travertine lake, where arsenic is transported associated to (possibly adsorbed onto) Fe oxyhydroxides. It is likely that the lake was characterized by a slightly alkaline pH [40], and by low redox potential (which stabilizes arsenites at the expense of arsenates). The combined effect of these two parameters favoured the desorption of the arsenic from the iron-oxyhydroxide surface, making it available for coprecipitation with calcite. In conclusion, this study demonstrates that in natural systems calcite may incorporate in its lattice arsenic as arsenite, thus hindering the mobility of the metalloid under physico-chemical conditions where iron-oxyhydroxide effectiveness in immobilizing arsenic is low. The ESE spectroscopy proved to be a key analytical tool to reach this conclusion, and may have additional applications in Earth and environmental sciences. Acknowledgements We thank for providing data and assisting us in our work Pierluigi Parrini, Elena Pecchioni, Mario Paolieri, Sara Bianchi, Miria Borgheresi, Marina Brustolon, Alfonso Zoleo, Antonio Barbon and Silvia Vettori. Two anonymous referees are also warmly acknowledged for their suggestions, which improved the readability and the understanding of the text. This work was supported by MURST funds (PRIN 2004). References [1] J. Matschullat, Arsenic in the geosphere — a review, Sci. Total Environ. 249 (2000) 297–312. [2] B.K. Mandal, K.T. Suzuki, Arsenic round the world: a review, Talanta 58 (2002) 201–235. [3] P.L. Smedley, D.G. Kinniburgh, A review of the source, behaviour and distribution of arsenic in natural waters, Appl. Geochem. 17 (2002) 517–568.
F. Di Benedetto et al. / Earth and Planetary Science Letters 246 (2006) 458–465 [4] J. Nriagu, Arsenic in the Environment. Part I: Cycling and Characterization, Wiley-Interscience Publication, NY, 1994. [5] J. Nriagu, Arsenic in the Environment. Part II: Human Health and Ecosystem Effects, Wiley-Interscience Publication, NY, 1994. [6] C.K. Jain, I. Ali, Arsenic: occurrence, toxicity and speciation techniques, Water Res. 34 (2000) 4304–4312. [7] V. Lenoble, O. Bouras, V. Deluchat, B. Serpaud, J.C. Bollinger, Arsenic adsorption onto pillared clays and iron oxides, J. Colloid Interface Sci. 255/1 (2002) 52–58. [8] S. Ouvrard, P. De Donato, M.O. Simonnot, S. Begin, J. Ghanbaja, M. Alnot, Y.B. Duval, F. Lhote, O. Barres, M. Sardin, Natural manganese oxide: combined analytical approach for solid characterization and arsenic retention, Geochim. Cosmochim. Acta 69 (2005) 2715–2724. [9] D.M. Sherman, S.R. Randall, Surface complexation of arsenic (V) to iron(III) hydroxides: structural mechanism from ab initio molecular geometries and EXAFS spectroscopy, Geochim. Cosmochim. Acta 67 (2003) 4223–4230. [10] L. Cheng, P. Fenter, N.C. Sturchio, Z. Zhong, M.J. Bedzyk, X-ray standing wave study of arsenite incorporation at the calcite surface, Geochim. Cosmochim. Acta 63 (1999) 3153–3157. [11] C. Le Guern, P. Baranger, C. Crouzet, F. Bodénan, P. Conil, Arsenic trapping by iron oxyhydroxides and carbonates at hydrothermal spring outlets, Appl. Geochem. 18 (2003) 1313–1323. [12] T.W. Horton, J.A. Becker, D. Craw, P.O. Koons, C.P. Chamberlain, Hydrothermal arsenic enrichment in active mountain belt: Southern Alps, New Zealand, Chem. Geol. 177 (2001) 323–339. [13] P. Costagliola, M. Benvenuti, M.G. Benvenuti, A. Innocenti, I. Mascaro, M. Paolieri, L. Rossato, G. Tanelli, Arsenic distribution in the Quaternary sediments of the median valley of the Pecora stream (Grosseto, Italy) Brownfields, in: A. Donati, C. Rossi, C.A. Brebbia (Eds.), WIT Press Southampton, UK, 2004, pp. 201–209. [14] P. Costagliola, M. Benvenuti, M.G. Benvenuti, S. Bianchi, M. Borzoni, F. Di Benedetto, A. Innocenti, M. Paolieri, L. Rossato, G. Tanelli, The arsenic anomaly of the Pecora Valley (Scarlino, Grosseto): interplay between Quaternary geology and minerogenic processes, Quinto Forum Scienze della Terra, Spoleto 21– 23 Settembre 2005. [15] G. Roman-Ross, L. Charlet, G.J. Cuello, D. Tisserand, Arsenic removal by gypsum and calcite in lacustrine environments, J. Phys., IV Fr. 107 (2003) 1153–1156. [16] R.S. Oremland, J.F. Stolz, J.T. Hollibaugh, The microbial arsenic cycle in Mono Lake, California, FEMS Microbiol. Ecol. 48 (2004) 15–27. [17] F. Di Benedetto, G.P. Bernardini, A. Caneschi, C. Cipriani, C. Danti, L.A. Pardi, M. Romanelli, EPR and magnetic investigations on sulphides and sulphosalts, Eur. J. Mineral. 14 (2002) 1053–1060. [18] M. Bonini, F. Sani, Extension and compression in the Northern Apennines (Italy) hinterland: evidence from the late Miocene– Pliocene Siena–Radicofani Basin and relations with basement structures, Tectonics 21 (2002) 1–35. [19] G. Serri, F. Innocenti, P. Manetti, S. Tonarini, G. Ferrara, Il Magmatismo Neogenico–Quaternario dell'area Tosco-LazialeUmbra: implicazioni sui modelli di evoluzione geodinamica dell'Appennino Settentrionale, Studi Geol. Camerti, Vol. Spec. 1991/1 (1991) 429–463. [20] M. Boccaletti, M. Bonini, G. Moratti, F. Sani, Tectonosedimentary evolution of the post nappe basins of the Northern Apennines, Notes Mém. Serv. Géol. Maroc 387 (1996) 177–193. [21] G. Tanelli, Mineralizzazioni metallifere e minerogenesi della Toscana, Mem. Soc. Geol. 25 (1983) 91–109.
465
[22] P.F. Lattanzi, M. Benvenuti, P. Costagliola, G. Tanelli, An overview on recent research on the metallogeny of Tuscany, with special reference to the Apuane Alps, Mem. Soc. Geol. Ital. 48 (1994) 613–625. [23] A. Pentecost, The Quaternary travertine deposits of Europe and Asia Minor, Quat. Sci. Rev. 14 (1995) 1005–1028. [24] CCRMP, Canadian Certified Reference Materials Project, Certificate of Analysis, RTS-1, RTS-2, RTS-3 and RTS-4 sulphide ore mill tailings reference materials, Ottawa 1996. [25] R.V. Gaines, H.C. Skinner, E.E. Foord, B. Mason, A. Rosenzweig, Dana's New Mineralogy, John Wiley and Sons, NY, 1997. [26] L. Cheng, N.C. Sturchio, M.J. Bedzyk, Impurity structure in a molecular ionic crystal: Atomic-scale X-ray study of CaCO3: Mn2+, Phys. Rev., B 63 (2001) 144104–144109. [27] D. Freund, Z. Wang, E. Rybacki, G. Dresen, High-temperature creep of synthetic calcite aggregates: influence of Mn-content, Earth Planet. Sci. Lett. 226 (2004) 433–448. [28] F. Di Benedetto, G. Montegrossi, L.A. Pardi, A. Minissale, M. Paladini, M. Romanelli, A multifrequency EPR approach to travertine characterisation, J. Magn. Reson. 177 (2005) 86–92. [29] G. Montegrossi, F. Di Benedetto, A. Minissale, M. Paladini, L.A. Pardi, M. Romanelli, F. Romei, Determination and significance of the Mn(II) Zero-Field Splitting (ZFS) interaction in the geochemistry of travertines, Appl. Geochem. (in press), doi: 10.1016/j.apgeochem.2006.02.008. [30] J.A. Angus, B. Raynor, M. Robson, Reliability of experimental partition coefficients in carbonate systems: evidence for inhomogenous distribution of impurity cations, Chem. Geol. 27 (1979) 181–205. [31] A. Abragam, B. Bleaney, Electron Paramagnetic Resonance of Transition Ions, Clarendon press, Oxford, 1970. [32] E.J. Elzinga, R.J. Reeder, S.H. Withers, R.E. Peale, R. Mason, K. M. Beck, W.P. Hess, EXAFS study of rare-earth element coordination in calcite, Geochim. Cosmochim. Acta 66 (2002) 2875–2885. [33] M. Sadiq, Arsenic chemistry in soils: an overview of thermodynamic predictions and field observation, Water Air Soil Pollut. 95 (1997) 117–136. [34] S. Goldberg, R.A. Glaubig, Anion sorption on a calcareous montmorillonitic soil — arsenic, Soil Sci. Soc. Am. J. 52 (1988) 1297–1300. [35] ILL (Institute Laue Langevin) press release, http://www.ill.fr/ pages/press/gb/press_releases/Arsenic.htm (2005). [36] J.H. Ryu, S. Gao, R.A. Dahlgren, R.A. Zierenberg, Arsenic distribution, speciation and solubility in shallow groundwater of Owens Dry Lake, California, Geochim. Cosmochim. Acta 66 (2002) 2981–2994. [37] B.J. Moldovan, D.T. Jiang, M.J. Hendry, Mineralogical characterization of arsenic in uranium mine tailings precipitated from iron-rich hydrometallurgical solutions, Environ. Sci. Technol. 37 (2003) 873–879. [38] P. Brandenberger, B. Louchouarn, H.P. Tissot, Geochemical and hydrodynamic controls on arsenic and trace metal cycling in a seasonally stratified US sub-tropical reservoir, Appl. Geochem. 19 (2004) 1601–1623. [39] M. Borgheresi, G.P. Bernardini, C. Cipriani, F. Di Benedetto, M. Romanelli, Electron paramagnetic resonance and electron spin echo spectroscopy study of natural bornite, Min. Pet. 85 (2005) 3–18. [40] N. Horvatincic, I. Krajcar Bronic, B. Obelic, Differences in the 14 C age, δ13C and δ18O of Holocene tufa and speleothem in the Dinaric Karst, Palaeogeogr. Palaeoclimatol. 193 (2003) 139–157.
