Aqueous speciation is likely to control the stable isotopic fractionation of cerium at varying pH

Abstract

Cerium (Ce) can be used as a plaeoredox proxy as shown by a recent study of stable isotopic fractionation of Ce during adsorption and precipitation. However, the experiments in that study were performed at pH conditions lower than that of natural seawater. In the current study, adsorption and precipitation experiments were performed at pH 6.80, 8.20, and 11.00 with 2.25 mM dissolved carbonate to simulate Ce isotopic fractionation in the natural environment and examine the relationship between isotopic fractionation and Ce speciation in the liquid phase. Mean isotopic fractionation factors between liquid and solid phases (α_Lq-So) of Ce adsorbed on ferrihydrite did not depend on pH conditions or dissolved Ce species. In the Ce/δ-MnO₂ system,α_Lq-So values decreased from 1.000411 (±0.000079) to 1.000194 (±0.000067) with increasing pH or number of carbonate ions, from Ce³⁺ to Ce(CO₃)₂⁻. In the Ce/precipitation system at pH 8.20 and 11.00 where Ce(CO₃)₂⁻ is present in solution, the α_Lq-So values were 0.999821 (±0.000071) and 0.999589 (±0.000074), respectively, meaning that lighter isotope enrichment was observed in the liquid phase, which is the contrary to those of the other systems.

Extended X-ray absorption fine structure (EXAFS) analyses were also performed to investigate the coordination structure of the adsorbed or precipitated Ce species that control the isotopic fractionation during adsorption. Even at higher pH, where Ce(CO₃)⁺ or Ce(CO₃)₂⁻ are the dominant dissolved species, the first coordination sphere of Ce in the solid phase in the Ce/ferrihydrite and Ce/precipitation systems was similar to that observed at pH 5.00 where Ce³⁺ was the main species in solution. A slight elongation in the CeO bond length in the solid phase at pH 11.00, where negatively charged dissolved species are dominant in the liquid phase, may cause a decrease in isotopic fractionation in the Ce/δ-MnO₂ system. The coordination environment of Ce may not change significantly during the adsorption onto ferrihydrite, because Ce binds to the neutral surface OH group on ferrihydrite at pH below 8.5–8.8 (i.e. the pH of the point of zero charge (PZC) for ferrihydrite), similar to other cations when the metal–O distance was similar in hydrated and adsorbed species. At pH above PZC, Ce bonds to the negatively charged surface OH group, while Ce also bonds with CO₃²⁻ in dissolved species. The reduced partition functions (ln β) for dissolved species (ln β_Lq) and adsorbed species (ln β_So) with the same trends canceled each other, because ln β of hydrated cation was reduced by the binding anion, resulting in small isotope fractionations. Thus, isotope fractionations for Ce/ferrihydrite may be quite small at the entire pH conditions in this study. The direction of the isotopic fractionation was estimated based on density functional theory (DFT) calculations, which confirmed that lighter Ce is enriched in the liquid phase when Ce forms a complex with carbonate ions. Therefore, this study indicates that the dissolved species can control stable Ce isotopic fractionation during precipitation reactions.

Introduction

Cerium (Ce) is a rare earth element (REE) that is stable in a trivalent state and has a unique ability to form a tetravalent cation under oxic conditions with much lower solubility than Ce³⁺. Ce(IV) can be generated either by oxidation after adsorption or by oxidation in the liquid phase. Oxidative adsorption mainly occurs on manganese oxide (e.g., Taylor and McLennan, 1988, Takahashi et al., 2000, Takahashi et al., 2007), whereas oxidation in the liquid phase is followed by precipitation as Ce(OH)₄ or CeO₂ (De Baar et al., 1988, German and Elderfield, 1989, Braun et al., 1990). These chemical properties result in anomalously high or low concentrations of Ce relative to its neighboring elements, lanthanum (La) and praseodymium (Pr), which are known as positive or negative Ce anomalies, respectively (e.g., Henderson, 1984). Large positive Ce anomalies are often observed in marine ferromanganese deposits (e.g., Piper, 1974, Elderfield et al., 1981, Bau et al., 1996), and this phenomenon can be caused by oxidative adsorption (Takahashi et al., 2000, Takahashi et al., 2007). By contrast, large negative Ce anomalies in the REE patterns are observed in seawater as a counterpart to these positive anomalies (e.g., Elderfield and Greaves, 1982, Piepgras and Jacobsen, 1992, Möller et al., 1994). This redox-sensitive property facilitates the estimation of the redox state of the paleocean (e.g., Shimizu and Masuda, 1977, Wang et al., 1986, Wright et al., 1987, Murray et al., 1991) and studies of the redox evolution of the atmosphere (e.g., Fryer, 1977, Murakami et al., 2001, Kato et al., 2006).

Cerium has interesting relationship with two other major redox sensitive elements, iron (Fe) and manganese (Mn). Thermodynamic data show that the Fe²⁺/Fe(OH)₃ boundary is at a lower Eh than that of Ce³⁺/CeO₂, whereas the Mn²⁺/MnO₂ boundary is higher than the Ce³⁺/CeO₂ boundary (Brookins, 1988; Fig. 1). Therefore, Ce can undergo the following three redox reactions as the redox conditions become more oxic: adsorption without oxidation on Fe (hydr-)oxides (field (i) in Fig. 1); spontaneous precipitation (associated with oxidation) as Ce(OH)₄ (field (ii) in Fig. 1); and oxidative adsorption on Mn (hydr-)oxides (field (iii) in Fig. 1). Taking into account the fact that stable isotopic fractionation is largely associated with the bonding environment (Bigeleisen and Mayer, 1947, O’Neil, 1986, Criss, 1999, Schauble, 2004), it is expected that stable Ce isotope fractionation varies with these three different redox stages. The first stable isotopic study of Ce focusing on these relationships clearly showed that the magnitude of isotopic fractionation between the liquid and solid phases increases as the redox conditions become more oxic. This finding suggests that Ce isotopes can be used as a paleoredox proxy (Nakada et al., 2013a). The experiments were performed at pH 5.0 under air-equilibrium, where non-carbonated Ce³⁺ species were dominant (shaded area in Fig. 2A), which were different conditions from the natural marine environment, where Ce mainly occurs as carbonate species such as Ce(CO₃)⁺ and Ce(CO₃)₂⁻ (Cantrell and Byrne, 1987, Byrne and Sholkovitz, 1996, Tang and Johannesson, 2003; Fig. 2B). Thus, the stable isotopic fractionation of Ce should be examined at circumneutral pH with dominant Ce(CO₃)⁺ or Ce(CO₃)₂⁻ species in the liquid phase before this method can be applied to (paleo)environmental studies.

Assessing isotopic fractionation in various systems with different dominant species is not only important to establish an enhanced Ce paleoredox proxy but also to clarify the factors controlling isotopic fractionation of heavy elements. Multicollector-inductively coupled plasma-mass spectrometry (MC-ICP-MS) has enabled us to study isotopic fractionation of heavy elements, which provides a better understanding of the ancient Earth and biogeochemical cycles (e.g., Halliday et al., 1998, Johnson et al., 2004, Anbar and Rouxel, 2007, Wiederhold, 2015). It has been suggested that under equilibrium conditions the heavy isotope of an element is preferentially concentrated in the component in which the element forms the stronger bonds (Bigeleisen and Mayer, 1947, O’Neil, 1986, Schauble, 2004). The bond strength is greater for species that have (i) shorter bond length, (ii) higher oxidation state, (iii) highly covalent bonds between atoms with similar electronegativities, and (iv) lower coordination number (CN) (Criss, 1999, Schauble, 2004). For REEs, isotopic fractionation of neodymium (Nd) and samarium (Sm) during the adsorption on Fe and Mn (hydr-)oxides followed rule (i), whereas it was proposed that of Ce was controlled by the distorted structure of the adsorbed species and high stability of the aqua ion (Nakada et al., 2013b). However, the relationship between isotopic fractionation and dissolved species is largely unknown. The reduction of hexavalent chromium (Cr) under acidic conditions (pH < 1) showed near-equilibrium fractionation, whereas under neutral conditions (pH ∼ 7), Cr showed larger fractionation with a kinetic signature (Zink et al., 2010). This isotopic fractionation can be related to the dissolved Cr species of Cr³⁺and Cr(OH)₂⁺under acidic and neutral pH conditions, respectively. Adsorption of cadmium (Cd) and zinc (Zn) on Mn oxyhydroxide showed that isotopic fractionation during adsorption of these elements depended on ionic strength (Wasylenki et al., 2014, Bryan et al., 2015). This result indicates that the isotopic fractionation of these elements can be affected by dissolved species. Therefore, examining Ce stable isotopic fractionation and dissolved Ce species is important to establish the Ce stable isotope system as a paleoredox proxy and identify the factors controlling isotopic fractionation of heavy elements.

To achieve these goals, stable Ce isotopic fractionation was measured during adsorption and precipitation, which was further evaluated using extended X-ray absorption fine structure (EXAFS) analysis and density functional theory (DFT) calculations. Because the coordination environment generally controls stable isotopic fractionation (Bigeleisen and Mayer, 1947, O’Neil, 1986, Criss, 1999, Schauble, 2004), analysis of adsorbed and precipitated structure, or coordination geometry, is important. EXAFS analysis is the best method to constrain the coordination environment of the heavy element causing isotopic fractionation (Juillot et al., 2008, Brennecka et al., 2011, Kashiwabara et al., 2011, Wasylenki et al., 2011, Wasylenki et al., 2014, Nakada et al., 2013a, Nakada et al., 2013b). Electronic structure methods such as DFT can be used to calculate vibrational frequencies of compounds, which can then be used to estimate the magnitude of isotopic fractionation (Wasylenki et al., 2008, Wasylenki et al., 2011). Therefore, the combination of EXAFS and DFT calculations can clarify the results of stable Ce isotopic fractionation experiments.